R/E/P Community

R/E/P => R/E/P Archives => Dan Lavry => Topic started by: Big Bri on October 02, 2004, 03:08:38 am

Title: Proper word clock implementation
Post by: Big Bri on October 02, 2004, 03:08:38 am
Dear Dan,

Man, I really appreciate you joining the forum.  Thank you so much for your time.

I've been experimenting with different word clock scenarios in our facility and am dumbfounded by the sonic differences of just changing which unit is the clock master.  One unit gives me a great clear 3d image while another is quite smeared and more 2D.

Can you please tell me why this is happening and what the best possible scenario is to set up a central word clock distribution and how to get the best results.  Also, how do you control digital units with no dedicated word clock input?(i.e. Dat machines, stand alone cd writers, outbord fx)

Once again, thanks for your time and your knowledge.
Brian
Title: Re: Proper word clock implementation
Post by: bobkatz on October 02, 2004, 04:40:42 pm
Big Bri wrote on Sat, 02 October 2004 03:08

Dear Dan,

Man, I really appreciate you joining the forum.  Thank you so much for your time.

I've been experimenting with different word clock scenarios in our facility and am dumbfounded by the sonic differences of just changing which unit is the clock master.  One unit gives me a great clear 3d image while another is quite smeared and more 2D.

Can you please tell me why this is happening and what the best possible scenario is to set up a central word clock distribution and how to get the best results.  Also, how do you control digital units with no dedicated word clock input?(i.e. Dat machines, stand alone cd writers, outbord fx)




Dan is the man! And this is what I bet he'll tell you!

If you hear differences when you change clocks "controlling" your converters, then you have a defective converter design! A well-designed converter should contain internal phase locked loops whose performance reduces any incoming jitter artifacts to inaudibility. An external clock is a bandaid for a "cure" which can only be done properly within a good converter design. In fact, any converter which does not perform equally as good or BETTER on internal clock than external is also defective.

BK
Title: Re: Proper word clock implementation
Post by: danlavry on October 02, 2004, 05:54:02 pm
If you hear differences when you change clocks "controlling" your converters, then you have a defective converter design! A well-designed converter should contain internal phase locked loops whose performance reduces any incoming jitter artifacts to inaudibility. An external clock is a bandaid for a "cure" which can only be done properly within a good converter design. In fact, any converter which does not perform equally as good or BETTER on internal clock than external is also defective.

BK



Yes indeed! Well said.

The best way to clock a converter is with internal clock, using a good fundamental frequency crystal (third order types are more jittery), and locating the crystal properly (good ground to the AD ample hold and so on). You now have a low jitter clock inside the machine.

What happens when you get a stand alone “almost no jitter clock”? You look AT THE OUTPUT CONNECTOR of that “super clock box” and it generally can work as well as the internal crystal clock  Now take a cable and hook it to the AD chassis. Now you have to go through some electronic circuit to receive the clock. At this point, you have accumulated a lot more jitter (I can list half a dozen causes).

Well, this is not the end of the road. The big one is the PLL circuit. Unlike the internal clock (fixed crystal case), you have a crystal that can be pulled up or down by some amount, we call it a VCXO (voltage controlled crystal oscillator). There is some circuitry in there that keeps comparing the incoming external clock rate to the VCXO, and makes the proper adjustment on an ongoing basis…
What is more steady? A mediocre internal crystal implementation is going to outdo even a good external clock implementation.

But there are times and reasons to use external clocks. For example, if one needs to sync many chassis…  

It is true that the PLL does better when fed a less jittery clock, but that is just a tiny portion of the overall issue. As Bob stated, most of the burden is on the PLL. A Good PLL, inside the AD chassis should clean most of the jitter out.

Why do you get such different results with different sources? I am not there to probe. I would not start with comparing how much jitter each source provides. I would look into issues such as driving coaxial lines, and proper termination impedance. Make sure the clock lines have no “branches” – Driver to point A, than to point B, than to C all in series.

I am no fan of distribution amplifiers either. You can not beat:
Driver to point A (with a BNC T), than to point B (with BNC T)… at the end the BNC T is terminated with the proper line impedance (if the cable is 75Ohm, so is the termination). It is a cost effective solution that yields the best results.

BR
Dan Lavry      
Title: Re: Proper word clock implementation
Post by: bobkatz on October 02, 2004, 07:25:58 pm
danlavry wrote on Sat, 02 October 2004 17:54




Why do you get such different results with different sources? I am not there to probe.





Dan said, modestly! While adding much of Dan Lavry's designer-level substance to my short reply to the question.

I'd like to add: It is possible that you can't EVER stop getting different results with different sources unless you throw out the entire poorly-designed converter (it's mostly due to the poor PLL design within the converter, though power supply and grounding is all part of that).

Quote:



I would not start with comparing how much jitter each source provides. I would look into issues such as driving coaxial lines, and proper termination impedance. Make sure the clock lines have no ?branches? ? Driver to point A, than to point B, than to C all in series.

I am no fan of distribution amplifiers either. You can not beat:
Driver to point A (with a BNC T), than to point B (with BNC T)? at the end the BNC T is terminated with the proper line impedance (if the cable is 75Ohm, so is the termination). It is a cost effective solution that yields the best results.




I'm sure Dan would like to point out that while this is the best way to do it, each of the devices that is to receive the wordclock must have a termination on/off switch. The termination must be turned off until the end of the line, or the source will be overloaded. In other words, the last device in the line must have the 75 ohm input. Unfortunately, the VAST MAJORITY of devices and converters I've encountered do not have such a switch, or else you have to go inside and remove a jumper or in worst case desolder a termination resistor.

Dan's points are very well taken, though, and you can save a lot of costs (both hardware and performance costs) by avoiding that distribution or clock amplifier circuit and just using his BNC-T approach. But you DO have to know how to measure input impedance if in doubt whether the input word clock is truly 75 ohm.

I'd like to point out that while wordclock has potential jitter superiority over AES/EBU "black" (there's an official AES number for AES black, I forget the number) there are many problems with using wordclock when synchronizing multiple channels and/or multiple devices. There is NO STANDARD for wordclock phase! You can end up with left and right channel reversal, or even a phase shift on the output of some of your supposedly "synchronized" devices. This is why AES/EBU sync is much preferred. However, there is a much higher jitter penalty with AES/EBU sync and even more attention has to be paid to the PLL on the input of any converter which is receiving AES/EBU sync.

BK
Title: Re: Proper word clock implementation
Post by: heinz on October 03, 2004, 02:03:19 am
bobkatz wrote on Sat, 02 October 2004 21:40

An external clock is a bandaid for a "cure" which can only be done properly within a good converter design. In fact, any converter which does not perform equally as good or BETTER on internal clock than external is also defective.


danlavry wrote on Sat, 02 October 2004 22:54

A mediocre internal crystal implementation is going to outdo even a good external clock implementation.


Wow! This thread has been very illuminating, kind of shatters some preconceptions I had. Thanks to all for the great information.
Title: Re: Proper word clock implementation
Post by: Ozzie Bostic on October 03, 2004, 10:58:44 am
danlavry wrote on Sat, 02 October 2004 16:54

If
I am no fan of distribution amplifiers either. You can not beat:
Driver to point A (with a BNC T), than to point B (with BNC T)… at the end the BNC T is terminated with the proper line impedance (if the cable is 75Ohm, so is the termination). It is a cost effective solution that yields the best results.

BR
Dan Lavry      



Hello and Welcome Dan,

I am currently enjoy the wonderful sound and clarity of your Blue series converters great product.

My question is since your not a fan of distribution clocks; what is the maximum number of devices that should be connected in series via the BNC T method mentioned above before the clock source is degraded?

Thanks in advance.
Title: Re: Proper word clock implementation
Post by: Albert on October 03, 2004, 11:49:49 am
Another question: Would you use a BNC t-bar and terminator on gear that is already internally terminated? Would that yield better results, or would it damage the gear?

Also, would you use a BNC t-bar on gear that has a WC in and out, or would you pass through the gear, using it's own circuitry? And if you did use the BNC t-bar to bypass the gears internal routing, would you then use a terminator on the units WC output or rely on its own termination?

Thank you, this is a great thread.
Title: Re: Proper word clock implementation
Post by: bobkatz on October 03, 2004, 01:08:13 pm
Albert wrote on Sun, 03 October 2004 11:49

Another question: Would you use a BNC t-bar and terminator on gear that is already internally terminated? Would that yield better results, or would it damage the gear?





It wouldn't damage the gear, but eventually it would load down the source wordclock so that the system would not function or would not function well.


1) I suggest you start with an oscilloscope on the wordclock generator. Measure its output, which should be 4 v p-p if possible. Any lower and some following devices will not like it, but most following devices are comfortable with as low as 1 v p-p---
TERMINATED!!! Hey, there's no standard! It's a wild wild west.

2) Then, put a terminator on it in parallel with the scope input. The level should drop by 1/2. If not, then the source is not truly 75 ohms. This is not a big big deal with wordclock frequencies, which are not so high generally as to be subject to reflection problems. I'm sure Dan can explain the compromises there, I don't know. It would cause some limit as to the length of the cable and how many devices you can attach.

3) Using this level as your goal (2 v p-p terminated), run the signal into each of your wordclock devices and use the scope to see if the device terminates the line. Chances are it does and you'll have to go inside the device and remove the 75 ohm resistor that's there, so you can then use the BNC-T.

If the wordclock generator is truly 75 ohm, 4 v p-p (unterminated), 2 v p-p terminated, then I don't see a problem using low loss coax for even 20-40 feet and having 3 to 6 high impedance (unterminated) dropoffs in between. Scope the end of the chain. If it's still a clean square wave at approximately 2 v p-p into the terminator at the end of the chain, then I'd say it's a win-win!

Quote:



Also, would you use a BNC t-bar on gear that has a WC in and out, or would you pass through the gear, using it's own circuitry? And if you did use the BNC t-bar to bypass the gears internal routing, would you then use a terminator on the units WC output or rely on its own termination?




The problem is that most word clock equipment has internally terminated inputs and will have to be modified (real real pity, since Dan's approach is absolutely valid). If you remove the 75 ohm resistor, then you will be able to use a BNC T, as this effectively bypasses the gear's internal routing. All you are concerned with at that point is the impedance of the gear's loading. Usually it's a small coil or transformer loaded by a 75 ohm resistor. Remove the resistor and (hopefully) it becomes high impedance, that is, negligible load on the rest of the circuit.

By the way, if the equipment to be slaved is NOT a converter, but simply a digital processor or DAT machine or whatever, then a wordclock distribution amplifier will do no harm, as the interface jitter is irrelevant. I use a Lucid Clk-X to feed wordclock to my Digital Timepiece, which is NOT feeding converters, to my DAT machine, to my TC Electronic System 6000, whose converters I am not using, etc. etc.

You could take the Clk-X box, remove the input terminator, and feed using Dan's BNC-T method through the input of the Clk-X, and feed (modified unterminated) critical converters like the Digidesign, and so on, through  the BNC T system, and use the Clk-X wordclock distribution box to feed the rest of things.

My Tascam DAT machine has a Wordclock through jack and some kind of automatic termination. How does that automatic termination sensing circuit work? I don't know!!!! I don't trust it, either, but I suppose I could sniff it out with a scope...
Title: Re: Proper word clock implementation
Post by: punkest on October 03, 2004, 01:53:42 pm
danlavry wrote on Sat, 02 October 2004 22:54




The best way to clock a converter is with internal clock,    


A mediocre internal crystal implementation is going to outdo even a good external clock implementation.



It is true that the PLL does better when fed a less jittery clock, but that is just a tiny portion of the overall issue. As Bob stated, most of the burden is on the PLL. A Good PLL, inside the AD chassis should clean most of the jitter out.

Dan Lavry      




    I thought this was only the case with older PLL that does not have reclocking circuits. I heard that many new converters used the PLL to sync and then regenerated the clock to get rid of jitter.

    Anyway, I always try to slave all to the converter?s clock, as you pointed out, Dan.

     Bob pointed out that AES sync is more jittery that WC, how would you rate the jitter when synced over ADAT optical?

     I have apogee AD. Optical cable to feed converted signals and clock to computer interfaces. Word clock cable to feed clock to Yamaha 03d console that also receive signal but no clock optically from computer interface. Would it be a better way???

     Is there potential problem with feeding parallely two sources instead of the usual serial A to B, B to C ???

Hans Mues
Title: Re: Proper word clock implementation
Post by: bobkatz on October 03, 2004, 06:20:35 pm
[quote title=punkest wrote on Sun, 03 October 2004 13:53]
danlavry wrote on Sat, 02 October 2004 22:54





    I thought this was only the case with older PLL that does not have reclocking circuits. I heard that many new converters used the PLL to sync and then regenerated the clock to get rid of jitter.






Good PLL design really separates the men from the boys. Now Dan's not going to mention brands but I don't have an afiliation so I'll just tell you that Apogee's claims of low jitter just did not bear out. The performance did not match their claims.

All PLL's "reclock" so that's not the issue. It's "how well they do it", and how well the output performance actually works that counts. I have measurements of converters' jitter performance, both good and bad, in my book.

Quote:



     Bob pointed out that AES sync is more jittery that WC, how would you rate the jitter when synced over ADAT optical?





Doesn't ADAT really require a wordclock anyway?  If that's the case, then ADAT sync would be irrelevant, except to the extent that clocks and power and grounds can interact in a poorly-designed box.

Quote:



     I have apogee AD. Optical cable to feed converted signals and clock to computer interfaces. Word clock cable to feed clock to Yamaha 03d console that also receive signal but no clock optically from computer interface. Would it be a better way???

     Is there potential problem with feeding parallely two sources instead of the usual serial A to B, B to C ???




Hans, can you please repeat the above, more slowly  Smile.  

1) What is the block diagram of your interconnections
2) What functions as the clock master?
3) What A/D and D/A converters do you use in this block diagram and what are their functions (monitoring, mixdown, etc.)

I know that many people feel this jitter "nonsense" is overrated. But if it weren't important, why would so many people be complaining about getting "great imaging and sound with one clock and real flat sound with another". So, "the sound" must be important to a lot of people, and discovering that clocking makes a sonic difference is enough to educate people as to why and how.

That's why getting jitter right is important! However, it is important to know when jitter makes a difference and when it does not. It does not make a difference with digital to digital transfers or digital processors (EQs, compressors, etc.).

BK
Title: Re: Proper word clock implementation
Post by: danlavry on October 03, 2004, 06:27:34 pm
Bob Katz said:
I'd like to add: It is possible that you can't EVER stop getting different results with different sources unless you throw out the entire poorly-designed converter (it's mostly due to the poor PLL design within the converter, though power supply and grounding is all part of that).


Of course. I was sort of implying that, but it is good to have it clearly stated.

I'm sure Dan would like to point out that while this is the best way to do it, each of the devices that is to receive the wordclock must have a termination on/off switch. The termination must be turned off until the end of the line, or the source will be overloaded…

Yes, thanks for reminding me of that. I decided long ago to not terminate internally, though I provide a jumper for internal termination. I guess I am not in the “vast majority” on that.  

Dan's points are very well taken, though, and you can save a lot of costs (both hardware and performance costs) by avoiding that distribution or clock amplifier circuit and just using his BNC-T approach. But you DO have to know how to measure input impedance if in doubt whether the input word clock is truly 75 ohm.

Or call the factory and ask them if the gear is internally terminated, and if so how to be sure and disconnect the internal termination. It is economically wise, or as you stated so well - save a lot of costs (both hardware and performance costs)…

I'd like to point out that while wordclock has potential jitter superiority over AES/EBU "black" (there's an official AES number for AES black, I forget the number) there are many problems with using wordclock when synchronizing multiple channels and/or multiple devices. There is NO STANDARD for wordclock phase! You can end up with left and right channel reversal, or even a phase shift on the output of some of your supposedly "synchronized" devices. This is why AES/EBU sync is much preferred.

This is the voice of experience from the field, thus very worth while points you bring up.

I am going to add some technical comments to your statements:

Is WC better than AES digital black?
Some of the “anti AES” for clock synchronization may be a “left over” from the days when people used AES with data. Such practice added a lot of jitter, but we now know we need to use a fixed pattern.
Also, as Bob stated, AES does not give you inconsistent polarity when doing a multi channel work with different gear (but is it ever a good idea?)
While AES digital black is not as bad as some people think, WC does have the advantage: The issue is signal transmission reflections (cable impedance and termination). When it comes to clocks, audio people can learn from disciplines such as ultra high speed signal transmission:

WHY TERMINATE CABLES?

Say you send a single transition (part of a square wave clock down) down a properly terminated cable. Say real fast rise from 0 to 5V.  
Lets next have some mismatch between cable impedance and the termination resistor. What happens?
The signal travels down the cable (at about 8nSec per foot – 3/4 the speed of light or so).  
If the termination is higher than the cable, say by 10%, when the transition gets to the cable end, about 5% (1/4V) gets added to the signal. That 1/4V now travels towards the clock source. When it gets there, it normally sees a near AC short, so it becomes inverted (-1/4V) and it now travels back to the load. There again we get about 5% (half the mismatch) of the new arrival – about 12.5mV reflected back to the source. Again it gets inverted and resent…  This of course will continue to go back and forth between source and load. The good thing is it decays down.

What if the load resistance is lower by 10% than cable impedance? It is a similar story. The reflections at the load are still 5% but inverted.

I do not wish to confuse anyone. But just think of the fact that such a mess of back and forth takes place for each and every transition in WC or AES. So the reason to PROPERLY TERMINATE is to have NO REFLECTIONS.

REAL WORLD PRACTICE

How real is the reflection issue? We assumed so far the transitions are very fast. Of course one can forget about transitions at DC or for very slow voltage changes. So the question is: what is “slow”? Slow (or fast) rise is always in reference to cable length. Digital signal transmission engineers look at the ratio between cable length and rise time. A 1 inch cable (150 pSec delay) is not a problem for a 10nSec rise time. But 100nSec cable is!  

For clocks, one has to also account for the time between transitions. A WC signal puts out a transition, and may cause a bunch of reflection activity due to mismatch. We have say 11uSec between transitions (for 44.1KHz square wave WC) thus “a lot of time” for decaying to take place. Each reflection is lower than the previous, and we have a lot of time for many back and forth action, thus decay. But say we use AES with transition rates in the MHz – much less than a 1uSec time for back and forth action before the next transition… The reflection decay is not as complete.

A new transition occurring before an incomplete decay is undesirable! We end up with interaction between cable length (delay) and AES data. Making it the “same data” (digital black) does NOT make for a repetitive behavior! With “everything rattling all over the place at the same time” the voltage levels on each point of the cable change over time, and with it comes jitter.

So, WC users can get great results with pretty bad terminations. In fact, having no termination with a reasonable short cable (say 100 feet) often works just fine. However “incorrect” the signal, it tends to be repetitive from clock edge to clock edge, and repetition is what we want most.

“AES black” users need to be much more careful with cable terminations.    

BR
Dan Lavry
Title: Re: Proper word clock implementation
Post by: Big Bri on October 04, 2004, 03:17:50 am
Dan & BK

Thank you both for your answers. The problem comes in when using different brands of converters on one system and trying to interface with digi design 888's and usd.  I found leaving my Apogee on internal and clocking the usd via the word clock inputs, the system sounded better than by clocking with the usd and using the apogee in AMbus sync.  So, is it more correct to use the Apogee AD 8000 clock or am i causing a problem i haven't discovered yet.

Thanks Again for your answers!!
Brian  
Title: Re: Proper word clock implementation
Post by: bobkatz on October 04, 2004, 10:16:27 am
The AD 8000 is about the only Apogee converter I'm "comfortable" with in terms of its sound quality (personal opinion). I tested it on internal and external clock with a good clock, for jitter artifacts, and it appears to perform about the same either way, which is a good sign.

If you have an FFT analyser, testing for jitter artifacts is very easy to do. See

http://www.tcelectronic.com/Default.asp?Id=1156

As to whether these artifacts correlate with audibility? This is a very difficult question. But at least you can quantify the amounts to which the converter is susceptible to external clock interference and it MAY help lead you to a conclusion as to whether you are hearing an improvement, or just a "subjective enhancement". Jitter can cause:

a) distortion (which can be perceived as euphonic)
b) image narrowing or image distortion (which can be perceived as euphonic)
c) loss of low level resolution (which can be perceived as euphonic, too!)

See what I mean!


Hope this helps,


Bob
Title: Re: Proper word clock implementation
Post by: punkest on October 05, 2004, 12:46:50 pm
bobkatz wrote on Sun, 03 October 2004 23:20



  All PLL's "reclock" so that's not the issue. It's "how well they do it", and how well the output performance actually works that counts. I have measurements of converters' jitter performance, both good and bad, in my book.




  Bob, thanks for the info, I thought older PLL?s did not recover from the shifting they perform in order to "trace" the incoming clock, good to know all do that or at least try to.

Quote:



  Doesn't ADAT really require a word clock anyway?  If that's the case, then ADAT sync would be irrelevant, except to the extent that clocks and power and grounds can interact in a poorly-designed box.




No, ADAT carries clock in its optical cable, together with 8 channels (@44.1 and 48) of audio signal.

Quote:

Quote:



     I have apogee AD. Optical cable to feed converted signals and clock to computer interfaces. Word clock cable to feed clock to Yamaha 03d console that also receive signal but no clock optically from computer interface. Would it be a better way???

     Is there potential problem with feeding parallely two sources instead of the usual serial A to B, B to C ???




Hans, can you please repeat the above, more slowly  Smile.  

1) What is the block diagram of your interconnections
2) What functions as the clock master?
3) What A/D and D/A converters do you use in this block diagram and what are their functions (monitoring, mixdown, etc.)




OK Bob, sorry for not being clear, the signal flow is:

Apogee converters?s optical out to MOTU 2408 MKIII optical in. (master clock is Apogee AD, MOTU?s clock is set to optical in). A BNC cable goes from WC of the Apogee AD to the Yamaha 03d WC in. (Yamaha clock is set to WC input). MOTU 2408 optical outs goes to Yamaha 03d optical input for control room monitoring and studios cues, but when mixing, I stay in the box.  Sounds fine to me, but anyway I was curious if you recommended otherwise.

Quote:



  That's why getting jitter right is important! However, it is important to know when jitter makes a difference and when it does not. It does not make a difference with digital to digital transfers or digital processors (EQs, compressors, etc.).

BK
Very Happy

 I knew that one, and I know the most sensitive part is the conversion.

Bob, thanks for the time you take in this and other posts, it is great to have you here.

Hans
Title: Re: Proper word clock implementation
Post by: bobkatz on October 05, 2004, 10:51:03 pm
punkest wrote on Tue, 05 October 2004 12:46


Hans, can you please repeat the above, more slowly  Smile.  

1) What is the block diagram of your interconnections
2) What functions as the clock master?
3) What A/D and D/A converters do you use in this block diagram and what are their functions (monitoring, mixdown, etc.)




OK Bob, sorry for not being clear, the signal flow is:

Apogee converters?s optical out to MOTU 2408 MKIII optical in. (master clock is Apogee AD, MOTU?s clock is set to optical in). A BNC cable goes from WC of the Apogee AD to the Yamaha 03d WC in. (Yamaha clock is set to WC input). MOTU 2408 optical outs goes to Yamaha 03d optical input for control room monitoring and studios cues, but when mixing, I stay in the box.  Sounds fine to me, but anyway I was curious if you recommended otherwise.

[/quote]

Hans, thanks for your nice thoughts. I enjoy being here, and even when I'm answering a question, I learn something.

Let's see. As far as I can see, I think it's a good idea you're using the Apogee's as the master clock. They're certainly going to do better on internal sync than locking to any of the other current devices in your system. And only a measurement and listening test of the Apogee's would reveal if they would do better when fed some external "high-end" clock. And if they did do better on external sync, that would be a denigration of the Apogees, not a praise of the external clock. As long as you are not using the converters of the Yamaha and mixing in the box, you are doing fine from the point of view of clocking but you certainly should expand your horizons and look into a good set of high quality outboard processing gear. Your weakest link in this setup is actually your monitoring, and you can solve that by using a high quality jitter-immune external  D/A converter and monitor level control for monitoring. The 2408 goes digitally into the Yamaha, and is processed digitally, so jitter is irrelevant there, and thus only the monitor DAC is in trouble, and the Yamaha's DAC is probably suffering there from point of view of jitter.

Where you are going to run into trouble is when you introduce external analog processing gear. And for that, I would recommend that on each analog insert you use the digital outs of the Yamaha to feed higher quality external converters. A good approach would be to replace the 2-channel Apogee with a single stable 8 channel A/D/A system running on internal sync and proven to perform best that way. Use pairs of that A/D/A system for your analog processing insert, and one pair for your monitoring DAC. You can do that with digital audio routing and still maintain that A/D/A as internal clock to drive the Yamaha and other gear.

That's my take on the affair!
Title: Re: Proper word clock implementation
Post by: punkest on October 06, 2004, 10:57:44 am
bobkatz wrote on Wed, 06 October 2004 03:51


   I think it's a good idea you're using the Apogee's as the master clock. They're certainly going to do better on internal sync than locking to any of the other current devices in your system. And only a measurement and listening test of the Apogee's would reveal if they would do better when fed some external "high-end" clock. And if they did do better on external sync, that would be a denigration of the Apogees, not a praise of the external clock. As long as you are not using the converters of the Yamaha and mixing in the box, you are doing fine from the point of view of clocking but you certainly should expand your horizons and look into a good set of high quality outboard processing gear.




 We?ve got some nice analog compressors and we have a lot of pre?s. We usually get the sound as close as we want BEFORE A to D. We are deciding if we are going to get an analog mixer and a 16 channel D to A,  (since we?ve got two 2408 MKIII and each have 8 outputs) and go the mix out of the box route, but then we would need some more compressors, eq?s and time/modulation processors, besides extensive cabling and patch bay.

Quote:



  Your weakest link in this setup is actually your monitoring, and you can solve that by using a high quality jitter-immune external  D/A converter and monitor level control for monitoring.




  Yes, for the actual setup, I have been thinking of getting something like the Benchmark DAC-1 or Lavry D to A, and feed it from the SPDIF out of the Yamaha mixer, and get a Coleman audio to control level and speaker sets.

Quote:



  The 2408 goes digitally into the Yamaha, and is processed digitally, so jitter is irrelevant there, and thus only the monitor DAC is in trouble, and the Yamaha's DAC is probably suffering there from point of view of jitter.




  I guess the new D to A would take care of this

Quote:



  Where you are going to run into trouble is when you introduce external analog processing gear. And for that, I would recommend that on each analog insert you use the digital outs of the Yamaha to feed higher quality external converters. A good approach would be to replace the 2-channel Apogee with a single stable 8 channel A/D/A system running on internal sync and proven to perform best that way.




  The apogee we have is 16 channel A to D, and we have as well 16 channel D to A from the 2 MOTU 2408MKII, I guess it is not fancy, but it'll have to do while we get better, and I prefer the MOTU D to A than the Yamaha 03d, so when I go external I go out of the MOTU to processor to Apogees and back to the box.

Quote:


 
That's my take on the affair!


Bob, again,  thanks for the insight.

Hans Mues
Title: Re: Proper word clock implementation
Post by: Albert on October 06, 2004, 12:20:40 pm
Bob and Dan, thank you for your very helpful and interesting replies. I've learned a lot from this thread and there is a lot more to think about.

Perhaps I missed this, but how many bnc t-bars can you put in a row before the signal degrades, or is this a concern at all?

As luck would have it, some of my gear does indeed have WC termination on/off switches, so I am going to try out this bnc daisy chain thing. I just bought a bunch of bnc t-bars and terminators on eBay! Smile

Bob, of particular interst to me was your comment regarding daisy chaining anything with a converter, but using a dsitribution box for digital to digital units. I had not heard that before, and in fact believed that jitter affected all digital gear equally. Apparently (or obviously) I was incorrect in that.

Incidentally, I am still enjoying "Mastering Audio", it has become one of a handful of essential reference books for me. Thank you again for sharing your experience and knowledge.
Title: Re: Proper word clock implementation
Post by: danlavry on October 06, 2004, 03:04:10 pm
Albert wrote on Wed, 06 October 2004 17:20


Perhaps I missed this, but how many bnc t-bars can you put in a row before the signal degrades, or is this a concern at all?



As I stated earlier in the thread, the issue of termination is much more impotent for higher frequency digital data (such as AES) than lower frequency (such as WC).

Lets me show an example with some numbers:

Say we have 1000 feet of coaxial cable (very long). That is 800nsec delay. Say we are working with a square wave WC at 96KHz. So a cycle time is about 10uSec. The time between transitions is about 5uSec (assuming a square wave). So how may reflections will occure between 2 transitions?
10usec/800nsec = 12.5 But since we need to account for twice the cable length (reflections travel from source to load and back to source), so we have over 6 round trips.

Now, lets get practical and assume a 5% cable to termination mismatch. So the first reflection yields 2.5% (half the mismatch), the second reflection is 2.5% of 2.5% = .0625%. The third round trip yields 2.5% of .0625%... and at 6 round trips we have 2.4 pico (a millionth of a millionth) of the original step. Say the step was 5V, the reflection activity is truly tiny 1.2nV (nano volt). And that is for 1000 feet of cable.

So the point is, proper termination is almost a joke at WC speed. The same 1000 feet cable with say 1MHz frequency (instead of 96KHz) will be a different story. The same exercise with 1uSec/800nSec = 1.25 - a new clock transition
is being sent down the cable before the first reflection had enough time to go through a round trip! So the first reflection which is 2.5%. or .125V "rides on" the 5V signal... That will much things up... That is at 1MHz, imagine faster rate such as AES signal.

Reflections and proper termination are about cable length vs period (frequency). The whole concept of proper termination plays no role at very low frequencies, where audio folks look at a cable as a capacitive medium. 44.1KHz-48KHz WC is still very low period about 20uSec. The signal contains high frequencies (fast rise and fall time square wave), but it is the basic period that counts for reflections.

So what is the bottom line? Don't worry about WC reflections. If you want to "be pure", go for short interconnects say 3 feet between units, and the first cable leading to the units can be long. That is all for WC, not for AES. How many units? A pile of units! Assume that each unit input looks like 10KOhm in parallel to say 20pF. Say you have 100 units hooked, thus a distributed load 100 Ohms and that will cost you some in amplitude, but still work fine. The capacitance will muck the purity of the 75Ohm line and so your starting point is no longer 2.5%. Say it is 25%, than the 1000 feet cable example will yield 4mV activity on the next edge. So yes 100 units is potentially on the hairy edge. I would not hesitate to have 25 units.

You could not hear the difference between the system with or without the termination resistor, that is in a double blind test. You would hear it with AES signal.

BR
Dan Lavry

   
Title: Re: Proper word clock implementation
Post by: Albert on October 06, 2004, 04:17:36 pm
Thanks for that explanation Dan. Especially the last couple paragraphs, which were simple enough for me to understand. I'm a composer/performer not an engineer, though I do try to educate myself as much as possible. The whole concept of reflections is new to me, so that will require some more thought for me to understand it.

I'll probably have around 15 pieces on one WC chain, so that seems to be well within the safety margin. The daisy-chaining approach simplifies things in my studio and actually saves me some money I was going to spend on another WC distribution device.

Dealing mostly with midi devices for so many years, I just assumed *any* daisy-chaining would be a bad thing. For midi devices, it certainly is. You can't go more than a few units and then you are in trouble.

Great thread, thanks again for the advice.
Title: Re: Proper word clock implementation
Post by: bobkatz on October 06, 2004, 06:42:07 pm
Albert wrote on Wed, 06 October 2004 12:20



Bob, of particular interst to me was your comment regarding daisy chaining anything with a converter, but using a dsitribution box for digital to digital units. I had not heard that before, and in fact believed that jitter affected all digital gear equally. Apparently (or obviously) I was incorrect in that.





Hi, Albert! Thanks for your very wonderful comments on my book. Well, to reiterate, jitter is ONLY important with converters! And a good converter is supposed to be immune to incoming jitter, as Dan has eloquently pointed out. So this means that you can use Dan's T-Bar strategy to minimize jitter in the feeds to critical gear that contains converters, and use a cheap distribution box to feed WC to gear that doesn't matter.

Hope this confirms your thoughts!

By the way, I'm not trying to hijack this forum away from Dan, it's just that on the subjects that I know a little something about, I do like to contribute! And at 130 words per minute typing, my thoughts tend to go virtually direct from brain to fingers to Internet   Smile


Take care,


Bob
Title: Re: Proper word clock implementation
Post by: bobkatz on October 06, 2004, 06:46:56 pm
Albert wrote on Wed, 06 October 2004 16:17



Dealing mostly with midi devices for so many years, I just assumed *any* daisy-chaining would be a bad thing. For midi devices, it certainly is. You can't go more than a few units and then you are in trouble.

Great thread, thanks again for the advice.



Well, it's not exactly "daisy-chaining". Unlike MIDI-through, which does add a delay, all you are doing with a BNC-T is putting a Y cord with a little load on it. Think of it like a big river flowing downstream, and little pipes on the side of the river taking a tiny tiny little drop out of the river at each pipe. That's effectively what the high-impedance connection is doing.

BK
Title: Re: Proper word clock implementation
Post by: danlavry on October 06, 2004, 07:06:54 pm
"I'm a composer/performer not an engineer, though I do try to educate myself as much as possible. The whole concept of reflections is new to me, so that will require some more thought for me to understand it."

I understand that in a forum one has a mix of interests, from the very technical to the very musical and all the in between. It is hard to talk to all at the same time, so I am just trying my best, and hope that it is understood.

There are a lot of EE's that do not know about reflections. Certainly the computer science types have little reason to get into it. On the other hand, designers of very high speed gear are heavily into transmission line theory. Imagine a 5GHz digital data transfer - that is 200pSec (pas is a billionth of a second). An electrical signal travels about 1.3 inch in 200pSec...
It is a bit "funny" for me to see how a concept such as proper line termination is followed at say 44.1Khz, but a line level cable with 22KHz does not call for impedance matching between audio cable and some temination. The practice is almost "brainless impulse", you see a coax, and you terminate.
The coax is good because of the nice shield, and the locking connectors are nice, but that is all, for audio WC.

I'll probably have around 15 pieces on one WC chain, so that seems to be well within the safety margin. The daisy-chaining approach simplifies things in my studio and actually saves me some money I was going to spend on another WC distribution device.

Save money and improve jitter performance at the same time.

Dealing mostly with midi devices for so many years, I just assumed *any* daisy-chaining would be a bad thing. For midi devices, it certainly is. You can't go more than a few units and then you are in trouble.

Of course midi is a different thing all. I do not know where midi basic clock rate is these days, it used to be 500KHz, than 1MHz... and it takes many clocks to construct a midi command (all the data is in series)... So you end up with serious delays, not to mention the ability to put too much data on the buss (especially with continuess controllers such as after touch, pitch bend and so on). So midi problems can be in the milliseconds. WC over cable is about 1.5 nano second per foot, or 2/3 the speed of light!

BR
Dan Lavry
Title: Re: Proper word clock implementation
Post by: danlavry on October 06, 2004, 07:50:24 pm
[quote title=Ozzie Bostic wrote on Sun, 03 October 2004 15:58
Hello and Welcome Dan,
I am currently enjoy the wonderful sound and clarity of your Blue series converters great product.
My question is since your not a fan of distribution clocks; what is the maximum number of devices that should be connected in series via the BNC T method mentioned above before the clock source is degraded?
Thanks in advance.[/quote]

Hi Ozzie,

Well, it took me a couple of days to try and figure how to answer it. I hope my latest posts with all the details are not "too much".

Assuming you are talking WC sync:

It is difficult to put a number on it, but I do not see a any reason why one can not hook 25 devices in series. You have to be sure that all the devices (except the last one)are not internaly terminated.

If the last device is internal terminated, do not terminate the last BNC T. In fact you do not need the lase BNC T and can plug directly.

If the last device is not internally terminated, you can put a BNC-T at the end and terminate the line there, at the last BNC.

If you are doing AES sync, I would go for a lot fewer devices. At that point I would want to know some things (such as line length, what sample rates are used...)

BR
Dan Lavry
Title: Re: Proper word clock implementation
Post by: 12345 on October 07, 2004, 08:52:31 pm
Hello All,

Great topic!  

When do you recommend using a grounded (with a grounding chain/wire attached) BNC T, and when do you recommend using an open one?  

Regards,
My World
Title: Re: Proper word clock implementation
Post by: danlavry on October 08, 2004, 12:48:36 pm
My World wrote on Fri, 08 October 2004 01:52

Hello All,

Great topic!  

When do you recommend using a grounded (with a grounding chain/wire attached) BNC T, and when do you recommend using an open one?  

Regards,
My World


BNC is for coaxial cables. The cables have one "center conductor" and a shield (typically braided). So it is aimed at sending signal on the center and the shield is for ground return path.

The question is which ground to connect the cable to. That is a system grounding question.

The BNC female connectors (what you plug the coax cable to) come in variety of configurations. You can get an all metal connector, designed to short your return to chassis ground if mounted on a metal chassis. Or you can get a BNC connector that does not short to a chassis. I am not sure why some come with a grounding chain and cover. My guess is for protecting the open connector in harsh environments. It is only a guess.

BR
Dan Lavry    

Title: Re: Proper word clock implementation
Post by: Ozzie Bostic on October 08, 2004, 02:35:28 pm
[quote title=danlavry wrote on Wed, 06 October 2004 18:50
Hi Ozzie,

Well, it took me a couple of days to try and figure how to answer it. I hope my latest posts with all the details are not "too much".

BR
Dan Lavry
[/quote]

Yes Dan, the previous post was exactly what I was looking for. I'm actually an EE and IT Major by degree and find your posts absolutely astounding. I am kinda sorry I didn't do more EE work, but more IT related. At any rate, great post.

Thanks again,

Title: Re: Proper word clock implementation
Post by: Albert on October 09, 2004, 12:32:28 pm
Dan and Bob, thanks again for your very illuminating posts, I appreciate it very much.

Okay, from my non-technical point of view, how would the pictured arrangement work as far as distributing WC? My question being, if it is possible to string 25 units in a WC chain, what if those same units were spread out on 3-4 shorter chains that originated from the same place?  The BNC t-bar configuration at the originating WC device would look like this:

http://www.albypotts.com/images/bnctbarx3.jpg

Crazy? Does that put too much of a load on the signal? Let's say there are the same number of devices as in a straight WC chain, would there be four times the load with this setup?

As you can see, I got my shipment of BNC t-bar connectors. Smile
Title: Re: Proper word clock implementation
Post by: danlavry on October 09, 2004, 04:36:50 pm
Okay, from my non-technical point of view, how would the pictured arrangement work as far as distributing WC? My question being, if it is possible to string 25 units in a WC chain, what if those same units were spread out on 3-4 shorter chains that originated from the same place?  The BNC t-bar configuration at the originating WC device would look like this:

Crazy? Does that put too much of a load on the signal? Let's say there are the same number of devices as in a straight WC chain, would there be four times the load with this setup?

As you can see, I got my shipment of BNC t-bar connectors. Smile[/quote]


From a purist stand point, don't do it - avoid branches. A transmission line (the coax cable) is a constant impedance line. Say you apply a 1V sudden step on a 50Ohm line. The “good old electrons” do not know what is ahead. They “can not know” how long the cable is, before they got to the end. So, the line at first look like a resistor. That 75Ohm line looks like a 75Ohms initially. So the current initially is 1V / 75 Ohm = 13mA and that current wave is traveling down the cable, and wherever it goes, what is ahead is “the rest of the cable”. Again, the electrons just see 75 Ohms.
At some point soon the wave (goes at about 8nSec/foot) gets to the end of the line. What does it see?  If the termination is proper, it sees a 75Ohm resistor. All that current (13mA) now goes through the resistor, thus 13mA*75Ohms=1V.
You can now see how a reflection can happen with wrong termination. Say, you terminated with 100Ohm resistor. The 13mA will “meet” an 100 Ohms thus we have 13mA*100Ohms=1.3V !!!  Now the load has .3V higher voltage than the source, so you have 4mA current going backwards…. It ends up as a lot of back and forth reflections.

The point is: The pure way to deal with transmission line is to have everything in series, and a correct termination at the end. If you split the line with a T into 2 long cables, that same 13mA current will divide with 6.5mA going to each cable. For “a while” (till the waves get reflected from their respective coax ends) each line has only 1/2V… The whole concept of impedance matching is disturbed. Where is the termination? On line A or B?...

For WC you can get away with some short branches if it is really necessary, a few feet here and there. I suspect it will work fine for WC whatever you do (say less than 1000feet), but why “push it” if you can go in series or as close to series as you can? I like to do things as cleanly as possible, when I can.

BR
Dan Lavry  
Title: Re: Proper word clock implementation
Post by: Kendrix on October 13, 2004, 12:32:45 pm
Great thread.  Its confirming things I knew and adding to my understanding in some areas.

As always, the devil is in the details.
When judging the possibility that an external WC might yield better conversion than an internal crystal the specific implementation of the internal clock must be understood.

In my case I have a Yamaha Aw4416.
It comes with 16 AD's on board and I added an I/O card wtih an additional 8 AD's.  How the "internal" clock gets distibuted to all these converters is not clear to me from the schematics provided with the unit.  However, it is clear that to transmit the internal clock to the add-on I/O board requires the clock signal passing over some distance and through an interface to the board.  

To be honest when I A/B  the Yamaha internal clock with a genx WC supplied via BNC ( I use each clock for both AD and DA conversions for the comparison) I do hear differences.  However im unable to clearly judge which is most accurate.  Ive used challengeing signals, such as a bell recorded via the same chain cept for varying the clock- and trying to judge which clock yeilded a sound closest to the actual bell ( difficult to do).  When converting and playing back a well recorded piece I have no objectivre reference of what the "correct" sound should be.  Depending on the material there can be plusses and minuses to each clock.  Generally the external clock seems a bit more defined and 3D but my monitoring environment (KRK7000's in a smalll room with some treatment for reflections) is marginal for making such fine judgments.  Reconciling this observation with the statement that an internal clock should almost always be better than external clock is bothering me.  Again, the details of how the internal clock gets distributed to the 24 individual converters in my system is unclear.

So here's a question.  Are the sonic artifacts of jitter random or highly program dependant OR can they be characterized in some way.  Is it the case that jitter always produces a harshness via unintended high frequency components generated  OR is it possible that jitter can manifest itself as a warmer/muddier sound beacause the highs get "smeared out". If there is some inherent characteristic of the sonic impacts of jitter then I could better judge what is most accurate in the context of my system.  

(I Hope all the above was coherent enough to allow for meaningful responses).

Title: Re: Proper word clock implementation
Post by: bobkatz on October 13, 2004, 07:49:09 pm
Kendrix wrote on Wed, 13 October 2004 12:32



To be honest when I A/B  the Yamaha internal clock with a genx WC supplied via BNC ( I use each clock for both AD and DA conversions for the comparison) I do hear differences.  However im unable to clearly judge which is most accurate.





You might be trying to choose the lesser of two evils, or between two "goods" each of which have some minuses, Kendrix.

I doubt that measurement would settle the issue; you'll probably see jitter artifacts at different frequencies with each implementation. But then again, you might think that one is more euphonic than another or hiding something better. If the music has a lot of distortion and the jitter is random noise, the additional noise could be masking the distortion! In that case you may be preferring "more jitter". Regardless, I tend to NOT try to let clocking drive my thinking as this will prevent you from having your ultimately good sound with a good source. Instead, I would try to find the clocking which is the most accurate with a very pure and good-sounding source. Then, if you think you need to "degrade" your sound with noise or whatever when you hear a lesser source through the better clocking, keep the better clocking and try to fix your source up with some other technique (equalization, distortion generation, compression, etc.)

Let me try this:

1) Which one has more depth?

2) Which one is warmer, or tonally more accurate?

3) Which one is wider and deeper?

I'll bet you dollars to donuts that one will be the clock with the least jitter!

Hope this helps,


Bob
Title: Re: Proper word clock implementation
Post by: danlavry on October 13, 2004, 08:04:29 pm
“Great thread. Its confirming things I knew and adding to my understanding in some areas.

As always, the devil is in the details.
When judging the possibility that an external WC might yield better conversion than an internal crystal the specific implementation of the internal clock must be understood.”


Well, it is possible that some fixed frequency crystal oscillator badly designed or poorly constructed is going to be absolutely terrible. But as a rule, a crystal oscillator tends to be very stable, and the crystal itself keeps things very constant.
On the other hand, external clock call for an adjustable crystal – the crystal is cut in such a way as to be sensitive to the amount of capacitance across it, and we are talking tenth of pF’s (small capacitance).

That capacitance change is driven by a voltage change. The capacitance is in fact a reversed biased diode called varactor, specially made to change capacitance with reverse voltage. All diodes do some of that (the less DC the more capacitance). Varactors are diodes made to enhance that effect.

So where is the voltage across the diode, which changes the capacity across that voltage controlled crystal, comes from? It is an error signal being made from the difference between the input frequency and the variable crystal…. Shell I go on? A lot of factors are at play here. Just a bit noise on that control voltage and you have some jitter. Can you filter the noise out of the controlling voltage? Yes, but only down to some frequency (be it 100Hz or 1KHz), but anything below that frequency becomes jitter.

And now you still have to distribute the clock… So which one is worse? The internal clock is much better.    

“So here's a question. Are the sonic artifacts of jitter random or highly program dependant OR can they be characterized in some way. Is it the case that jitter always produces a harshness via unintended high frequency components generated OR is it possible that jitter can manifest itself as a warmer/muddier sound beacause the highs get "smeared out". If there is some inherent characteristic of the sonic impacts of jitter then I could better judge what is most accurate in the context of my system.”

Random noise makes random jitter. The noise that is correlated to music or to the data that carry the music – in fact any non random noise can modulate the jitter. Go to my web at www.lavryengineering.com and look at a paper called ON JITTER (under the support section). Look at the plots and you will get much more detail.

BR
Dan Lavry
Title: Re: Proper word clock implementation
Post by: Roland Storch on October 14, 2004, 01:04:55 pm
danlavry wrote on Thu, 14 October 2004 01:04



So where is the voltage across the diode, which changes the capacity across that voltage controlled crystal, comes from? It is an error signal being made from the difference between the input frequency and the variable crystal?. Shell I go on? A lot of factors are at play here. Just a bit noise on that control voltage and you have some jitter. Can you filter the noise out of the controlling voltage? Yes, but only down to some frequency (be it 100Hz or 1KHz), but anything below that frequency becomes jitter.

And now you still have to distribute the clock? So which one is worse? The internal clock is much better.    
BR
Dan Lavry




So what is the best solution if I need 24 AD converters for 24 tracks? Cascading 8-channel ADC with the first as master? Do the external clocked second and third 8-channel converters have a worse performance, because they are externally clocked?

Or should I prefer a 24 channel modular ADC system?
Title: Re: Proper word clock implementation
Post by: danlavry on October 14, 2004, 01:52:47 pm
Roland Storch wrote on Thu, 14 October 2004 18:04

danlavry wrote on Thu, 14 October 2004 01:04



So where is the voltage across the diode, which changes the capacity across that voltage controlled crystal, comes from? It is an error signal being made from the difference between the input frequency and the variable crystal?. Shell I go on? A lot of factors are at play here. Just a bit noise on that control voltage and you have some jitter. Can you filter the noise out of the controlling voltage? Yes, but only down to some frequency (be it 100Hz or 1KHz), but anything below that frequency becomes jitter.

And now you still have to distribute the clock? So which one is worse? The internal clock is much better.    
BR
Dan Lavry




So what is the best solution if I need 24 AD converters for 24 tracks? Cascading 8-channel ADC with the first as master? Do the external clocked second and third 8-channel converters have a worse performance, because they are externally clocked?

Or should I prefer a 24 channel modular ADC system?


For 24 tracks, you probably need to go for an external clock source and send a cable (for a clock signal) to each chassis.

My statement is: when possible, use internal clock. But when you have more than one chassis it is not possible.

My first reaction is to suggest that all clocks arrive at the same time (or nearly so) and the proposed BNC and T connection is a good one.

Regarding having one chassis as master and doing daisy chain? Probably not a good idea, because there may be some time delay from the master AD to the clock output, and additional delays in the other chassis. Does it matter? Read the thread I started yesterday about time delays.

BR
Dan Lavry  
Title: Re: Proper word clock implementation
Post by: Roland Storch on October 19, 2004, 05:58:02 pm
danlavry wrote on Thu, 14 October 2004 18:52


My first reaction is to suggest that all clocks arrive at the same time (or nearly so) and the proposed BNC and T connection is a good one.

Regarding having one chassis as master and doing daisy chain? Probably not a good idea, because there may be some time delay from the master AD to the clock output, and additional delays in the other chassis. Does it matter? Read the thread I started yesterday about time delays.

BR
Dan Lavry  



I went through the time delay thread and think time delay is not an issue in my situation, where I have a main pair/array (AB, ORTF, Decca Tree,...) with maximum 5 microphones plus spot mics.

The main pair/array will be converted by an 8-channel converter, so there is no time delay. The spot mics follow the 3:1 rule anywhere. Some are even delayed a few miliseconds (Haas-effect).

Wouldn?t it be better in this case to take the internal clock of the first converter and and take the wordclock out to clock the ADC of the 8+ channels?

The idea is that, as you said, a converter always works best with its internal clock. And I would like to have the best performance on my main pair/array.
Title: Re: Proper word clock implementation
Post by: danlavry on October 19, 2004, 07:12:37 pm
Roland Storch wrote on Tue, 19 October 2004 22:58

danlavry wrote on Thu, 14 October 2004 18:52


My first reaction is to suggest that all clocks arrive at the same time (or nearly so) and the proposed BNC and T connection is a good one.

Regarding having one chassis as master and doing daisy chain? Probably not a good idea, because there may be some time delay from the master AD to the clock output, and additional delays in the other chassis. Does it matter? Read the thread I started yesterday about time delays.

BR
Dan Lavry  



I went through the time delay thread and think time delay is not an issue in my situation, where I have a main pair/array (AB, ORTF, Decca Tree,...) with maximum 5 microphones plus spot mics.

The main pair/array will be converted by an 8-channel converter, so there is no time delay. The spot mics follow the 3:1 rule anywhere. Some are even delayed a few miliseconds (Haas-effect).

Wouldn?t it be better in this case to take the internal clock of the first converter and and take the wordclock out to clock the ADC of the 8+ channels?

The idea is that, as you said, a converter always works best with its internal clock. And I would like to have the best performance on my main pair/array.


I do not have much experience with placing mics and the 3:1 rule. The math shows that one can get into trouble when adding 2 or more signals that share a portion of the same signal. Clearly if mic A picks up say a clarinet, and mic B, pointing elsewhere picks up the same signal but with -20dB attenuation, having non equal delays will make the addition and cancellation impact to the tune of -20dB (the impact is dictated by the weaker signal). So the more you isolate the mics, the less the impact of time delay...

The time delay stuff is what happens after you have taken care of all the acoustic considerations. If you are sure that there is no significant sharing of sounds between channels of different delays, you could indeed be free to run your "mains" at an internal clock, and the others as slaves. That is of course better from jitter standpoint.

BR
Dan Lavry


Title: Re: Proper word clock implementation
Post by: Albert on October 21, 2004, 11:11:41 am
Okay, to continue regarding chaining many WC clock devices together as opposed to using a distribution box...

I've got all my BNC t-bars, terminators, and short WC cables, and have found chaining WC devices to be very fiddly. In short, I haven't been able to get it to work "as it should" although I have been able to get it to work with some fudging.

When chaining through a number of units using the t-bars, some problems occur:

A) at some point the WC signal seems to get lost and all the rest of the devices on the chain don't lock.

B) when "A" happens, often the lock will be lost all along the chain, or a good part of it.

C) WC lock can be lost by the first unit in the chain (which will then revert to internal), but everything else later in the chain locks to it fine.

D) WC seems to be very sensitive as far as which devices it "likes" and "doesn't like". Even different version of the same basic model can cause problems.

E) Without fail, WC lock is lost when using an external terminator on the last device rather than its internal termination. Let's just say I've found the chain "hates" external termination.

The units being chained are two digital mixers, Tascam DM-24's, one Akai S6000 sampler and five Akai S5000 samplers. All these units allow WC termination to be turned off. My clock source is an Aardvark AardSync II, the best clock I have.

I have been able to work around these issues by doing the following:

The Tascam's have WC in and out. So I go into the WC in on the first Tascam, then use its WC out to feed the next Tascam. That Tascam WC in has a t-bar connector, and the five Akai S5000's are fed in a chain from there. The WC output of the second Tascam is used to feed the S6000. I did it this way because while the S5000's locked together happily, they didn't when the S6000 was involved. So I had to send it signal separately.

So it's not a pure chain using t-bars only, the WC signal goes through the Tascam's in order to feed units further down the line. I've found this works very reliably. However, I have a question: Do I lose the advantages of chaining the units when using the WC "thru" on the Tascams? In other words, is that basically the  same as using a distribution box?

My next challenge is to figure out how to chain my second rack of Akai's.

Incidentally, I do think the Akai's sound better using the AardSync II clock rather than their own internal clocks. Even with my bastardized version of a wordclock chain, I've been able to lock these units to a better clock, something I wasn't able to do otherwise without spending a bundle on another distribution box.

I'll be interested in hearing any suggestions for further troubleshooting. Thanks again for this very informative thread!
Title: Re: Proper word clock implementation
Post by: bobkatz on October 21, 2004, 11:36:33 am
Albert wrote on Thu, 21 October 2004 11:11



When chaining through a number of units using the t-bars, some problems occur:

A) at some point the WC signal seems to get lost and all the rest of the devices on the chain don't lock.





I'm sure you are running into multiple terminating word clock inputs. You will have to remove the 75 ohm resistor inside the box(es) for this to work.

BK
Title: Re: Proper word clock implementation
Post by: Albert on October 21, 2004, 11:48:58 am
All the units I'm chaining have switches to turn WC termination on/off. In all cases I've turned it off, except for the last unit in the chain. That's why I'm a bit puzzled.
Title: Re: Proper word clock implementation
Post by: danlavry on October 21, 2004, 01:09:34 pm
"So it's not a pure chain using t-bars only, the WC signal goes through the Tascam's in order to feed units further down the line. I've found this works very reliably. However, I have a question: Do I lose the advantages of chaining the units when using the WC "thru" on the Tascams? In other words, is that basically the same as using a distribution box?"

I can not answer question regarding all the gear you have. But in fundamentally, the impact on a WC signal due to a short coax cable is less then going through a circuit. The circuit is made out of semiconductor device which will a some jitter. Maybe not much but a little here and a little there and it does add up. The mechanisms of increased circuit jitter are many. One is Power supply noise (to the "through" circuit). Another is due to the fact that the detection threshold is in fact not fixed, it moves around slightly (semiconductor noise). There may be other issues, such as grounding.

You do not always get to optimize everything. All it takes is one device to "mis behave". But if you can it is a good idea.
I know hooking things up and setting up gear can be complicated and it is tempting to "just getting it to work"...

It is difficult to guess what is going on there. Removing the termination at the end does increase the amplitude.
1. Do you have enough WC amplitude?
2. Is the WC source back terminated? If so, end termination will cut the signal by 1/2
3. Is one of the units stuck at internal termination?  

If your combined cables are not say thousands of feet you may just try it without termination. Read what I said about WC and reflections. At those low speeds, with pretty limited rise times and and reasonably short cables, termination just does little good, while reducing the amplitude some.

BR
Dan Lavry
Title: Re: Proper word clock implementation
Post by: bobkatz on October 21, 2004, 01:50:49 pm
Albert wrote on Thu, 21 October 2004 11:48

All the units I'm chaining have switches to turn WC termination on/off. In all cases I've turned it off, except for the last unit in the chain. That's why I'm a bit puzzled.



Oh dear. But that may be a good sign. Maybe one of the termination switches is bad or sticky. It's real easy with those BNC T's to pull them off the box while watching the scope and find the culprit.

BK
Title: Re: Proper word clock implementation
Post by: Albert on October 23, 2004, 12:35:30 pm
Thank you for your suggestions, I'm going to troubleshoot along those lines. I don't have a scope, but I can try everything else. Thanks again, I'll report back.
Title: Re: Proper word clock implementation
Post by: jcoyle on October 24, 2004, 04:00:06 pm
Hi Dan,

Some of what you are saying is contrary to what others have told me, and I am starting to feel like my mind has been filled with CRAP to make me think I have to buy a GOOD clock.

I am a home recorder.   I send my stero mic pre signal directly to an Apogee PSX-100 stereo converter, and then send the Apogee Lightpipe signal to an Alesis HD24 Hard Drive recorder.  The HD24 locks to the PSX-100 signal of course.

This is a pretty simple setup.

I was told the best thing I could do to improve my sound to tape (HDrive) would be to clock the PSX-100 with a real clock!!. The Lucid GenX6-96 was recommended.

But you are saying the internal clock of the PSX-100 will be better than externally clocking.   As I have no multiple convertors issue, requiring an external clock.

Am I a sucker for believing others up to this point???  Is everyone just clueless, has marketing hype led everyone astray?

thanks,
joey

Title: Re: Proper word clock implementation
Post by: danlavry on October 24, 2004, 05:44:19 pm
jcoyle wrote on Sun, 24 October 2004 21:00

Hi Dan,

Some of what you are saying is contrary to what others have told me, and I am starting to feel like my mind has been filled with CRAP to make me think I have to buy a GOOD clock.

I am a home recorder.   I send my stero mic pre signal directly to an Apogee PSX-100 stereo converter, and then send the Apogee Lightpipe signal to an Alesis HD24 Hard Drive recorder.  The HD24 locks to the PSX-100 signal of course.

This is a pretty simple setup.

I was told the best thing I could do to improve my sound to tape (HDrive) would be to clock the PSX-100 with a real clock!!. The Lucid GenX6-96 was recommended.

But you are saying the internal clock of the PSX-100 will be better than externally clocking.   As I have no multiple convertors issue, requiring an external clock.

Am I a sucker for believing others up to this point???  Is everyone just clueless, has marketing hype led everyone astray?

thanks,
joey




This is a technical forum. I stand by my statements that when possible to do the AD conversion with internal crystal, you are much better off doing so. Is everyone else wrong? Yes they are. I already elaborated about the difference between an internal crystal that is MADE FOR A FIXED FREQURNCY, vs. all the jitter sources when using  a VOLTAGE DEPENDENT CRYSTAL (+/- 100ppm at least), with a PLL circuit that hunts the average frequency (going up and down constantly), with a cable (pickup of external signals such as 60Hz), thus a semiconductor transmitter, and a semiconductor receiver… both very suceptible to supply noise...I did not say everything, there is more!

One can never argue with subjective opinions. But if less jitter is what you are after, use internal clock whenever possible.

BR
Dan Lavry

Title: Re: Proper word clock implementation
Post by: bobkatz on October 24, 2004, 07:16:24 pm
jcoyle wrote on Sun, 24 October 2004 16:00




But you are saying the internal clock of the PSX-100 will be better than externally clocking.   As I have no multiple convertors issue, requiring an external clock.

Am I a sucker for believing others up to this point???  Is everyone just clueless, has marketing hype led everyone astray?

thanks,
joey





Dan is saying that, "all other things being equal, in a well-designed converter, internal clock will always perform better than, or equal to an external clock."

You have to prove whether the Apogee is a well-designed converter! If it's a poor converter, it may very well perform better on external clock. A left-handed "compliment" if I ever saw one!

And yes, everyone is clueless... very very few people have grasped the implications of what Dan has to say. John Watkinson, among many other competent writers, myself included, have been haranguing this topic for a long time. Manufacturers, it seems are the most clueless.

The BEST article I've seen on this topic from a competent manufacturer is TC Electronic. Go to

http://www.tcelectronic.com/TechLibrary


and download the article: "Clock and Synchronization in the TC System 6000". It demonstrates, measures, and summarizes the issues for all to see that when it comes to poorly-designed converters, the Emperor has no clothes. Every external clock that makes (or seems to make ) a converter "sound better" is a bandaid, not a cure.
Title: Re: Proper word clock implementation
Post by: Joe Bryan on October 25, 2004, 01:54:01 am
danlavry wrote on Sun, 24 October 2004 14:44

This is a technical forum. I stand by my statements that when possible to do the AD conversion with internal crystal, you are much better off doing so. Is everyone else wrong? Yes they are.


Dan is 100% correct. It's extremely unlikely an external clock can outperform an internal clock unless there are serious problems with the design.

I wrote an article about this for our webzine a while back. It's available here:  http://www.uaudio.com/webzine/june/content/content2.html

-Joe

Joe Bryan
VP Engineering
Universal Audio
Title: Re: Proper word clock implementation
Post by: bobkatz on October 26, 2004, 09:59:23 am
Joe Bryan wrote on Mon, 25 October 2004 01:54


Dan is 100% correct. It's extremely unlikely an external clock can outperform an internal clock unless there are serious problems with the design.





Obviously, Joe, you're also one of the good guys! In my measurement and listening tests I've encountered quite a few defective converters! In the "not-defective" department, it took TC Electronic ONE MAN YEAR to design the PLL circuit currently in use in the System 6000. It measures one tiny tiny bit "less good" than its internal crystal! As to whether the amount of jitter that remains in the TC on external clock is audible, I doubt it.

BK
Title: Re: Proper word clock implementation
Post by: Roland Storch on October 27, 2004, 04:55:42 am
Reading all this for me it seems the best way of clocking for a hybrid studio (mixing digital with digital and analog outboard) is to use an integrated AD/DA converter where both the ADC and DAC are clocked internally by a very stabil clock. And this clock at the same time is the master clock for all digital devices.

Also for monitoring with a better outboard DAC than the internal DAC of a digital mixing desk it would be better if the DAC for monitoring is the master. I just compared the sound of the DAC-1 and another DAC (Digital Audio Denmark) against a very good CD player by hearing CDs through the analog output of the player or the digital output converted by DAC-1 or DAD.

The CD player analog out to me sounded always better than either the DAC-1 or the DAD (regardless if connected optical or with BNC). This confirms the statements by Dan, Bob and all the other appreciated technical explanations, that the internal clock outperforms an external clock.

BTW: I prefer either the DAC-1 or the DAD to the analog out of my digital mixer. But the DAC of my 1500,-
Title: Re: Proper word clock implementation
Post by: jcoyle on October 27, 2004, 10:03:53 am
I may be wrong here, but with multiple DACs, they don't need to be word clocked together...    Because they clock to the incoming digital audio, buffer this, and then tightly clock this internally and convert it to Analog.

I think the need for external clocking of multiple hardware, is when you have multiple devices with digital outputs that send their digital outputs to another common digital device.

please correct if this is wrong.

joey
Title: Re: Proper word clock implementation
Post by: danlavry on October 27, 2004, 11:19:23 am
The issue is different for AD and DA:

Clock jitter on the AD will ALTER THE DATA FOREVER.  

Clock Jitter on DA will only impact the specific playback. You can play it again "later" through another DA and if the device is better (or worse), it would sound better (or worse).

Also, it is not always possible to have an AD and DA of the same clock, from jitter standpoint. Just a few gate and flip flop circuit delays, and you are not as clean as the crystal oscillator. Of course it still beats the external clock with cable, receiver, PLL, voltage controlled crystal (VCXO)...  


BR
Dan Lavry
Title: Re: Proper word clock implementation
Post by: Kendrix on October 27, 2004, 12:51:24 pm
Roland Storch wrote on Wed, 27 October 2004 09:55

Reading all this for me it seems the best way of clocking for a hybrid studio (mixing digital with digital and analog outboard) is to use an integrated AD/DA converter where both the ADC and DAC are clocked internally by a very stabil clock. And this clock at the same time is the master clock for all digital devices.

Also for monitoring with a better outboard DAC than the internal DAC of a digital mixing desk it would be better if the DAC for monitoring is the master. I just compared the sound of the DAC-1 and another DAC (Digital Audio Denmark) against a very good CD player by hearing CDs through the analog output of the player or the digital output converted by DAC-1 or DAD.

The CD player analog out to me sounded always better than either the DAC-1 or the DAD (regardless if connected optical or with BNC). This confirms the statements by Dan, Bob and all the other appreciated technical explanations, that the internal clock outperforms an external clock.

BTW: I prefer either the DAC-1 or the DAD to the analog out of my digital mixer. But the DAC of my 1500,-
Title: Re: Proper word clock implementation
Post by: Max on October 27, 2004, 04:34:11 pm
Lucas Van Der Mee, Apogee's Senior Design Engineer, has been following this thread and felt compelled to respond to some of the things being said. Since Lucas is already at the AES show, I have taken the liberty to post some comments on his behalf:

jcoyle wrote on Sun, 24 October 2004 16:00




But you are saying the internal clock of the PSX-100 will be better than externally clocking.   As I have no multiple convertors issue, requiring an external clock.

Am I a sucker for believing others up to this point???  Is everyone just clueless, has marketing hype led everyone astray?

thanks,
joey





bobkatz wrote


Dan is saying that, "all other things being equal, in a well-designed converter, internal clock will always perform better than, or equal to an external clock."


True, but all things are never equal when it comes to clocking. Anyone who is using a D/A converter of any kind is re-clocking, for example.

bobkatz wrote


You have to prove whether the Apogee is a well-designed converter! If it's a poor converter, it may very well perform better on external clock. A left-handed "compliment" if I ever saw one!


The only external clock that will improve the clock performance of the PSX-100 is the C777 used in Big Ben. Incidentally, this clock also improves performance of every converter that we have tested, with the exception of SRC based D/A converters.  Based on this criteria, every high-end converter with a PLL design is a "poor converter" other than the AD-16X and DA-16X, which include the C777 clock.

bobkatz wrote


And yes, everyone is clueless... very very few people have grasped the implications of what Dan has to say. John Watkinson, among many other competent writers, myself included, have been haranguing this topic for a long time. Manufacturers, it seems are the most clueless.


Not everyone is clueless, it is just that some people are still living in the 20th century technology-wise and clinging to theories that don't hold up in objective listening tests. There is plenty of empirical evidence that the C777 clock disproves this theory and has the impact on sound performance that is being refuted on this forum.

The problem with discussing the how and why of Apogee's advanced science in this field is that there are things we are doing that we do not want to advertise to our competitors, frankly, yet without mentioning those things, it becomes hard to mount a credible argument against the theorists. Based on this, my suggestion is for all those interested in reality to perform an objective listening test and make the determination with your ears, and not a text book.
Title: Re: Proper word clock implementation
Post by: Roland Storch on October 27, 2004, 06:09:56 pm
Kendrix wrote on Wed, 27 October 2004 17:51



I think you need to be carefulo abou attributing any perceived sonic differences to the clock.  ther are analog circuits at the output stage of the DA converters that must be accounted for.  FYI- the DAC1 claims to eliminate jitter on incomming digital signals via a propietary approach (I understand it does not use a PLL based on the clock embedded in the input signal).

So, I dont think you can conclude that jitter was the cause of any sonic differences you heard in this case.



I admit you are probably right.
On the other side I hope the reason for sonic differences was the clock because the afformentioned 1500
Title: Re: Proper word clock implementation
Post by: Terry Demol on October 28, 2004, 10:20:10 am
Max wrote on Wed, 27 October 2004 21:34

Lucas Van Der Mee, Apogee's Senior Design Engineer, has been following this thread and felt compelled to respond to some of the things being said. Since Lucas is already at the AES show, I have taken the liberty to post some comments on his behalf:

jcoyle wrote on Sun, 24 October 2004 16:00




But you are saying the internal clock of the PSX-100 will be better than externally clocking.   As I have no multiple convertors issue, requiring an external clock.

Am I a sucker for believing others up to this point???  Is everyone just clueless, has marketing hype led everyone astray?

thanks,
joey





bobkatz wrote


Dan is saying that, "all other things being equal, in a well-designed converter, internal clock will always perform better than, or equal to an external clock."


True, but all things are never equal when it comes to clocking. Anyone who is using a D/A converter of any kind is re-clocking, for example.

bobkatz wrote


You have to prove whether the Apogee is a well-designed converter! If it's a poor converter, it may very well perform better on external clock. A left-handed "compliment" if I ever saw one!


The only external clock that will improve the clock performance of the PSX-100 is the C777 used in Big Ben. Incidentally, this clock also improves performance of every converter that we have tested, with the exception of SRC based D/A converters.  Based on this criteria, every high-end converter with a PLL design is a "poor converter" other than the AD-16X and DA-16X, which include the C777 clock.

bobkatz wrote


And yes, everyone is clueless... very very few people have grasped the implications of what Dan has to say. John Watkinson, among many other competent writers, myself included, have been haranguing this topic for a long time. Manufacturers, it seems are the most clueless.


Not everyone is clueless, it is just that some people are still living in the 20th century technology-wise and clinging to theories that don't hold up in objective listening tests. There is plenty of empirical evidence that the C777 clock disproves this theory and has the impact on sound performance that is being refuted on this forum.

The problem with discussing the how and why of Apogee's advanced science in this field is that there are things we are doing that we do not want to advertise to our competitors, frankly, yet without mentioning those things, it becomes hard to mount a credible argument against the theorists. Based on this, my suggestion is for all those interested in reality to perform an objective listening test and make the determination with your ears, and not a text book.



Hi Max,

Do you have a phase noise graph for the C777 in dBc versus F(Hz)
away from carrier.

This is an effective way to compare it's measured performance to
a good VCXO or ULN XO.

The first people I know to use DDS for clocking in audio
converters was Mark Levinson. They used the AD9850 chip which
was pretty good WRT phase noise if not quite up to the spec
of a good XO. There are probably better ones available now.

Cheers,

Terry


Title: Re: Proper word clock implementation
Post by: bblackwood on October 29, 2004, 03:05:16 am
Max wrote on Wed, 27 October 2004 15:34

Not everyone is clueless, it is just that some people are still living in the 20th century technology-wise and clinging to theories that don't hold up in objective listening tests. There is plenty of empirical evidence that the C777 clock disproves this theory and has the impact on sound performance that is being refuted on this forum.

So I assume you've tried used the 'big ben' to reclock the da924 for better results? I mean, it's using '20th century technology' so it must suck, right?
Title: Re: Proper word clock implementation
Post by: bobkatz on November 02, 2004, 12:06:19 am
bblackwood wrote on Fri, 29 October 2004 03:05

Max wrote on Wed, 27 October 2004 15:34

Not everyone is clueless, it is just that some people are still living in the 20th century technology-wise and clinging to theories that don't hold up in objective listening tests. There is plenty of empirical evidence that the C777 clock disproves this theory and has the impact on sound performance that is being refuted on this forum.

So I assume you've tried used the 'big ben' to reclock the da924 for better results? I mean, it's using '20th century technology' so it must suck, right?


Sarcasm will get you everywhere, Brad!

But seriously: The bottom line is that if an EXTERNAL clock improves or even CHANGES the sound of a converter, then its internal crystal is not well designed and its phased locked loop is not effective enough. This is fact, not conjecture.

So, the conclusion is that if the Big Ben improves the sound of the Apogee, then its internal clock has to be improved.

The goal of the designers of the new Grimm AD-1 converter expresses quite clearly their clear understanding of the issue:

"Our goal was to produce a converter whose performance does not change nor degrade when on internal or external clock."

The bottom line IF the Big Ben improves the sound of the Apogee is that it is a band aid, not a cure.

BK
Title: Re: Proper word clock implementation
Post by: Max on November 02, 2004, 07:38:35 pm
bblackwood wrote on Fri, 29 October 2004 08:05

Max wrote on Wed, 27 October 2004 15:34

Not everyone is clueless, it is just that some people are still living in the 20th century technology-wise and clinging to theories that don't hold up in objective listening tests. There is plenty of empirical evidence that the C777 clock disproves this theory and has the impact on sound performance that is being refuted on this forum.

So I assume you've tried used the 'big ben' to reclock the da924 for better results? I mean, it's using '20th century technology' so it must suck, right?


Yes, and that does not mean the box "sucks", as you put it. As a matter of fact, that is just the opposite of what I am saying. My point is that there is no perfect converter or perfect clock and to say that when an external clock helps a good converter means that the converter is flawed is to say that you probably have not tried Big Ben, which Bob Katz has not. As a matter of fact, Bob has not tried any Apogee products since the original AD-8000 (seven years old), so most of what he is saying in this thread is based on history, if not histrionics.

There are many golden-eared end users with high end gear that have purchased Big Ben because they understand what I am talking about. Bob Ludwig, Doug Sax, Bernie Grundman, Steve Marcussen, Clark Germain, Jack Joseph Puig, Ronald Prent, and Steve Levine, to name a few off the top of my head. Not all of these guys use Apogee converters, and most who do use Apogee converters use others as well. Yet, what they all have in common is Big Ben.... Does their gear suddenly suck after all this time because Big Ben makes it sound good? No. Call Big Ben an exception to the rule, or not, since that would mean that you might actually have to listen to the product to make a determination, and who would want to do that, right?
Title: Re: Proper word clock implementation
Post by: chap on November 02, 2004, 08:48:30 pm
I don't mean to stir the pot but a couple of things come to mind.
First of all, most studios have at least 2 digital devices.
Most have more.  How would we run our studios ( I have 10 devices that need to speak to each other) unless they are properly clocked?
Regardless of whether a stand alone converter sounds better with it's internal clock, it's not practical to use a single digital device by itself.  I like using my tc6000 with my converters.
It simply won't work unless they are mutually clocked.
So is the argument academic with regards to practical application?

The second is more anecdotal but since I'm in the anecdote, it's true.
This summer, I was mastering a project at Sterling Sound with my favorite mastering engineer, Greg Calbi.  We have been working hard to try and popularize the notion of 'stem' mastering.  (Hopefully a new president will encourage this 'stem' research).
After working for awhile, I suggested that he might like Big Ben for stemming.  I arranged for him to try one.
A week later he called me and said it did exactly what I suggested it would.
Tighter low end, better depth of field and a wider image.
He wasn't proclaiming Ben as a miracle box.  He was saying that it did exactly what it was supposed to do.  He kept the box.
Greg has some of the most amazing ears in the business and I would not drop his name casually (this is the first post where I've mentioned him) but I think it lends credibility to the notion that a qualified listener can hear the improvements that Ben imparts on multiple digital device situations.
I should also say that Greg appears to remain happy with his dB
converters.
best,
chap
Title: Re: Proper word clock implementation
Post by: bobkatz on November 02, 2004, 11:39:18 pm
Max wrote on Tue, 02 November 2004 19:38



tried Big Ben, which Bob Katz has not. As a matter of fact, Bob has not tried any Apogee products since the original AD-8000 (seven years old), so most of what he is saying in this thread is based on history, if not histrionics.

There are many golden-eared end users with high end gear that have purchased Big Ben because they understand what I am talking about. Bob Ludwig, Doug Sax, Bernie




I am pissed because this thread is being taken over by salesmanship and ignorance of the basic design constraints of converters. Bob Ludwig knows better, he's got the tech chops. So I suspect you're taking him out of context.

If I gave you an endorsement, like Ludwig, Calbi and Sax, of how "great" the Apogee sounds with the Big Ben it is really a political spin of the converse.  I really should be giving you a negative endorsement of how poorly it sounds WITHOUT THE BIG BEN.

By the way, I have nothing personally against you, Max, but my anger comes from that you approached this thread as a salesman and without concrete knowledge of how your own equipment works. Otherwise I would not be so confrontational. I hope you will learn something from this post; Apogee makes excellent converters, but there is great room for improvement once you learn the science and the facts and forget the hype....

Saying that a converter sounds better with a "great" external clock is attempting to justify poor engineering. You are rationalizing the purchase of a Big Ben when the whole purpose of a converter should be to AVOID the NEED for a Big Ben! It is a big Left-handed compliment to say that a converter sounds better with an external clock!

Let's remember how a PLL works and how an internal crystal works, as described clearly by Dan Lavry earlier in this thread.

I really don't care how much the Apogee is improved by the Big Ben. I can buy numerous well-designed converters which perform as well or better without it, and reject the jitter well-enough from external clocks so as not to need the Big Ben.

The very fact that it is IMPROVED is a demonstration that not enough attention was made to the internal design of the converter itself. That includes both the PLL and the internal crystal. Both of which should perform excellently without any external BANDAIDS, because that's what the Big Ben is, a BANDAID. The CURE, not the BANDAID, should have been within the converter's internal design not external to it. That includes the PLL and the crystal. The ABSOLUTE best performance of a well-designed converter MUST be that the internal crystal performance is indistinguishable from the external PLL performance. If the PLL performs one iota better than the crystal, then the designers have not done their homework.

For the record, I have auditioned two other Apogee converters post the AD-8000 and been underwhelmed by their sonics. I found the Rosetta and the PSX-100 to be colored and veiled, under blind tests, compared with a straight wire, analog sources, and against other models of converters. I am prepared to be impressed by the later designs; everyone makes a mistake sometime. But don't mess with me with this Big Ben Nonsense. The science and the basic principles of engineering design are against you. This is not-brand-specific. There is no mysterious "voodoo" about the Big Ben and what it is or is not doing for the Apogee or any other converter.

It costs about $35 in parts, coupled with good grounding, layout and power supply design, to build a crystal oscillator in your converter that will outperform any external Big Ben or whatever. So quit the hype, please.

Was it an economic decision to cripple the Apogee? I can understand if you want the consumers to save $600 (or whatever the Big Ben costs), buy the converter, get fair performance without it at a decent price (I assume).  If you are willing to admit that it was an economic decision to design the converter in this way, then I have no quarrel with you. Every designer has to face economic compromises, and maybe the Apogee's clock circuit was one of those compromises.

Just don't tout the Big Ben as a big achievement when it is really the equivalent of a better PCV valve on a poorly-designed internal combustion engine!

Do you have FFT measurements of your converter's jitter performance on internal and external clock comparable to that of, for example:

DCS, Lavry, Benchmark, or TC system 6000.

All of the above converters do not require a Big Ben to demonstrate textbook-perfect jitter performance.

Let's see your measurements, stick to the facts and skip the hype. In a converter that sounds poorly or worse with poor external clocks, or poorly with internal clock, it will demonstrate poor jitter rejection and poor internal crystal performance. So you only get an A+ with Superman on the external wordclock. You'll have to justify that decision with economics. If the cost of the Apogee WITH Big Ben justifies its sonics, then ok... but it's not a good sign.
Title: Re: Proper word clock implementation
Post by: Bob Olhsson on November 03, 2004, 12:18:48 pm
FWIW Bob Ludwig made it very clear during his workshop the other day at AES that what sounds best to him in his room is using the internal clock of an A to D and, for playback only, clocking the entire system off the internal clock of the D to A converter. He spoke very highly of using the Big Ben for locking to video but never suggested he uses it in place of an internal clock when that's possible.
Title: Re: Proper word clock implementation
Post by: Max on November 03, 2004, 02:42:52 pm
[quote title=bobkatz wrote on Wed, 03 November 2004 04:39]
Max wrote on Tue, 02 November 2004 19:38



tried Big Ben, which Bob Katz has not. As a matter of fact, Bob has not tried any Apogee products since the original AD-8000 (seven years old), so most of what he is saying in this thread is based on history, if not histrionics.

There are many golden-eared end users with high end gear that have purchased Big Ben because they understand what I am talking about. Bob Ludwig, Doug Sax, Bernie




Quote:

I am pissed because this thread is being taken over by salesmanship and ignorance of the basic design constraints of converters. Bob Ludwig knows better, he's got the tech chops. So I suspect you're taking him out of context.


The thread is not being taken over by salesmanship, it has been rife with salesmanship from the beginning.

Quote:

If I gave you an endorsement, like Ludwig, Calbi and Sax, of how "great" the Apogee sounds with the Big Ben it is really a political spin of the converse.  I really should be giving you a negative endorsement of how poorly it sounds WITHOUT THE BIG BEN.


Again, your own salesmanship shows itself. You of course keep pointing to the Apogee converters in the context of Big Ben, yet refuse to acknowledge the entire point of my post: that many other converters mentioned in this thread are also improved when Big Ben is used as the master clock. Does that mean that all of these other converters sound poorly as well? Are they all poorly designed? I don't think so.

Quote:

 By the way, I have nothing personally against you, Max, but my anger comes from that you approached this thread as a salesman and without concrete knowledge of how your own equipment works.


With all due respect, Bob,  I have a very solid foundation in how our equipment works. On top of that, I am sitting here every day surrounded by some of the best engineers in this business, whom I have consulted and completely agree with everything I have posted.

Quote:

 Otherwise I would not be so confrontational. I hope you will learn something from this post; Apogee makes excellent converters, but there is great room for improvement once you learn the science and the facts and forget the hype....


I agree with you that there is always room for improvement. There is also always room for new and better ideas, and theories are not facts until they are proven scientifically. Even then science has a way of breaking through and changing what we previously believed to be right and prove all of us wrong over and over. I would also add that the belief in what sounds good is not scientific or objective in any way. To say that the science disproves something that can be qualified audibly is to say that the science is incomplete at this time or to say that sound and music is subjective and still more art than science, even when it comes to engineering.


Quote:

 Saying that a converter sounds better with a "great" external clock is attempting to justify poor engineering. You are rationalizing the purchase of a Big Ben when the whole purpose of a converter should be to AVOID the NEED for a Big Ben! It is a big Left-handed compliment to say that a converter sounds better with an external clock!

Let's remember how a PLL works and how an internal crystal works, as described clearly by Dan Lavry earlier in this thread.

I really don't care how much the Apogee is improved by the Big Ben. I can buy numerous well-designed converters which perform as well or better without it, and reject the jitter well-enough from external clocks so as not to need the Big Ben.


Well, again there are other converters improved by Big Ben besides the Apogees, perhaps even in your own studio. When one of our engineers contacted you and offered you a chance to try Big Ben and find out for yourself, you turned us down.

Quote:

There is no mysterious "voodoo" about the Big Ben and what it is or is not doing for the Apogee or any other converter.


Actually, Bob, there is. I would be happy to arrange to get you a Big Ben to try for yourself. At this point, of course, I would insist that at least one of us were there to insure that the test was done as objectively as possible, since I believe you are a bit biased against Apogee for whatever reason and this bias may unfairly skew your results if left to your own devices.

Quote:

Was it an economic decision to cripple the Apogee? I can understand if you want the consumers to save $600 (or whatever the Big Ben costs), buy the converter, get fair performance without it at a decent price (I assume).  If you are willing to admit that it was an economic decision to design the converter in this way, then I have no quarrel with you. Every designer has to face economic compromises, and maybe the Apogee's clock circuit was one of those compromises.


No compromise in the Apogee clock circuit. As a matter of fact, it is one of the best performing analog PLLs on the market. It is just nothing compared to the C777, which by the way, is the standard clock in the AD-16X and DA-16X converters. These two converters of course are not improved with Big Ben, for obvious reasons.

Quote:

Just don't tout the Big Ben as a big achievement when it is really the equivalent of a better PCV valve on a poorly-designed internal combustion engine!


The C777 is a lot more than that, Bob.

Quote:

Do you have FFT measurements of your converter's jitter performance on internal and external clock comparable to that of, for example:

DCS, Lavry, Benchmark, or TC system 6000.

All of the above converters do not require a Big Ben to demonstrate textbook-perfect jitter performance.

Let's see your measurements, stick to the facts and skip the hype. In a converter that sounds poorly or worse with poor external clocks, or poorly with internal clock, it will demonstrate poor jitter rejection and poor internal crystal performance. So you only get an A+ with Superman on the external wordclock. You'll have to justify that decision with economics. If the cost of the Apogee WITH Big Ben justifies its sonics, then ok... but it's not a good sign.


You of all people should know that the measurements wont prove everything. Almost every converter on the market measures perfectly at this point, yet they all sound different. Why, because designing a converter is about making choices; choices on where to minimize the loss inherent in conversion and emphasize the sound quality.  

At the end of the day, this is all still about how it sounds. Again, there are too many people out there insisting that Big Ben improves the sound of what are to be considered great sounding converters with good internal clock circuitry. To discount what all of these folks are hearing and then not be willing to listen for yourself is disingenuous at best.
Title: Re: Proper word clock implementation
Post by: chap on November 03, 2004, 04:17:52 pm
Hi Bob,
I just want to clarify a few things as this seemed to be about clocking and has now become a personal attack thread.
I've read your book and respect your contributions to the field.
I've used Dan's converters and they sound great.
I, in no way, implied that Calbi 'endorsed' Big Ben and feel that
you may have taken his name out of context.
If you read my post (after you calm down), you'll see that Greg told me that Ben did exactly what it was supposed to do.
He bought it but I don't see his name on the Apogee site and it embarrasses me that his name is used out of context.

I've known Max for some time now.  He is quite careful not to appear on a forum in the guise of a salesman.  In fact, VP of Sales is only a title.  Max was involved in design aspects of the new series as were other Apogee employees.  I'm sure you're aware of Apogee's history and that they are a small company requiring people to wear many hats.  Max can converse on a technical level with most people here.  This thread was about clocking and not Apogee converters.  I think that by narrowing your focus to Apogee products of years gone by, you miss the point.
Again, my tc6000 sounds better when clocked to Big Ben.
I am not nor have I ever been employed by an equipment manufacturer.  I make music for a living.  Everyday.
I don't have the luxury of using my digital gear 1 piece at a time so I need a clock.  I see no evidince of salesmanship taking place here.  I see someone trying to clarify a point and someone else jumping down their throat.  Shouldn't this be a cool, informative forum and not Lord of the Flies?
Sincerely,
chap
Title: Re: Proper word clock implementation
Post by: bobkatz on November 03, 2004, 04:28:34 pm
[quote title=Max wrote on Wed, 03 November 2004 14:42

At the end of the day, this is all still about how it sounds.

[/quote]

Yup! How it sounds. That's why I care, too. Which should be evident from my messages. I just won't go around bragging about a converter that improves when it should have been better-designed in the first place.

So it's ok to clobber together a car out of a bunch of partseven if we can make it run better by building it better from scratch?

Quote:



Again, there are too many people out there insisting that Big Ben improves the sound of what are to be considered great sounding converters with good internal clock circuitry.




You should be a politician. The English language makes it clear what the word "great" means. I hesitate to call a converter a "great" converter in the same sentence which also says that it can be improved! Take your hyperbole down a notch. A GREAT converter, by definition, is one which already sounds good and CANNOT be improved by any external means.

Which by inference makes it clear that there are not very many good converters out there! I've only found a half dozen or so that meet the above definition.

Let me get this straight:

1) Apogee is saying that you're offering the Big Ben as a service in order to improve the sound of other manufacturer's converters? When syncing to video, for instance? Or just to replace the internal sync?

2) But not your own models? Except for the models of Apogee that do not contain the Big Ben internally? Are you selling it strictly to improve the sound of older Apogees that do not contain Big Ben? You're not selling it to improve "all" Apogees per se? (Only to improve the sound of Apogee Converters that do not contain the Big Ben)

In 35 years of design, listening and testing experience, I can only name HALF A DOZEN or so great converters, that meet the above definition. I also own a "Big Ben Quality" external clock that I can use for testing converters; it is a VCXO with extremely low phase noise in a test box designed by Fred Forssell and Bascom King, recognized experts in this field.

I wonder if any of these "great" converters which users found improved by the Big Ben are part of the half dozen or so that are really great. Send me a list privately, please.
And if I own or can find any of them, then I will test them. I wager that I will discover that these converters on their own are not as good performers in the first place as they are claimed to be by their owners.

In the pantheon of the probably 50 to one hundred brands and models of converters that have been manufactured, half a dozen is a very small subset.

Quote:



To discount what all of these folks are hearing and then not be willing to listen for yourself is disingenuous at best.


Max, read my post again, I DID NOT DISCOUNT WHAT THESE FOLKS ARE HEARING. I simply said, "So What!" I've already heard many converters dramatically improve with external clock. But I do not own, nor do I wish to own any of them. Thank you, Apogee, for making inferior converters sound better....

I also don't remember ever getting a call or offer from Apogee to audition their Big Ben. I have gotten calls from you to audition the Apogee converters and in my copious free time I'd love to....

I have auditioned (and measured) many brands and models of converters in the past 25 years. I've even built one rather advanced converter myself and I know all of its defects and virtues. In all this time, I have NEVER heard a converter which was improved by an external clock that could not be further improved by redesigning its internal circuits. Or, better still, replacing it with a jitter-immune model, perhaps even one of your newest models.

Any engineer who raves over the sound of their improved converters with the Big Ben should be complaining to the manufacturer of the converter who didn't design a "Big Ben quality" clock in their converter in the first place.

If a converter's sonics and performance does not meet up with the definition of "great" at the top of this message, the I choose not to own it.
Title: Re: Proper word clock implementation
Post by: Albert on November 03, 2004, 05:20:00 pm
Okay, I hate to interrupt this heady and intense discussion with the ridiculous, but I have a simplistic nuts and bolts type question.

First, most dedicated WC generators have multiple outputs, like the AardSync II I use, the Big Ben, and others. However, most AD/DA converters have only one WC output.

So, while I would prefer to chain all my digital devices from a master clock, as has been recommended, I'm finding that to not work reliably for reasons I haven't quite sorted out yet. This makes he concept of the whole studio clocked off one WC source difficult if not impossible for me.

So here's the question, and it may be unanswerable:

Which is better: to take the WC of your best converter into a WC distribution device and chain from there, or to have a master WC generator with multiple outs feeding the WC chains directly? In that second scenario the converter would be locked to the master clock, not its internal clock. Assume in both setups that high quality converters and WC generators are being used. Which scenario will introduce the most jitter?

Again, I am pondering such possibilities because I have yet to be able to successfully set up one long WC chain, although several shorter chains off the same clock are working.
Title: Re: Proper word clock implementation
Post by: danlavry on November 03, 2004, 06:15:06 pm
Albert wrote on Wed, 03 November 2004 22:20

Okay, I hate to interrupt this heady and intense discussion with the ridiculous, but I have a simplistic nuts and bolts type question.

First, most dedicated WC generators have multiple outputs, like the AardSync II I use, the Big Ben, and others. However, most AD/DA converters have only one WC output.

So, while I would prefer to chain all my digital devices from a master clock, as has been recommended, I'm finding that to not work reliably for reasons I haven't quite sorted out yet. This makes he concept of the whole studio clocked off one WC source difficult if not impossible for me.

So here's the question, and it may be unanswerable:

Which is better: to take the WC of your best converter into a WC distribution device and chain from there, or to have a master WC generator with multiple outs feeding the WC chains directly? In that second scenario the converter would be locked to the master clock, not its internal clock. Assume in both setups that high quality converters and WC generators are being used. Which scenario will introduce the most jitter?

Again, I am pondering such possibilities because I have yet to be able to successfully set up one long WC chain, although several shorter chains off the same clock are working.


Go to the thread:
"Time delay problems, real or not?"

Read the first message I posted. It sugests that it is a good idea to keep all time delays equal when there are common signals on different channels to be mixed.

But if you can be sure that you are not dealing with such a case, I would go for "best converter is a master clock".

Most people I know go for an external master. Some very good pro's insist on equal delay (while using good gear of course).

BR
Dan Lavry
Title: Re: Proper word clock implementation
Post by: danlavry on November 03, 2004, 06:41:26 pm
My reply (to Max) consists of 3 main points:

1. This is a technical forum. Your statements change the focus away from technical into personal.

2. Jitter destroys proper AD conversion. Your clock is based on the wrong technology (DDS) when lowest jitter is a goal. DDS is worse (more jitter) than crystal based clock for both internal and external applications.

3. An external clock is necessary in some cases but is inferior to internal clocking. You are selling external clock as an improvement for applications where internal clock is possible.


You said:

“Not everyone is clueless, it is just that some people are still living in the 20th century technology-wise and clinging to theories that don't hold up in objective listening tests. There is plenty of empirical evidence that the C777 clock disproves this theory and has the impact on sound performance that is being refuted on this forum. “

Max,

This forum is not about listening tests! It is not about “empirical evidence” based on listening tests either. Jitter is something to be measured, just like voltage, or frequency. My frequency counter is 1000 more accurate than any piano tuner’s ear, and my voltmeter is better than feeling voltage with my fingers. Your objective listening tests and empirical evidence is marketing, not technical.

If we are going for LOW JITTER, then the DDS technology is INFERIOR to a dedicated correctly done crystal oscillator!!! The AD9850 and similar DDS devices provide flexibility, and LOW COST (about $25 for the IC) but are more jittery than the crystal circuits that I use for in my LavryBlue converters. Using a AD9850 or DDS technology is terribly inappropriate.

A quick look at the IC data sheets show much higher jitter for DDS devices than for crystal oscillators. Even the basic crystal for “clock in” to the DDS chip is at best a very high frequency crystal, thus it runs on an overtone, not on fundamental resonance. An overtone crystal is much more jittery than one oscillating in a fundamental mode, not to mention the other circuits in the signal path. Those of us who really care about jitter, pay attention to such facts.

I will be speaking about “Jitter” in Feb.2005 at the New York AES meeting and April ’05 and in Montreal at the Annual Symposium in Montreal for the Association des Professionals en Audio. Also, if interested you can read my white paper on Jitter on my website under “Support”


“You said

The problem with discussing the how and why of Apogee's advanced science in this field is that there are things we are doing that we do not want to advertise to our competitors, frankly, yet without mentioning those things, it becomes hard to mount a credible argument against the theorists. Based on this, my suggestion is for all those interested in reality to perform an objective listening test and make the determination with your ears, and not a text book.

Max Gutnik
Director of Sales
Apogee Electronics”


Regarding your statement "it becomes hard to mount a credible argument against the theorists". My reply is: What do you suggest? To forgo any argument because of your claim regarding “things we are doing that we do not want to advertise to our competitors”? We should all just believe what you say, stay away from science, engineering and measurements, or else be labeled a book worm theorist of the last century?

You said “argument against the theorists.” Max, I am not just a theorist. I am a designer and you should know it. I designed the first AD and DA converters that Apogee made their name on, and you are still using the analog soft saturation concept (or circuits?) from those days. My friend Jerry Goodwin and I developed the UV 22 (though it was in the last century), and I am the guy that designed the dB technologies products, which is now Lavry Engineering. You are wrong to suggest that I am only a theorist. I am a theorist AND a “hands on” design engineer. And if I ever learned anything, it takes BOTH theory and practice to be an industry leader! Good theoretical understanding is THE KEY to good practice in my field of electronics.

I find your statements and general attitude towards technologists (and that does include theory) disrespectful. Your “Apogee's advanced science in this field “ is based on buying IC’s filled with concepts developed by theorists. Where would your products be without theoretical developments of concepts such as over sampling, up sampling, sigma delta… all concepts from the last century. I have recently seen some unqualified people take a stand against the fundamental Nyquist theory of the early 20th century, advocating 192KHz sampling rate. Audio is full of marketing hype!

What qualifies you to talk down at theorists and draw lines between advanced science of the 21st century and the old last century? My challenge to you is to come up with an electronic product that is not based on 20th century theory.

I hope you are not trying to suggest in your statement “how and why of Apogee's advanced science in this fieldis that there are things we are doing that we do not want to advertise to our competitors” that you have figured how to greatly improve on Analog Devices IC specifications. The finished product is not going to be as good as the stand alone device specifications. That AD9850 DDS device has only one hard jitter spec (for one tiny section, the comparator at 80ps p-p jitter!). There is NO WAY a PLL (phase lock loop) device can get near a fixed crystal, Period! The DDS IC’s are great for many applications, including synthesis of a lot of frequencies (we need only a few) and phase modulation (in the audio case we need zero phase modulation). Not good enough to enter the arena of very low jitter fixed clocks. Crystal technology is.

Regarding your comments about the 20th century and “Apogee's advanced science in this field”: Indeed the last 5 years, belong to the 21st century. And here I am, the old relic, clinging to old physics of the last century.

But the real point in that thread is not just about the fact that DDS puts out MORE JITTER than an average crystal design.

There are times when one needs an external crystal clock, such as when one is operating a lot of chassis at the same time. However, there are numerous times when one can operate with an internal clock, such as stereo AD conversion, or an 8 ch. AD in one chassis and more. One should operate with internal clock whenever possible. Why? Because it creates less jitter. I have heard people that like tube sound, warm sound, compressed sound… name it. But I do not know anyone that likes the sound of more jitter!

Let us assume for a moment that your external clock is as good as a good crystal based clock. Say you provide an output with ZERO JITTER! That clock needs to travel via a cable (grounding and interference issues), into a semiconductor receiver (more jitter and also supply sensitivity) to a PLL circuit (that circuit adds A LOT OF JITTER relative to a crystal). A marketing guy can talk about “Apogee's advanced science in this field” but this forum endeavors to elevate discussions way above those considerations.

I know you sold a lot of clocks, but this is a technical forum. “At the end of the day” the issue of jitter is about how many pico seconds of jitter are AT THE AD CHIP. To be -clear – the crystal technology beats the jitter of both internal clock modules and stand alone clocks based on DDS technology. Internal clocks modules based on DDS are more jittery. Advocating the use of any external clock in cases where one can use internal crystal is WRONG.

Again, this is a technical forum. Please read the rules under Announcement (first message).

Again: JITTER IS BAD FOR CONVERSION!
Again: the best DDS technology is much more jittery than reasonable crystal.
It is not about who YOU call a great engineer, or an old relic.
It is not about saying “it is better” over and over.
You are constantly changing focus from the real issue at hand.

The issue at hand is:
How many pico seconds jitter is AT THE AD LOCATION. The less, the better. Period!

That is the difference between a “low jitter clock” and a “low jitter crock”.

Regards
Dan Lavry
Title: Re: Proper word clock implementation
Post by: Roland Storch on November 03, 2004, 06:50:07 pm
This question has already been answered (I already asked almost the same qustion).

Take the internal clock of your AD converter (lower jitter than external clcocking). For the clocking of the other digital units take the WC out of that internal clocked AD converter and connect the other units directly via BNC cables and T-connectors and the right dermination.

A WC distribution amp is not necessary and even worse (regarding jitter) than just taking BNC cables an T-connectors.
Title: Re: Proper word clock implementation
Post by: bobkatz on November 03, 2004, 06:55:25 pm
Albert wrote on Wed, 03 November 2004 17:20

Okay, I hate to interrupt this heady and intense discussion with the ridiculous, but I have a simplistic nuts and bolts type question.




It's not ridiculous, Albert, you're cool! We need lots of practical questions and examples to help illustrate the theory versus the practical.

In order to answer your question, let me ask: did you catch the part of this thread where Dan Lavry suggests making a chain with a series of T-Connectors, to avoid the jitter which can be added in a distribution amplifier?

How many actual CONVERTERS and channels do you want to sync up? Which models are they and what are their syncing possibilities?

If any of the things that you want to sync up ARE NOT CONVERTERS, then you can just use plain ol' AES/EBU sync or a WC distribution amplifier to feed those. Tell us more and maybe we can come up with a plan that answers your question!

-----------

Back to the "Big Ben" discussion, maybe we should split the thread...

On listening tests for jitter. We have to separate the hype from the facts. It is EXTREMELY difficult to do an A/B comparison between clocks by the way, scientifically and blindly. The reason is that EVERY clock circuit takes time to stabilize; and many converters mute during the stabilization period or switch period. This very gap in sound (can be as long a 5-10 seconds or longer in some models) totally destroys the ear's memory for sound. If you have a preconceived notion of what effect an external clock will have on the sound, then it can be very difficult to prove whether we are not just "dreaming it all up". The ONLY fair listening test for D/A converters on external versus internal clocking would be to have two identical models of the same converter and be able to switch between them. Fair enough? Max is invited to supervise my tests, but I think he'll find that I am a very fair and open-minded listener. Even though I am biased against DSD, for example, I've found some surprising and puzzling sonic improvements with upsampling to DSD in blind tests.

My listening record demonstrates that several times in the past I have discovered and verified OPPOSITE conclusions to my own biases. In the case of the famous "filter tests" for example.

So I think that I can conduct fair and open listening tests, but Max is welcome to supervise, as I said.

For the DACs, he should bring TWO of each converter to the test. Any other listening test is bogus or suspect. How many of you out there have performed clock comparisons with identical converters? I admit it, I haven't! So all my tests are suspect, at least on the D/A side.

Clock listening comparisons can be done on the A/D side with a single converter since the results are recorded into a file. And the two files then can be auditioned into a single D/A.

We shall see, if Apogee sends me a Big Ben and two converters. My assistant's ex college roommate is now working for Apogee. I don't know Max personally and I feel very bad for being so confrontational. But this is science, and he's been responding on the "feeling good" level with no measurements to support the voodoo. No measurements I've seen, at least. How many objective listening tests? How many measurements of the jitter?

The other poster who claimed that his TC 6000 improved with Big Ben. I had the opposite experience; the System 6000 was MARGINALLY superior on internal clock, exactly as its measurements imply. But I did not apply the Big Ben. I used another stable clock; I'm willing to listen (and measure) again with the Big Ben, voodoo not withstanding. Ironically, I'm no longer using the System 6000 converters... I've "moved up".

I am trying to remain on the scientific level in this thread, despite that I am a very critical listener by profession and by instinct. On the scientific and technical level, my points, Dan's, and several others' are very hard to refute. Dan is even aware of the technology behind the Big Ben; depending on the nature of the jitter which this technology produces, in combination with the PLL of the receiving circuit, it could produce jitter at the converter which could produce a "vaguer image" rather than a "wider, stable image". This can fool a lot of ears...it can easily be a case where "more jitter" sounds better than less.

So we have to learn how to be objective in our jitter listening tests as well! Think of it like coincident pair miking versus spaced miking. Is "more pleasant to the ear" more correct? So to really know whether the converter is lying or telling the truth, we need an analog source to compare it to. Or, at the least, we have to know a lot about the miking which was used in the recording which we are evaluating; if a coincident pair starts to sound like a spaced pair with the Big Ben, then I would call it "jitter-induced hypnosis"  Smile. The space has to increase while the apparent stability and size of the center image has to remain rock steady, in order to qualify for a genuine sonic improvement.

Regarding user "bias"---it is true... Dan has product to sell, but so does Max, and we must recognize that money drives the world, even though I think Dan has done an excellent job nearly completely separating the theory from the "brand practice". If Max is not going to provide the measurements or the theory, then he has to post more than just listener endorsements by name. He has to post some written critiques, fine descriptions of the sonic effects, and descriptions of the objective nature of the test methods so we can effectively criticise and evaluate the results.
Title: Re: Proper word clock implementation
Post by: Albert on November 03, 2004, 07:02:28 pm
Dan, thank you, I will read that time delay thread.

Roland Storch wrote on Wed, 03 November 2004 23:50

This question has already been answered (I already asked almost the same qustion).

Take the internal clock of your AD converter (lower jitter than external clcocking). For the clocking of the other digital units take the WC out of that internal clocked AD converter and connect the other units directly via BNC cables and T-connectors and the right dermination.

A WC distribution amp is not necessary and even worse (regarding jitter) than just taking BNC cables an T-connectors.


The issue in my case is that I have over 16 devices that need WC, and for some reason t-bar chaining is not working properly. All the terminations are off as they should be, short BNC cables, etc., but it's just not happening. The only way I've been able to get it to go is with a few WC streams coming out of my AardSync II and feeding different chains. Sounds okay, but I'm just trying to get optimum performance out of the gear I have. Hopefully someday I'll be able to get a single chain to function. Still working on it.
Title: Re: Proper word clock implementation
Post by: danlavry on November 03, 2004, 08:09:13 pm
Bob Said:

Regarding user "bias"---it is true... Dan has product to sell, but so does Max, and we must recognize that money drives the world, even though I think Dan has done an excellent job nearly completely separating the theory from the "brand practice". If Max is not going to provide the measurements or the theory, then he has to post more than just listener endorsements by name. He has to post some written critiques, fine descriptions of the sonic effects, and descriptions of the objective nature of the test methods so we can effectively criticise and evaluate the results.

Bob,
I think your initial reaction was more to the point. That is when you said: “I am pissed because this thread is being taken over by salesmanship and ignorance of the basic design constraints of converters.”

Yes, Dan has product to sell .

But why do you think I do not sell a stand alone clock? After all, look at the Msync board of my LavryBlue converter. It is ready to package clock circuit with 2 very expensive fundamental frequency circuits, and 2 even more expensive pullable crystals. plus a wide lock circuit, plus a word clock/AES in and a BNC out, and a whole front panel with LED and switches to control it all. This is low jitter technology! You buy an AD and you have it included!  

Everyone is selling a clock. Apogee, Symetrix, Advark, Prism… I am the “fool” telling people to use internal clock when possible instead of selling one myself. I could easily make one. The electronics is done, all I need is a "box". I am the one to point out that for the cases requiring external clock, the jitter is going to be mostly due to the transmission and internal PLL. So it is not entirely “fair to me” to be put into the same category as stated: “Dan has product to sell, but so does Max, and we must recognize that money drives the world”.
Of course I do appreciate your comment:  “I think Dan has done an excellent job nearly completely separating the theory from the "brand practice".

Back to jitter and clocks. I have been familiar with jitter problems long before digital audio (such as in medical, instrumentation, telecom and more AD gear). Jitter destroys the signal tracking. Random jitter is bad news. Non random jitter (coupling from the signal itself or from a non random source) is much worse, and the energy concentrations occur at sums and differences and aliasing of sums and differences – as un musical as converter anti aliasing problems. Once we agree jitter is bad, and less jitter is good, everything falls into place – go for as low jitter as you can.

I understand people liking tube sound, more even harmonics, re-verb, warm sound, harsh or edgy sound… name it. Some people like chocolate ice cream, others like cheese cake or apple pie.

But one can not extend the concept of like or dislike it to jitter. To use the analogy above, liking more jitter is likening poison. It is a bad thing. It does not call for a listening test.
What is next, after listening to jitter? Should we add noise to the converter DC voltage reference and listen to it? How about modulation of the coupling capacitors with a varactor? Or injecting noise into the PC board ground plane? Chances are someone will buy it. Don’t you think?    

What you said earlier was right. A design helped by external clock is bad news (real bad internal clock). Not even subtly so. Why get sucked in to listening to jitter? If someone tells you that some external device jitter is going to cancel the internal jitter, don’t buy it.
It can not be true!

Let me remind you all that this forum is not about listening tests. Please take such topics of conversation to another site.

BR
Dan Lavry
Title: Re: Proper word clock implementation
Post by: Roland Storch on November 04, 2004, 06:15:39 am
Albert wrote on Thu, 04 November 2004 00:02

Dan, thank you, I will read that time delay thread.

Roland Storch wrote on Wed, 03 November 2004 23:50

This question has already been answered (I already asked almost the same qustion).

Take the internal clock of your AD converter (lower jitter than external clcocking). For the clocking of the other digital units take the WC out of that internal clocked AD converter and connect the other units directly via BNC cables and T-connectors and the right dermination.

A WC distribution amp is not necessary and even worse (regarding jitter) than just taking BNC cables an T-connectors.


The issue in my case is that I have over 16 devices that need WC, and for some reason t-bar chaining is not working properly. All the terminations are off as they should be, short BNC cables, etc., but it's just not happening. The only way I've been able to get it to go is with a few WC streams coming out of my AardSync II and feeding different chains. Sounds okay, but I'm just trying to get optimum performance out of the gear I have. Hopefully someday I'll be able to get a single chain to function. Still working on it.


In this case I would take a master clock with over 16 outputs (for over 16 devices if all devices are used at the same time.
Title: Re: Proper word clock implementation
Post by: Max on November 04, 2004, 12:55:48 pm
I had written a very long and drawn out response to all of this, unfortunately it somehow became lost in the ether (urrrgghh, what a frustrating waste of time). Now I guess I'll sum up:

Dan, to your question as to whether we improved the specs on the DDS chip:

Quote:

I hope you are not trying to suggest in your statement how and why of Apogee's advanced science in this field is that there are things we are doing that we do not want to advertise to our competitors that you have figured how to greatly improve on Analog Devices IC specifications. The finished product is not going to be as good as the stand alone device specifications.


The answer is yes, we have figured out how to do this. And while we are not using the AD9850, we have improved on the original DDS spec in similar fashion to how we (and you for that matter) have been able to improve A/D conversion beyond just what the A/D chip is capable of. Again, I cannot comment specifically on how we did this, as that would have a direct impact on our business.

As for not building an external clock box, we felt similarly for many years, until we developed technology that would make a difference.

Bob, I will try and get in touch with you offline to make arrangements in the near future to do this listening, if you are serious.

With regards to being off topic, keep in mind that the rhetoric and disparaging remarks that were being thrown around specific to Apogee is what prompted me to respond in the first place. I would not have felt compelled to respond had you guys kept the comments above board to begin with. That being said, everything stated about Big Ben is valid and pertinent to this discussion and the readers have a right to know the truth about this technology.

Finally, Dan, I understand that there is a history with you and Apogee, but that was over a decade ago. Everything has changed many times since then. For example, Soft Limit is in its 4th generation now and has been greatly improved since its inception. UV22 was completely rewritten by Jerry Goodwin a long time ago and is now UV22HR. While we respect the contributions you have made to the audio industry, we do not appreciate you taking credit for work done by Apogee since your departure long ago. Taking credit for that which is not yours or even suggesting as much is unfair to Apogee's hard working designers and everyone else who is putting forth an honest effort to make the best products we can.

In the interest of getting back to work and putting this to rest, I will opt out of the rest of the discussion. I hope going forward you can strive to give the readers and honest assessment of what you believe to be relevant, without making personal attacks and throwing around needless rhetoric.
Title: Re: Proper word clock implementation
Post by: danlavry on November 04, 2004, 07:29:10 pm
"Dan, to your question as to whether we improved the specs on the DDS chip:

Quote (what Dan said earlier):
"I hope you are not trying to suggest in your statement how and why of Apogee's advanced science in this field is that there are things we are doing that we do not want to advertise to our competitors that you have figured how to greatly improve on Analog Devices IC specifications. The finished product is not going to be as good as the stand alone device specifications."

The answer is yes, we have figured out how to do this. And while we are not using the AD9850, we have improved on the original DDS spec in similar fashion to how we (and you for that matter) have been able to improve A/D conversion beyond just what the A/D chip is capable of. Again, I cannot comment specifically on how we did this, as that would have a direct impact on our business."


I do not believe the impossible. You can not do the impossible, and just because you claim that you can does not make it so. Improving a DDS chip beyond it’s capability is marketing hype. Those chips are tested and specified by excellent teams at the IC factory, with the best gear, they are powered by laboratory supplies in extremely clean environments. Low jitter is about the physics of a device, not DSP and filters, and a fixed properly deigned Crystal oscillator is about as good as it gets, and can not be improved with DSP or filtering of DDS.

When push come to shove, what you are saying cannot stand up to any reasonable theoretical scrutiny and will end up as no more than unsubstantiated listening tests or subjective sonic opinions.

THE FOCUS IS:  HOW MANY PICO SECONDS OF JITTER IS AT THE AD CONVERTER. NOTHING ELSE.

AND

YOUR SOULUTION IS NOT IN PAR WITH A REASONABLE CRYSTAL BASED SOLUTION.  

First you tried “brushing it away” with comments about “theorists” of the last century. Instead of answering my comments, you are now “piling it up” with claims of super natural ability. Get some perspective man. You seem to believe your own comments. Use your head. Do you believe in some “Apogee super technology” capable of “opening an IC” and mucking with the insides? DDS is MORE JITTER than fixed fundamental CRYSTAL circuit!  Also external clocking is MORE JITTER than using internal crystal.

NO AMOUNT OF ARGUMENTS WILL CHANGE THAT PHYSICAL REALITY!

”As for not building an external clock box, we felt similarly for many years, until we developed technology that would make a difference.”

Marketing marketing and marketing. What does “make a difference” mean? Better jitter performance than internal clock? NO SO. Worse jitter? YES!

Look, Max. You are asking for it. I just printed the stuff about Big Ben from your site. Yes, I see the Tech Award 2004. That is marketing too, and another topic, not a technical  one. But at the middle of the page, under “A Cure for the Jitters” it starts by saying a correct fact: "the better clock performance, the less flexibility it has”. This is true! That is why a dedicated crystal, In fact, one crystal per frequency (or divide by n) is the best solution.

But then your literature says: “Apogee has developed ………using the most advanced DDS…along with DSP based digital filtering…level of performance unequaled in the world of clocking.”  

That is SNAKE OIL. Performance unequaled in the world of clocking is a WRONG!
Unless I do not understand the word PERFORMANCE. For me it means LESS JITTER, certainly under your heading “A Cure for the Jitters”.

You guys screwed up! You can not cover B.S. with more B.S.

"With regards to being off topic, keep in mind that the rhetoric and disparaging remarks that were being thrown around specific to Apogee is what prompted me to respond in the first place."

No such comments on this thread. My comments were not specific to anyone, nor did I get into the DDS vs. Crystal issue. I just pointed out that one should use internal clock when possible. I guess it cuts into yours and other’s sales goals, but I am for good audio and for the truth. Your entrance onto this site brought the name of Apogee and a whole slew of nonsense to this technical site.  I never mentioned any clock manufacturer’s names. I was dealing with the concept of clocking. Since you brought your marketing dogma here I am simply telling you what I think of it.

“I would not have felt compelled to respond had you guys kept the comments above board to begin with. That being said, everything stated about Big Ben is valid and pertinent to this discussion and the readers have a right to know the truth about this technology.”

I understand that as a salesman you felt compelled, but this is a site for engineers. We welcome your engineer to take part in a TECHNICAL discussions and you can feel free to return to your sales job.

I agree that the readers have the right to know the truth about this technology. It is sold as a device to improve jitter, and it is not capable to compete with a reasonable fixed internal crystal solution.

“Finally, Dan, I understand that there is a history with you and Apogee, but that was over a decade ago.”“Everything has changed many times since then. For example, Soft Limit is in its 4th generation now and has been greatly improved since its inception. UV22 was completely rewritten by Jerry Goodwin a long time ago and is now UV22HR.”

But you kept the concepts, and those are the 2 real “theoretical ideas” used by Apogee, both developed before I did any work at Apogee. I have greatly improved upon those old design concepts long ago. It appears you are recycling them but then I do not know this to be a fact since I am far too busy to be poking around in your current circuitry.

"While we respect the contributions you have made to the audio industry, we do not appreciate you taking credit for work done by Apogee since your departure long ago. Taking credit for that which is not yours or even suggesting as much is unfair to Apogee's hard working designers and everyone else who is putting forth an honest effort to make the best products we can."

Does it sound like I am trying to take credit for current implementation?  I am telling you that your clock is incorrectly implemented only because you are bringing it to this forum and are confronting me with nonsense.

I did design Apogee’s first AD and DA and did contribute concepts such as the soft saturation, and the UV which was done in my house (with Jerry Goodwin) – 18 or so months o 7 days 12-14 hours a day. But that was a long time ago. However, the value is in the concept not only the specific implementation of a concept from an inventor’s perspective.

I have continued to design audio products for my companies and there are some real contributions I am proud of.  

"In the interest of getting back to work and putting this to rest, I will opt out of the rest of the discussion. I hope going forward you can strive to give the readers and honest assessment of what you believe to be relevant, without making personal attacks and throwing around needless rhetoric."

You are the one that was “throwing around needless rhetoric” at the middle of a technical thread. I find you statement regarding  “honest assessment of what you believe to be relevant” interesting. I guess what a non technical guy “will believe to be relevant” has a lot to do with who they choose to listen to, so the responsibility for spreading misinformation goes back elsewhere. No self respecting EE will take an issue with my comments about DDS being more jitter than crystal, and Internal Crystal being better then external lock. While everyone is “going back to work", I am standing here in the open telling the truth, and the focus is:

HOW MANY PICO SECOND JITTER IS AT THE AD CONVERTER LOCATION.

The answer is:

1.   LESS JITTER WHEN USING A CRYSTAL THAN DDS
2.   LESS JITTER WITH INTERNAL CRYSTAL THAN EXTERNAL CLOCKING

Dan Lavry
 

Title: Re: Proper word clock implementation
Post by: chap on November 04, 2004, 09:06:17 pm
well, I for one, am happy to see some 'simmer down now' taking place.  I'm hoping we get back to knowledge and not 'impart' 'infer'.
peace,
chap
Title: Re: Proper word clock implementation
Post by: Bill B on November 04, 2004, 09:18:39 pm
Wow. Maybe not the prettiest of threads, but these things need to be said to keep the focus on the facts. It can be so hard for little guys like me to decipher the hype of some(most)  manufacturers (without the knowledge and experience that many of you have) to try to spend our hard-earned dollars wisely.
Thanks Dan, BK and all for a very enlightening thread.


Title: Re: Proper word clock implementation
Post by: Lucas van der Mee on November 05, 2004, 03:53:49 pm
Dan,

You seem to have a problem with sales guys, even the ones with knowledge and integrity. Max never states anything without backing it up with engineering.
So here is an engineer, who speaks your language and tells you, you are wrong. Sure, jitter is bad, we never said anything different. Just like you, I was convinced that internal clocking was the best thing to do. But here is what I found:

I have designed several PLL’s most of them hybrid and as you very well know there is not a perfect world. The VCO solution is more jittery but has a wider capture range, the VCXO solution is almost perfect but has a very narrow capture range. You see, I worked in recording studios for two decades and found myself 9 out of 10 switching to wide capture range mode, because the situation forced me. So that is why you never find a VCXO on an Apogee device, it is simply a pretty useless PLL, not to mention the lack of being able to varispeed. BUT we need PLL’s and PLL’s have diametrically opposing requirements. They need to be fast to follow sudden changes in the clock source, they need to be slow to have very low jitter. So I started to think of a complete new solution, one that can marry both requirements and more. That is where the DDS comes in. You can use a DDS in a PLL in many different ways, just read the textbooks and the app. Notes. But just like you, I did not see the performance required. Until I found a DDS which I really liked, one that I could use in a non-standard solution and I was able to improve on, in its specs. Let me give you some hints, its filtering, layout…all that analog crap again, we have to deal with and…of course our implementation. And that is as far as I am willing to go in telling how it works. We even decided not to get a patent on it, because that would be a giveaway as well. So yes we are very proud of our work and can imagine you don’t like to see someone succeed in something you thought was impossible.
During the development of or C777 PLL we also did some listening tests and found to our surprise that a lot of other converters sounded better (meaning more accurate, closer to the source) when clocking to the C777, against all theory! This was one of those moments where the engineer in me was dumbfounded, it did not make sense at all. Yet the tests were conclusive and repeatable. So we researched that and developed a clock that not only performs great as a PLL but as a masterclock as well. Again, I would hurt my own research if I tell you why that is, we did find something and as usual it is not that complex. But why should I give it away? It is my product and it allows me to make a decent living, sorry. To give you an idea, we have the same C777 in the AD16X. There is no difference in performance whether you have the unit on internal or external clock, if you have been to the Rhode & Schwartz booth on the AES show in San Francisco you could have seen the performance yourself, since we loaned the unit to demo their new test-equipment: -111 dB THD+N at -1 dBfs. Now that is what I call “unsurpassed excellence” (to quote your website), for a 16 channel AD converter retailing for $3500.- It is up to you to beat that…At the same booth you were also able to see all kinds of other measurements, including an FFT and those who have been there, can acknowledge the “crystal” performance we get out of our C777.

Bottom line is Dan, technology and measurements are only one part of developing equipment. Of course we all want absolute values, facts. But experience has taught me that objectivity always comes after the subjective: why does something sound good? Instead of: this will make it sound good. The list of examples when this happened is infinite. So I suggest, and I am told you have good ears, to listen to a Big Ben clocking a converter and then come again…you will be surprised.

Now finally, about Soft Saturate and UV22, when you see our current line Soft Saturate does not exist anymore, it has been replaced by the much more subtle and better sounding Soft Limit circuitry, but that is besides the point I want to make. These are still part of our current converters, of course, why wouldn’t it be? We also continued to design our products around a converterchip! Are you going to claim that as one of your achievements as well?

Regards,
Lucas van der Mee
Sr. Design Engineer
Apogee Electronics
Title: Re: Proper word clock implementation
Post by: danlavry on November 05, 2004, 08:18:56 pm
Dan,

You seem to have a problem with sales guys, even the ones with knowledge and integrity. Max never states anything without backing it up with engineering.


Those are confrontational comments, not technical ones.

So here is an engineer, who speaks your language and tells you, you are wrong. Sure, jitter is bad, we never said anything different. Just like you, I was convinced that internal clocking was the best thing to do. But here is what I found:

OK, lets hear why internal clock is not the best thing to use:

I have designed several PLL’s most of them hybrid and as you very well know there is not a perfect world. The VCO solution is more jittery but has a wider capture range, the VCXO solution is almost perfect but has a very narrow capture range.

Good you admit that the VCXO is almost perfect. The fixed Crystal is better than VCXO. The difference is that fixed crystal is optimized fro fixed frequency, the VCXO must operate over a tiny range such as +/-100 parts per million. So again, the fixed crystal is better, and that is what one uses (or should) for internal clock.  

You see, I worked in recording studios for two decades and found myself 9 out of 10 switching to wide capture range mode, because the situation forced me. So that is why you never find a VCXO on an Apogee device, it is simply a pretty useless PLL, not to mention the lack of being able to varispeed.

My gear uses VCXO for external lock, and we do not have customers loosing lock, not 9 out of 10, not 1 out of 1000.

BUT we need PLL’s and PLL’s have diametrically opposing requirements.

You are getting off the track. The subjects at hand are:
1. INTERNAL CLOCK vs PLL
2. THE BEST TECHNOLOGY FOR EXTERNAL CLOCK SOURCE

So you are redefining the “starting point” as a need for PLL. I say we need a fixed crystal with no PLL for best jitter. You are getting off the subject.

They need to be fast to follow sudden changes in the clock source, they need to be slow to have very low jitter…. Until I found a DDS which I really liked, one that I could use in a non-standard solution and I was able to improve on, in its specs. Let me give you some hints, its filtering, layout…all that analog crap again…

Nuts. Let me give you some hint: Before you talk about filtering, remember that a fixed crystal is a far superior filter to anything you are talking about. It is about DEVICE PHYSICS, where the alignment of the quartz molecules is such that the device yields the narrowest bandwidth possible, thus the bandwidth of operation is tiny, the Q factor is HUGE. Putting a device like that in a loop is THE BEST way to overcome power supply variations, and what you call “analog crap’. By the way, I love analog and it is not crap to me.  

A VCXO for external locking is almost perfect (YOU said it yourself). A fixed crystal (for internal applications) is so much better.

“…of course our implementation. And that is as far as I am willing to go in telling how it works. We even decided not to get a patent on it, because that would be a giveaway as well. So yes we are very proud of our work and can imagine you don’t like to see someone succeed in something you thought was impossible.”

You totally side stepped the issues of:  
1. INTERNAL CLOCK vs PLL
2. THE BEST TECHNOLOGY FOR EXTERNAL CLOCK SOURCE

During the development of or C777 PLL we also did some listening tests and found to our surprise that a lot of other converters sounded better (meaning more accurate, closer to the source) when clocking to the C777, against all theory! This was one of those moments where the engineer in me was dumbfounded, it did not make sense at all.

Ah, so we are back to marketing. I agree that the engineer in you is dumbfounded and what you say still does not make sense, nor is it to the point.  

“Yet the tests were conclusive and repeatable. So we researched that and developed a clock that not only performs great as a PLL but as a master clock as well.”

More marketing.

“Again, I would hurt my own research if I tell you why that is, we did find something and as usual it is not that complex. But why should I give it away? It is my product and it allows me to make a decent living, sorry. To give you an idea, we have the same C777 in the AD16X. There is no difference in performance whether you have the unit on internal or external clock, if you have been to the Rhode & Schwartz booth on the AES show in San Francisco you could have seen the performance yourself, since we loaned the unit to demo their new test-equipment: -111 dB THD+N at -1 dBfs. Now that is what I call “unsurpassed excellence” (to quote your website), for a 16 channel AD converter retailing for $3500.- It is up to you to beat that…At the same booth you were also able to see all kinds of other measurements, including an FFT and those who have been there, can acknowledge the “crystal” performance we get out of our C777.”

Maybe I should start talking about the music I play with my Klezmer band. A bunch of words petting yourself on the back, but NOTHING about the subject at hand!!! It certainly seems like you are running away from the technical. I know why you are.

Bottom line is Dan, technology and measurements are only one part of developing equipment. Of course we all want absolute values, facts. But experience has taught me that objectivity always comes after the subjective: why does something sound good?
Instead of: this will make it sound good. The list of examples when this happened is infinite. So I suggest, and I am told you have good ears, to listen to a Big Ben clocking a converter and then come again…you will be surprised.


Ah, the famous lecture about sounding good. I thought we are taking about JITTER. You literature claims that you C777 is “THE CURE FOR THE JITTERS”!

Now finally, about Soft Saturate and UV22, when you see our current line Soft Saturate does not exist anymore, it has been replaced by the much more subtle and better sounding Soft Limit circuitry, but that is besides the point I want to make. These are still part of our current converters, of course, why wouldn’t it be? We also continued to design our products around a converter chip! Are you going to claim that as one of your achievements as well?
Regards,
Lucas van der Mee
Sr. Design Engineer
Apogee Electronics


I did not invent the resistor, nor did I invent the IC. I did co-invent the UV22. Trying to attack me and lower me is a reflection on you.

Lucas,
I did not go after Apogee. I was talking about clocking. Max came in and now you. This is a technical forum.

Lets get back to technical. If you can not do it, I can:

THE FIRST ISSUE IS ABOUT USING INTERNAL CLOCKS VS EXTERNAL.

YOU ADMITTED THAT JITTER IS BAD. Then, you ramble on saying nothing short of a few words about filters or implementation of a non fixed source, then name throwing (Rohde and Schwartz), then how things sound.

Meanwhile, you guys are telling people that using your clock will be a “CURE FOR ALL JITTERS”. Technically speaking, even if your external clock had no jitter, THERE IS NO WAY THAT IT COULD REMOVE, CANCEL, REDUCE the noise picked up at the other end of the interconnecting cable, the receiver, the internal PLL. Your clock box DOES NOT KNOW WHAT IS AHEAD, and therefore CAN NOT COMPENSATE FOR IT!

THE SECOND ISSUE IS COMPARING YOUR CLOCK JITTER TO A GOOD CRYSTAL CIRCUIT.

Your DDS is, at best, based on a crystal that operates on an overtone, thus the “engine” of your design is already inferior. Go to analog Devices site and read about the impact of the reference clock on the DDS. It can not be corrected for, by DSP or any filter. That clock is the ONLY referance. If it moves, there is nothing else to know that there is a need for correction! The ONLY reference is your reference clock, and when it jitters, everything follows!!! Also, think about all that “analog crap” as you call it, where every device in series at the DDS demands the cleanest supply. Think about the fact that the more devices in series, the more jitter accumulates. That a Crystal oscillator can operate with one or two transistors in series and everything is designed to be FIXED, while your DDS has all the extra circuitry to deal with (for additional features) thus more buildup of intrinsic device jitter. Realize that the best way to filter the jitter noise is with the best filter – a crystal. Realize that a supper high Q narrow bandwidth device – a crystal, when operated with the proper feedback gain, can be optimized to yield a pure sine wave at one frequency, in contrast to the square wave of the digital devices.

Did you know any of it, or are you out of your league? So far, you did not come up with a single technical point to refute what I said. I am surprise you showed up here saying what you said, letting yourself be so exposed, almost all sales and no engineering. I told Max to send the engineer.

It would not be uncommon for a manufacturer to sell something with poor specs saying “it sounds good”. But it is an altogether different to call it “A cure for the jitters”. I would not argue with someone that decided they like the sound of more jitter, more distortions or anything that is a matter of taste. I do not hear you admitting that the internal clock is less jitter. That using any external clock to drive an internal PLL yields more jitter than internal fixed crystal. I do not hear you admit that a fixed crystal is less jitter than your DDS. You stated that a VCXO (pull able crystal) is almost perfect. Are you saying your DDS is better than almost perfect?

THE BOTTOM LINE:

QUESTION: WHICH SOLUTION PROVIDES LESS JITTER AT THE AD LOCATION?

1.   A REASONABLE FIXED CRYSTAL
2.   BIG BEN DRIVING A CLOCK INTO A CHASSIS EQUIPED WITH A PLL VIA A 10 FOOT CABLE?

QUESTION: CAN YOU EVER CANCEL OR REDUCE A RANDOM NOISE AT ONE END OF A CABLE WITHOUT KNOWING WHAT IT IS AT THE OTHER END?
(can you cancel, remove or reduce the receiver and PLL noise, some of it unknown, some totally random, by driving it with a clean low jitter clock? Driving it with any clock?)

That is in fact what you are selling!

I remind you that before you or your sales guy entered into this forum, no one was saying names, or directly criticizing any company. If you read the posts, for every question we have tried very hard to remain technical and elicit technical responses. We are not about subjective listening here but you can certainly find chat rooms that are. Perhaps you believe what you say as does your sales director. The facts remain that putting aside your personalized comments you have not edified or elevated the technical level.  I speak not only for myself, but for others who are reading these posts, we want very much to have intelligent conceptual inputs in the spirit of positive interaction.  I gain no joy at having to put you down.

You are missing much by not understanding analog design as are so many other young designers.  

Dan Lavry
Title: Re: Proper word clock implementation
Post by: bobkatz on November 05, 2004, 08:49:44 pm
Lucas. I would like to perform some objective and subjective (blind) listening tests on the Big Ben. What I like about your post is the "against all theory" part. Because no one can refute that statement except with measurements and tests. I'd like to see the theory broken!

However (and this is important):

Will the measurements show the improvement as well as the listening tests? You describe some Rohde and Schwartz measurements. Does this include FFTs of a high frequency -1 dBFS test tone to look at the artifacts?

If I discover with simple measurements that the jitter artifacts of selected converters measure worse with the Big Ben, will you understand that I need not proceed with subsequent listening tests? Because that would refute your statement of "against all theory"?  

Are you game? Send me a Big Ben for testing. While I may seem stubborn when it comes to the technical, I am stubbornly open-minded when it comes to listening. After all, I worked for an audiophile label for more than 10 years! And you know audiophiles. I WILL separate my opinion from the facts, and if we do listening tests they will include other listeners, and will be BLIND.

May the best clock win!

Please Send the Big Ben to:

Digital Domain
931 NSR 434 Suite 1201-168
Altamonte Springs, FL 32714

It will either be purchased, or returned, in like-new condition. Promise. Furthermore, I will not write or publish anything until you have seen what I am to write, and have been given an opportunity to respond.

BK
Title: Re: Proper word clock implementation
Post by: Curve Dominant on November 07, 2004, 05:11:39 am
Quote:

posted by Lucas van der Mee:
technology and measurements are only one part of developing equipment. Of course we all want absolute values, facts. But experience has taught me that objectivity always comes after the subjective: why does something sound good? Instead of: this will make it sound good.


Lucas,

That statement makes it appear as if you should get out of the engineering business, and be an artist instead.

Let me explain further what I mean by that:

Artists depend upon engineers to do just that: engineer.

Once you engineering blokes start to wax philosophical, you lose our confidence. Please. We have enough of the fuzzy stuff inherent in OUR field as artists. We don't look to you for the fuzzy stuff. We look to you engineering folks for solid ground.

We need solid tools to accomplish our sometimes abstract goals.

The realm of  "this will make it sound good" is the realm of the artist's subjectivity, NOT the engineer's. The engineer's mission is to provide tools and services which will establish an even playing field for ALL artists to accomplish their goals.

Once the engineer imposes his predgudice upon the medium, he immediately skews the result in a most unproductive fashion, as you have demonstrated.

I would suggest that you and Max Gutnick re-examine your role in our industry, and then re-position your company's mission in a pro-active fashion.
Title: Re: Proper word clock implementation
Post by: bobkatz on November 07, 2004, 10:25:53 am
Eric Vincent wrote on Sun, 07 November 2004 05:11

Quote:

posted by Lucas van der Mee:
technology and measurements are only one part of developing equipment. Of course we all want absolute values, facts. But experience has taught me that objectivity always comes after the subjective: why does something sound good? Instead of: this will make it sound good.


Lucas,

That statement makes it appear as if you should get out of the engineering business, and be an artist instead.




Well, I disagree. Lucas's primary contention is absolutely important. It IS the role of engineers and scientists to find out WHY something sounds good. Psychoacoustics is a science; just not as well-understood a science as the design-engineering science that we're discussing in this board.

If there is an apparent discrepancy between any or some of the measurements, it IS OUR job as engineers and scientists AND listeners to try to resolve that discrepancy. This will only improve the science and the engineering, and help to make better equipment in the long run.

For example, we need to subjectively/objectively test Apogee's contention that an external clock will improve the sound of "any converter", regardless of the scientific theory. Regardless that EVERY other respectable converter designer would disagree.

And why? Because we do not know everything about the subjective issues here. I personally disagree with their contention, not only because it disagrees with the science, but also because my past tests and experiences contradict their contentions. But I am  willing to test the box, as objectively as I can, both listening and measurements---because there we have to leave room for doubt, or we would not be true scientists.

Let's summarize: Apogee claims that the Big Ben improves the sound (and lowers the jitter, it's in their ads) of EVERY converter (except ones which contain the Big Ben ). Would it be sufficient to disprove their claim if I find a converter which does not improve with the Big Ben? Would it be sufficient to disprove their claim if I find a converter which sounds audibly worse when fed from the Big Ben? What about the measurements? What if the measurements and double-blind listening tests disagree? What then?

Responsible engineers should be prepared to deal with all of the above possibilities.

In the past, when investigating inferior converters that are susceptible to jitter, I discovered that measuring the absolute RMS (or peak to peak) value of the jitter at the converter's clock pin----DID NOT CORRELATE well with the listening experience. The spectrum of the jitter itself had an influence on the sound of the converter! In fact, to some degree, you could even "tune" the sound of a converter by the spectrum of the jitter, and whether you made it random or correlated to some degree.

But I did find that ALL OF THIS nonsense became a moot point when we finally learned how to make converters with sufficiently low measurable levels of jitter such that the jitter spectrum was insignificant. The point being that an inferior converter could conceivably sound better with the Apogee clock. (Ho hum....)

My point being that there are explanations to everything... we just have to find them! Science and engineering includes the science of psychoacoustics, after all! And we're going to have to live with fuzzy-logic for a long time, guys... until we have a much firmer grasp on the psychoacoustics.

My biases not withstanding, I promise to do a thorough examination of the Big Ben. This could take weeks to months. Every weekday I have is booked, and I've been working weekends (mastering or doing studio maintenance/updates) more often than you can believe.
Title: Re: Proper word clock implementation
Post by: maarvold on November 07, 2004, 03:03:05 pm
Mr. Lavry & Mr. Katz (Dan & Bob didn't seem respectful enough): how do we know what to listen for?  I bought Aardsync in the Digidesign 888/24 days and used it to provide a discreet Superclock feed to each 888/24 (and 1 to the USD) and I seem to remember thinking that:
1. Recordings I made of acoustic piano sounded more 'real' when played back with Aardsync as the master clock (rather than USD in the 'approved' Digi daisy-chain arrangement)
2. Apparent depth and width of reverb seemed more fully 'fleshed out' at the rear of the soundstage with Aardsync
3. The apparent frequency response in the 200-500 Hz range seemed to change by a dB or more with Aardsync instead of USD.  I seem to remember it was 'leaner-sounding' with USD

Now I own Pro Tools HD and monitor the AES (internal) out via a TacT RCS 2.0 [digital] room correction device.  Since reading this thread (in its earlier phases), I have switched, from Aardsync, back to using "Internal" of the 192 I monitor with as my master clock.  I do hear a difference (certainly more subtle than with 888/24 and Aardsync vs. USD) between Aardsync and Internal as the master clock, but which is better does seem to be much more (potentially) subjective than in the past.  Can you say what parameters, listening-wise, tell you that you are on the right track with clock integrity?  Thank you both in advance.  
Title: Re: Proper word clock implementation
Post by: Curve Dominant on November 08, 2004, 03:32:21 am
Quote:

posted by Bob Katz:
Eric Vincent wrote on Sun, 07 November 2004 05:11
Quote:

Quote:

posted by Lucas van der Mee:
technology and measurements are only one part of developing equipment. Of course we all want absolute values, facts. But experience has taught me that objectivity always comes after the subjective: why does something sound good? Instead of: this will make it sound good.


Lucas,

That statement makes it appear as if you should get out of the engineering business, and be an artist instead.




Well, I disagree. Lucas's primary contention is absolutely important. It IS the role of engineers and scientists to find out WHY something sounds good. Psychoacoustics is a science; just not as well-understood a science as the design-engineering science that we're discussing in this board.



Bob,

Let me try to explain the context of the point I was attempting to make.

Every artist has tools at his disposal which can introduce non-linearities into the audio so long and so far as he chooses, and to the desired degree.

But control over that audio depends upon a baseline signal chain which is clean, minimal, and accurate.

What I had suggested to Dr. Van Der Mee, is that the engineer's preference for coloration of that signal chain is counter-productive to the artist's means.

Let's use DSD as an example. The one-bit conversion inherent in DSD introduces correlated distortion which cannot be removed from the source audio. Some people may find that distortion "pleasing," but those who do not are stuck with it nonetheless.

Such as it is with the euphonics inherent in external clocking. Certain demonstrations of the "Big Ben Sound" may be "pleasing" to some ears. But let's face it: It is an unnecessary step introduced into the signal chain, which introduces artifacts which once recorded cannot be removed from the audio after the fact.

Which brings us around to the point I attempted to make: Let artists distort the audio, IF THEY WANT TO. But, let's not give design engineers carte blanche when it comes to building what they percieve to be "euphonic" effects into the kit.

If Apogee's kit is essentially a distortion box, then let them sell it as that. Hell, maybe I'll buy one, if that's what I know I'm getting. I LOVE distortion. But as a composer, I would like to have a clean signal path to work with as the baseline audio. THEN introduce "euphonic" distortion effects to the degree I desire, in a manner I can control, IF and WHEN I want it, and to a degree which I can determine.

Does that make sense?
Title: Re: Proper word clock implementation
Post by: Curve Dominant on November 08, 2004, 04:07:00 am
Quote:

posted by Bob katz:
Science and engineering includes the science of psychoacoustics, after all! And we're going to have to live with fuzzy-logic for a long time, guys... until we have a much firmer grasp on the psychoacoustics.


Look here, mate. I would encourage you engineering blokes to ditch the psychoacoustics shtick, and ditch the fuzzy logic stuff while you're at it.

I'm a lyricist and composer, so it might seem weird to hear that from me.

But when I'm recording an artist, I switch hats, and then it's all about the logic and the science.

If you want to go "fuzzy," then write some songs, and you'll be an "artist" and you can get as fuzzy as you damn well please.

But when I'm recording a compelling artist performing a profound statement....I don't want no friggin' fuzz in the signal chain.

If I do, I can instantiate the SansAmp plugin, and add fuzz. But I don't want design engineers forcing it on me defacto.

Quality A-D conversion is pretty cut-and-dry business. Jitter, monotonicity, opamp rise time, capacitor dialectric absorption coefficients...y'all do your job, and we do ours.
Title: Re: Proper word clock implementation
Post by: bobkatz on November 08, 2004, 11:37:36 am
Eric Vincent wrote on Mon, 08 November 2004 03:32



If Apogee's kit is essentially a distortion box, then let them sell it as that. Hell, maybe I'll buy one, if that's what I know I'm getting. I LOVE distortion. But as a composer, I would like to have a clean signal path to work with as the baseline audio. THEN introduce "euphonic" distortion effects to the degree I desire, in a manner I can control, IF and WHEN I want it, and to a degree which I can determine.

Does that make sense?



What you are advocating is that engineers advertise their "color-making" boxes as making colors. At the end of the day, it can be effectively argued that EVERY single processor and converter is a color-making box. Just that some produce less color than others. Your argument has merit but I don't think in practicality it will fly.
Title: Re: Proper word clock implementation
Post by: jfrigo on November 08, 2004, 12:48:20 pm
re: Apogee

Dan's "bedside manner" may be brusque, but without Apogee opening their strong technical claims to peer review, we can't be sure of their merit. I understand the need to keep trade secrets, but perhaps some objective test results could be posted so that at least we could see the proof in the performance, if we can't get a decription of the innovation.


re: Eric & psychoacoustics

We can't just ditch psychoacoustics. Though it should not be used (or abused) as an excuse for bad engineering, it's an important part of making the technology work. Without knowing a lot about how we hear, one can't make good decisions about the sample rate and bit depth required for transparent reproduction, including how low dither must be before we don't hear it, and the best way to impliment noise shaping curves, and even some filter issues. The problem is when the factual basis is perverted into pseudo-science and used to falsely support fantastic claims; but this is not the only science that is used so.

Title: Re: Proper word clock implementation
Post by: Lucas van der Mee on November 08, 2004, 06:04:45 pm
Bob, we would love having you test the Big Ben. Please email Max to arrange for the unit.

But let me put a couple of things straight, because there seems to be some misinterpretation on what we claim and say.
We never said it improves every converter. We would not be able to substantiate this, since we simply haven’t tried them all, that would almost be impossible. But we did find improvements with most of the converters we tried, the exception being converters using a SRC for clocking, and got similar reports from users in the field. One of the known examples is the popular Digidesign 192 I/O unit.
However, we do not claim we lower the jitter of a converter by clocking it to a Big Ben, that is not what we are saying.
Let me quote from our website:

A Cure for the Jitters – Apogee’s new “C777” Clock
At the heart of any word clock regeneration is a phase lock loop (PLL). The PLL is a combination of analog and digital elements without the ability to dynamically adapt to the nature of the external clocks. In other words, the better the clock performance, the less flexibility it has. And conversely, the more flexibility there is in a PLL, the less effective it is at jitter reduction. Apogee’s new C777 is built to address these inadequacies. The C777 utilizes an entirely digital process that Apogee has developed using the most advanced Direct Digital Synthesis (DDS) technology available along with DSP based digital filtering. The result is the most aggressive jitter reduction ever. With the flexibility and stability provided the C777, Big Ben is able to re-clock devices with excessive jitter and function as if it were the master clock. The ultimate in flexibility combined with unprecedented levels of performance make the C777 unequaled in the world of clocking.


In other words, when you HAVE to clock to an external , jittery, source, you will find that your results will be much better if you have the Big Ben do the clocking first and clock your device to the Big Ben. This is objective, measurable and verifiable.

Then there is the much more difficult to explain phenomenon of converters starting to sound “better” when clocked to the Big Ben as a master clock only; the C777 running on internal. That is something we found empirically and like I said in my previous posting, we researched and optimized.


Eric, I have learned in my life that the danger with many (technical) engineers is that they start to think they are God; they think they know it all and can explain it all. I take a different approach, which is not unique, there are many with me. I see the scientific world as a way to explain what we experience.
The process of learning is isolating entities from its environment. By isolating we simplify and are able to make a model that resembles our experience. Yet very often we find that an entity we isolate, turns out to be much more complex than we originally thought. Or the connection we thought that was there, is not the connection, but the result of something bigger we had not identified yet. My school book example: In the medieval period scientists believed, mice would be created by old rags (Spontaneous Creation or Abiogenesis). Sounds funny and stupid, right? Still, nowadays scientists find everyday they make “goofs” like that all the time.
I have a university degree in Electronics comparable to a Masters, have a scientific background, I rely on math for the design of my converters, I use measuring equipment to get verifiable and objective test-results of what I build. But I also know there are parameters that are not being measured, simply because we haven’t quantified them yet. For that reason I trust my ears as much; they very often tell me a lot more than what I measure. A good example is the choice of op-amps in my analog design (which I love btw, Dan. I was just being ironic).
Last, the Big Ben is not a distorto box. The most common comments we get from experienced listeners, who use the BB are: converters sound less harsh, imaging is greatly improved, more depth and a much clearer soundstage. If you think that is the result of distortion, I am happy to learn more about this new insight, fill me in.

Jay, do you think we would allow everyone to use the best audio measuring gear in the world by Rohde & Schwarz (…does this get me a free unit?...) on their booth to test our equipment, if we felt our units were not up to par? After the show, the R&S rep raved about how impressed people were by the measurable quality of the AD16X.

Lucas van der Mee
Sr. Design Engineer
Apogee Electronics
Title: Re: Proper word clock implementation
Post by: bobkatz on November 08, 2004, 07:00:24 pm
[quote title=Lucas van der Mee wrote on Mon, 08 November 2004 18:04]Bob, we would love having you test the Big Ben. Please email Max to arrange for the unit.

I may have lost Max's email, can you please provide it in a PM.

Sounds good.  The good thing is that you have come to save the day, Lucas, and that I urge you to get together with Max to change the misleading language that he uses to sell the box. It is 180 degrees apart from the language that you use in your post. I would have (little or) no quarrel if Max had used the correct technical language.

Quote:



However, we do not claim we lower the jitter of a converter by clocking it to a Big Ben, that is not what we are saying.




Well you and Max need to agree on what language you are going to use. Max was trying to tell us that engineer A and B and C are raving over the Big Ben's ability to make a converter perform better than on Internal Clock! Which as Dan Lavry points out, is technically impossible unless the internal clock of the converter under test is a piece of crap.

Quote:



In other words, when you HAVE to clock to an external , jittery, source, you will find that your results will be much better if you have the Big Ben do the clocking first and clock your device to the Big Ben. This is objective, measurable and verifiable.




Guys...the above is VERY different in meaning and intent from what Max was saying in this thread. It agrees with Bob Ludwig's reactions as well. "If you must externally clock (as in "lock to video") then consider the Big Ben.

So I think the extent of my testing should be to see how TYPICAL converters perform with the Big Ben or without the Big Ben when they MUST be externally clocked. Now we're cooking with gas!!!

Quote:



Then there is the much more difficult to explain phenomenon of converters starting to sound ?better? when clocked to the Big Ben as a master clock only; the C777 running on internal. That is something we found empirically and like I said in my previous posting, we researched and optimized.




This is a much harder contention to sell and disagrees with the theory IF a converter is well-made and contains a good internal clock. But there are a lot of interior converters on the market. We shall see.

I'm curious if the stable of converters that I can test will improve with the Big Ben:

Cranesong HEDD-192, Benchmark A/D, TC Electronic System 6000, DCS


Place your bets  Smile


Best wishes,


Bob
Title: Re: Proper word clock implementation
Post by: Max on November 08, 2004, 07:46:42 pm
Well, I really intended to be done with this thread. I am a little confused at your responses, Bob. you said:

Quote:

Sounds good.  The good thing is that you have come to save the day, Lucas, and that I urge you to get together with Max to change the misleading language that he uses to sell the box. It is 180 degrees apart from the language that you use in your post. I would have (little or) no quarrel if Max had used the correct technical language.


What language was it that mislead you specifically as I do not see the conflict?

Quote:

Well you and Max need to agree on what language you are going to use. Max was trying to tell us that engineer A and B and C are raving over the Big Ben's ability to make a converter perform better than on Internal Clock! Which as Dan Lavry points out, is technically impossible unless the internal clock of the converter under test is a piece of crap.


I stand by this statement. it is the same thing Lucas is stating at the end of his post where he says:

Then there is the much more difficult to explain phenomenon of converters starting to sound ?better? when clocked to the Big Ben as a master clock only; the C777 running on internal. That is something we found empirically and like I said in my previous posting, we researched and optimized.

Like Lucas is saying above and as I said in my previous posts, there are many many users out there that have used a Big Ben to clock a single converter and noticed a marked improvement in the sound quality and this includes converters across the quality spectrum. Again, Bob, my point in bringing this up to you was your comment that if a converter sounds better externally clocked there must be something wrong with the converter. I do not agree with this assessment based on our results, as I am not prepared to say that many of the converter brands we tested and have heard about from end users are bad or poorly designed. To do so would be the blatant salemanship and marketing that you previously were accusing me of and I am not prepared to do that, even though doing so would most likely result in us selling more Apogee products.


Quote:

This is a much harder contention to sell and disagrees with the theory IF a converter is well-made and contains a good internal clock. But there are a lot of interior converters on the market. We shall see.

I'm curious if the stable of converters that I can test will improve with the Big Ben:

Cranesong HEDD-192, Benchmark A/D, TC Electronic System 6000, DCS


Place your bets  Smile


Best wishes,


Bob


Bob, we look forward to getting you the unit and hearing back regarding your results. While under the circumstances it will not be possible for us to be there as I would have wished, I believe the process you outlined in your previous post as to how you would go about testing will work for us. I would still appreciate interfacing with you directly on this so feel free to email me at your convenience at max@apogeedigital.com
Title: Re: Proper word clock implementation
Post by: danlavry on November 08, 2004, 07:54:44 pm
Hello Bob.
I.   Re: External box driving A/D

You said:

"Well, I disagree. Lucas's primary contention is absolutely important. It IS the role of engineers and scientists to find out WHY something sounds good. Psychoacoustics is a science; just not as well-understood a science as the design-engineering science that we're discussing in this board."

The short answer is: If this manufacturer wishes to sell a solution based on “sound better” that is fine. However, to fulfill as you suggested, an honest attempt at figuring what is going on, takes more then “we were dumbfounded” or “it sounds good”. An honest attempt could be to say “MORE JITTER but it sounds better”, which is very different from “A cure for all jitters”.

With all due respect, it is my considered opinion that your “open minded approach” postpones and delays the obvious. Their claims are akin to being able to fly by waving ones arms.  

At some point, the technical engineer should put his foot down! More jitter is worse. Using fewer bits is worse.  Audio is full of “I don’t know why but it sounds better”, which in my view is, most often, an excuse for selling lesser performance gear. That is “marketing”. Name the car that breaks down “reliant”, the car with bad gas mileage “econo-line”, and the slow one “Cheetah”.

"If there is an apparent discrepancy between any or some of the measurements, it IS OUR job as engineers and scientists AND listeners to try to resolve that discrepancy."

I do not claim to be a listener. I work at creating transparent uncolored sound that the artists then color according to their own sensibilities but measurements are indeed what I pay attention to. In this case, there is no need to measure jitter because the concept is fundamentally flawed. An external clock box cannot reduce jitter inside an A/D box.

"For example, we need to subjectively/objectively test Apogee's contention that an external clock will improve the sound of "any converter", regardless of the scientific theory. Regardless that EVERY other respectable converter designer would disagree."

At some point things get out of hand. What about a claim that 8 bits sound better? How about a claim that we need 100MHz of audio bandwidth? There is not enough time to investigate every claim made up by some sales guy. We engineers know there are some lines that cannot be crossed. More jitter is bad. Lucas admitted that in his first post.

"Let's summarize: Apogee claims that the Big Ben improves the sound (and lowers the jitter, it's in their ads) of EVERY converter (except ones which contain the Big Ben ). Would it be sufficient to disprove their claim if I find a converter which does not improve with the Big Ben? Would it be sufficient to disprove their claim if I find a converter which sounds audibly worse when fed from the Big Ben? What about the measurements? What if the measurements and double-blind listening tests disagree? What then?"

What would you do if I told you that doubling the voltage on a resistor reduces instead of doubles current flow? It would contradict Ohm’s Law! Will you bother to test that? I know I will not. As you know and stated, listening tests are most susceptible to subjectivity. Basic science and engineering is not. We are not philosophizing, we are dealing with Ohms law, Nyquist, Laws of thermodynamics, basic engineering theories.

"Responsible engineers should be prepared to deal with all of the above possibilities."

I disagree! A responsible engineer is one that while using his/her solid background can decide where to draw the line between what is fundamentally correct, fundamentally flawed and what may be fuzzy. There is no perpetual motion. Ohms law is solid. Nyquist was right. And so on.

"My point being that there are explanations to everything... we just have to find them!"

Some explanations are marketing based and go against engineering. Let’s get back to the subject at hand, and do a “think like a designer” exercise.

Problem: There is a box. It is a closed box, you can not see what is inside. It may be an AD with a perfect zero jitter clock circuit. It may be a jittery circuit. It may even be completely broken, not working. Your mission, should you accept it, is to send a message to the box telling the circuitry to do a better job. You are allowed to send any signal imaginable into that box, via a cable connection. But remember one thing, you are allowed to talk to the AD box but the box cannot talk to you.

THE BOX DOES NOT SEND ANY INFORMATION BACK TO YOU. YOU DO NOT KNOW EVEN FOR AN INSTANT, WHAT IS TAKING PLACE INSIDE THAT CLOSED BOX.

Say you knew that the PLL and clock inside the AD work perfectly. Then you should send a message with a “content to do nothing.” Say you knew that every third clock edge is late by 1 nsec. Then you should (assuming it is possible) send a message to generate every third clock 1 nsec earlier.

Sending a message to a closed A/D box to correct what it does (WHILE NOT KNOWING what it does or what its internal problems may be eg: noisy power supply) is fundamentally an impossible task. I am sorry, the Big Ben box cannot possibly do this.
.
In a very GENERAL sense one can know what is inside each A/D. There is a clock with some level of random jitter activity. Then perhaps some signal correlated jitter (or maybe not), perhaps some power line induced jitter (or maybe not), or jitter due to noise on the supply (or not)? These factors differ from AD to AD…

Conceptually and theoretically speaking, one has a chance to make some corrections when there is feedback. But here there is NO FEEDBACK. Feedback REQUIRES that the AD chassis send information back to the external clock unit, telling it what is going on at the AD. But in this case, Big Ben is driving the AD box, and there is one way communication- NO FEEDBACK. Therefore the claims stand against ENGINEERING INTEGRITY.

I will be generous and say that giving credence to such fallacies only encourages a form of marketing that confuses those who want to be well informed. By insisting that this forum be a technical one we provide a service to be trusted! This is the place where Ohm’s Law always works and jitter is always bad…

NEXT TOPIC:
II. Re: Clock itself

How good a clock can one make with the DDS technology? Again, it is a simple matter to see through the faulty logic.

To talk about secret technology of the 21st century is bogus. Hints about filters and “analog crap”  is  also bogus. How do I know so much about what Lucas did without knowing what he did? I will explain. Again, let’s “think like a designer”:

Whatever one measures or generates calls for a REFERENCE. A yard stick. Your old voltmeter had a scale, like a ruler. The “needle” moved against the scale telling you the voltage (the printed “ruler scale” on the front panel was the reference). The new digital voltmeter has a precision internal DC reference (such as a reference zener diode). Any measurement is always compared to a reference. This procedure is fundamental to science and engineering. If the reference is moving up and down by 1%, the measurements are off by 1%.

One can only get as good as ones reference. The only way to do better is to get a more accurate reference, a better, more accurate “yard stick”. If that better reference is say good to 1mV than you could know that your outcome to 1mV. One cannot correct things beyond the finest reference one has.

The same statements are true for measuring or generating current, resistance, frequency and mechanical construction to name a few..  One can have all sorts of circuitry help the outcome get closer to the reference, but one can not have a better performance then the reference itself. The DDS technology relies on a reference clock. If this reference clock can not beat the jitter of an internal crystal, all the circuits in the world can not help it. IT CAN NOT.

So I do not need to know any “secrets” those guys are talking about. No amount of DSP or hints about filters or circuits AFTER the reference clock are going to do a bit of good, and no amount of disrespectful talk towards this “theoretical guy” is going to alter the reality of what I say. The “war” for low jitter takes place at the reference clock first and foremost. Call it theory if you wish, but it is also a fact of life.

A good implementation of internal clock uses a crystal that resonates in a fundamental mode.  But fundamental crystals can not be made for very high frequencies.  DDS requires a higher frequency reference clock, thus the crystal that operates on more jitter. If the reference has jitter, you can not do a thing about it. Can you filter it with DSP? Build a notch filter to remove unwanted energy? Nuts! The notch frequency moves with the reference clock (it moves with the jitter). And so on…

Moving from theory to practice and “hands on” engineering, it is best to get a reference clock to drive a signal as directly as possible to the AD. Going from a higher jitter reference clock through all sorts of DDS circuitry, then to a driver circuit, then to a cable (possible interference), then receiver IC, then a PLL (more jitter) driving a VCXO (or anything with higher jitter then a fixed fundamental crystal)…. That can not compete with a good fixed crystal very near the destination. Check Mate!

I just explained why two of the Apogee arguments are incorrect. So why did they get a tech award? Must be marketing $ certainly not technological achievement.

At this forum technical know-how has value, without being subject to questions by listening tests which are accompanied by hype, name dropping,  subjective arguments and last but certainly not least- politics.

Best Regards
Dan Lavry

Title: Re: Proper word clock implementation
Post by: jfrigo on November 09, 2004, 01:11:03 am
Lucas van der Mee wrote on Mon, 08 November 2004 15:04

Jay, do you think we would allow everyone to use the best audio measuring gear in the world by Rohde & Schwarz (...does this get me a free unit?...) on their booth to test our equipment, if we felt our units were not up to par? After the show, the R&S rep raved about how impressed people were by the measurable quality of the AD16X.

Lucas van der Mee


I'm sure you worked very hard on your unit and are proud of its performance. I don't doubt Apogee's pursuit of quality. My only point is that it's reasonable and even expected for the audio public to say "show me." This is not to say whether the box is a huge success or or not as I haven't tried it;  just that it's a smart consumer who doesn't simply take the word of every advertiser he encounters in the magazines.
Title: Re: Proper word clock implementation
Post by: Schallfeldnebel on November 09, 2004, 11:36:42 am
When I was looking on the internet site of Apogee at the Big Ben, I got the impression this unit is much more then a simple clock generator.  Maybe I did not understand this correctly. I get the impression the BB also functions as a format converter and a sample rate converter.

If you combine those functions with an ultra low jitter internal clocking, this unit starts to work as a de-jitter generator. Mr. Daniel Weiss made such a device years ago, and called it Clockwork.

I have tested in the past intensively with the Clockwork, and it made a lot of sense hanging it between CD transports and DA converters. Often it sounded like having a better transport, or having beter DA converter.

All those studio’s which are claiming to have a better sound using the BB, are not very specific in how they use it. If they use it, to clock their workstation or recorders, it might be that what they are doing is, sending a better clocked signal to the DA converter. In my experience indeed  reclocking the signal gives better sound at the DA stage.

Or, if a signal coming from an AD converter is sent into the BB, and the BB is used to clean up that signal, so BB functions as a reclock generator, again it would be possible that the end result is sounding better. But it is not the AD conversion which sounds better, it is the final signal going to the DA converter which is cleaned up. In this case the output signal from that AD converter must have had quite much pollution.

Also when BB is used as a central clocking unit, and it is clocking an AD converter, but also your digital workstation and recorders, again you are cleaning up the whole signal path, and it is again the DA converter which resposible for the better sound, because it got a clean clock.

Only when an AD converter is clocked from the BB, and the AD converter is clocking with it’s own AES output signal your workstation, and the workstation is clocking with it’s output signal your  DA converter, I cannot believe it will sound better, at least I would be very surprised.

I have clocked several AD converters external, and the sound was always worse, and when the output signal jitter was measured, it was always 2 to 3 times higher than when internally clocked. There was one exception where it did not matter, it sounded the same and measured the same, and this experience I had with a converter from Daniel Weiss.

So we should not make the mistake when concluding external clocking of AD converter gives a better sound, that we clock all other equipment after the AD also directly out of the BB, because then you clean up the signal of the whole system, which can have a large effect on the DA converter, so it is the DA converter making us believe our AD sounds better.


Erik Sikkema

Title: Re: Proper word clock implementation
Post by: Duardo on November 09, 2004, 03:04:27 pm
Quote:

Or, if a signal coming from an AD converter is sent into the BB, and the BB is used to clean up that signal, so BB functions as a reclock generator, again it would be possible that the end result is sounding better.


That wouldn't really make a difference, would it?  Once a signal's been converted, there's not really anything you can do to "clean up" the signal, is there?  This would make the signal sound better if the clock was less jittery than the DA's clock, but it wouldn't do anything about "cleaning up" the signal coming out of the A/D.  Only way to do that would be to clock the A/D to the BB, right?

-Duardo
Title: Re: Proper word clock implementation
Post by: Schallfeldnebel on November 09, 2004, 03:32:37 pm

What I mean is as follows, and I will illustrate this with a  situation in my work.

When I am recording, my control room can be more than 300ft away from stage. I have AD converters on stage. At the end of the AES cables the jitter gets much higher. Although this is not a problem for my recorders, I usually place a Weiss Clockwork in the middle of the cable at 150ft to clean up. If I would hang a DA converter at the end of the cable at 300ft and switch the Clockwork off, it sounds worse than switched on.

If the BB can be used to clean up the signal path between AD and DA, you might think your AD is sounding better, the cause for the better sound is the DA gets a less jittery signal.

Erik Sikkema
Title: Re: Proper word clock implementation
Post by: Max on November 09, 2004, 04:13:52 pm
Duardo wrote on Tue, 09 November 2004 20:04

Quote:

Or, if a signal coming from an AD converter is sent into the BB, and the BB is used to clean up that signal, so BB functions as a reclock generator, again it would be possible that the end result is sounding better.


That wouldn't really make a difference, would it?  Once a signal's been converted, there's not really anything you can do to "clean up" the signal, is there?  This would make the signal sound better if the clock was less jittery than the DA's clock, but it wouldn't do anything about "cleaning up" the signal coming out of the A/D.  Only way to do that would be to clock the A/D to the BB, right?

-Duardo


Correct. Once the signal has been converted, there is nothing you can do to clean up the jitter during the A/D conversion. In the above scenario where you are using a CD transport digitally out to a D/A, inserting Big Ben in the path would reduce the jitter created by the transport before it gets to the D/A, therefore making the D/A converter perform and sound better. Any artifacts due to jitter that have already been recorded during the A/D conversion is of course not being corrected by Big Ben (it's too late once it is recorded).
Title: Re: Proper word clock implementation
Post by: Max on November 09, 2004, 04:23:36 pm
Quote:

So we should not make the mistake when concluding external clocking of AD converter gives a better sound, that we clock all other equipment after the AD also directly out of the BB, because then you clean up the signal of the whole system, which can have a large effect on the DA converter, so it is the DA converter making us believe our AD sounds better.


Erik Sikkema




In some cases this is true as I described above. This is also what I was referring to in my first post when I said that all things are never equal when it comes to clocking. Anyone who is using a separate D/A converter of any kind is re-clocking, for example.

On the other hand it is important to note that there are many devices nowadays such as the Digidesign 192 and Rosetta 800 for example, that have A/D and D/A in the same box. Typically, these types of units are are designed to clock from a crystal for both A/D and D/A conversion when set on internal mode. When not being clocked internally, the A/D and D/A converters switch over to the PLL. Again, the interesting thing here is is that we have experienced improvements in sound quality when units such as these are clocked externally to Big Ben.
Title: Re: Proper word clock implementation
Post by: Schallfeldnebel on November 09, 2004, 04:36:37 pm
Max wrote:

"Correct. Once the signal has been converted, there is nothing you can do to clean up the jitter during the A/D conversion. In the above scenario where you are using a CD transport digitally out to a D/A, inserting Big Ben in the path would reduce the jitter created by the transport before it gets to the D/A, therefore making the D/A converter perform and sound better. Any artifacts due to jitter that have already been recorded during the A/D conversion is of course not being corrected by Big Ben (it's too late once it is recorded). "


Max, you are right. You know this, I know this. But it is not what I mean. Those engineers who have found out that their AD conversion sounds better when a BB is used, since it is not clear from the Apogee homepage how they use the BB, it is for me unclear if they are only cleaning up the path in between their converter or cleaning up the converter itself.

Again, there is often a lot inbetween AD conversion and DA, workstation, EQ, cabling, name it, all places where jitter can added to the process, and if the BB is cleaning that path up, you might think it has cleaned up the AD, and it has not.
I would only believe in one test situation, wordclock out BB to Wordclock in converter, AES converter out directly in DA, and listen.

As long as the BB can be used as inbetween reclocker, it can clean up jitter anywhwere in the chain, not only in the AD converter.  

Erik Sikkema
Title: Re: Proper word clock implementation
Post by: danlavry on November 09, 2004, 07:17:05 pm
“What I mean is as follows, and I will illustrate this with a situation in my work.

When I am recording, my control room can be more than 300ft away from stage. I have AD converters on stage. At the end of the AES cables the jitter gets much higher. Although this is not a problem for my recorders, I usually place a Weiss Clockwork in the middle of the cable at 150ft to clean up. If I would hang a DA converter at the end of the cable at 300ft and switch the Clockwork off, it sounds worse than switched on.

If the BB can be used to clean up the signal path between AD and DA, you might think your AD is sounding better, the cause for the better sound is the DA gets a less jittery signal.

Erik Sikkema”


Erik, your comment regarding the confusion between AD and DA is technically valid. Here is my reply:

If  that “jitter cleaning device” which is a PLL box  (phase lock loop circuits) is 50ft from the DA  OK. How about at 10 feet? WHAT ABOUT INSIDE the DA? That is what I do. –A good PLL based circuit (or better) is inside my DA! And my final re-clocking circuit IS based on the MUCH SUPERIOR VCXO crystal technology.

I am willing to accept that external re clocking box for a DA can improve jitter when the DA PLL is poor. But this is not the best solution. Internal jitter removing is better. Just get a good DA with good jitter handling capabilities.

Why?

Say you removed “all” the jitter and you have the perfect AES signal (or SPDIF). It is going to get pretty jittery as soon as you enter the DA.

Hawksford in his AES paper (about 10 years ago) “Is the AES/EBU Digital Audio Standard Flawed?” sheds light on the jitter problem .The AC coupling of a transformer (or capacitor) sets a limit on the capability of the digital audio link. That limitation makes the whole AES (or spdif) waveform wobble up and down. That wobble is DATA DEPENDENT and causes DATA MODULATED JITTER. The conclusion is to do the jitter removal AFTER the AC coupling at the DA input..

True, every little bit helps. I have nothing against getting into the DA with zero jitter but that help will prove minimal. The lion share of the jitter removal task needs to take place after the AC coupling – inside the DA.

At the risk of repeating myself lets get back to AD clocking:

1. It is best to use internal crystal when possible  

2. Even when using external or internal master clock, crystal technology is superior. It is true that DDS provides lower cost and more flexibility, but it is at the expense of more jitter. That is true for AD and DA.

3. Regarding “cleaning jitter” at all sorts of places in the chain: It matters most at the AD. It matters at the DA. It matters where there is an SRC. The rest is just about data transfer. The jitter requirements for data transfer are a lot less stringent – the idea is to make sure you don’t drop bits.

Max, it is technically impossible for your FUNDAMENTALLY FLAWED CONCEPT of external clocking an AD, to improve A/D performance and your statement about D/A is marginal.  Also, the Apogee Big Ben clock cannot outperform other clocks that do have fixed crystal and VCXO based designs. Jitter removal inside a box for DA is FAR BETTER to jitter removal externally. Jitter reduction of AD with external clock is an imposible task, as I explained in my previous post.

Also, External jitter removal leave open a path, (between the external clock and the AD or DA) where more jitter is accumulated and/or generated. The internal jitter removal circuit approach eliminates that very serious fact of life (I explained it above for the DA case).  

I have previously mentioned that this forum is the wrong place for you to make appearances.  I have no more time to spend responding to sales talk. Your company engineer already said that your advertisements incorrectly state his product design’s capabilities. There are many chat rooms entirely devoted to listening. Please find one.

Regards
Dan Lavry
Title: Re: Proper word clock implementation
Post by: bobkatz on November 09, 2004, 07:59:23 pm
ErikS wrote on Tue, 09 November 2004 15:32



If the BB can be used to clean up the signal path between AD and DA, you might think your AD is sounding better, the cause for the better sound is the DA gets a less jittery signal.

Erik Sikkema



How about using a D/A converter that is immune to jitter in the first place? That kind of functionality should be built into converters. There should be no need for a $1500 add-on. And as Dan Lavry points out, the dejittering really MUST be inside the converter. Because any external dejittering box has to face the reality that the following box DOES NOT SEE ITS clock directly; there is interface jitter and the jitter gain of the PLL within the following device.

I'll say no more, because I don't want to prejudice my listening tests any more than they already are. They will have to be blind in order to convince anyone.
Title: Re: Proper word clock implementation
Post by: Terry Demol on November 09, 2004, 08:42:19 pm
bobkatz wrote on Wed, 10 November 2004 00:59

ErikS wrote on Tue, 09 November 2004 15:32



If the BB can be used to clean up the signal path between AD and DA, you might think your AD is sounding better, the cause for the better sound is the DA gets a less jittery signal.

Erik Sikkema



How about using a D/A converter that is immune to jitter in the first place? That kind of functionality should be built into converters. There should be no need for a $1500 add-on. And as Dan Lavry points out, the dejittering really MUST be inside the converter. Because any external dejittering box has to face the reality that the following box DOES NOT SEE ITS clock directly; there is interface jitter and the jitter gain of the PLL within the following device.

I'll say no more, because I don't want to prejudice my listening tests any more than they already are. They will have to be blind in order to convince anyone.


Bob,

What dacs are you going to use for evaluation? Benchmark
is out as Apogee has clearly stated that the effect of BB
does not apply to ASRC based dacs.

Cheers,

Terry

Title: Re: Proper word clock implementation
Post by: danlavry on November 10, 2004, 01:41:27 pm
I agreed to moderate this forum for the sake of communicating TECHNICAL knowledge. There are many other forums for discussing listening tests, psychoacoustics and the ear.
I am going to use my discretion to remove and block massages that go against the intent of this forum, as stated by the rules.

Best Regards

Dan Lavry
Title: Re: Proper word clock implementation
Post by: danlavry on November 10, 2004, 02:26:41 pm
“Apogee has clearly stated that the effect of BB
does not apply to ASRC based dacs.

Cheers,
Terry”


I heard such statements before. Is it true that SRC based DA is completely immune to jitter? If such were the case, all the DA makers would be putting an SRC in front of the DA’s. Other then increased latency (which is an application specific drawback) the DA jitter would be completely solved. Is it so?

The idea of SRC is to take data arriving with a jittery clock, and make it into data with a steady clock. Indeed, one can use an internal crystal oscillator, which beats any “jitter external cleaning device”, and is even slightly better than a  VCXO (pull able crystal) based PLL circuits.

But this is not the “end of the story”. There is something else that is taking place while substituting the jittery input clock with a clean fixed internal crystal. That something else is a complete and elaborate computation calls SRC.

The relationship between the incoming jittery clock (at the input of the SRC) and the fixed clean clock (at the output of the SRC) is not fixed. The relative timing between each input clock and the output “clock train” is not well defined. The phase is totally unknown, and the “specific per sample time location” is “bouncing” forwards and backwards in time. The more jitter the more it bounces.

Each individual output sample (the data itself), is figured out with it’s own unique set of coefficients. The selection of the coefficient set is based on the relative timing of the INDIVIDUAL output sample to the incoming data sample time locations. One ends up with a filter computation, where a complete set of coefficients are is being “replaced” changed from sample to sample…

So the DATA ITSELF is being altered. Not only does it get to go through a complete LPF (low pass filter) computation, the filter coefficients are timing dependent. Jitter does modulate the coefficient selection process. The more jitter, the more it alteration of the data.

Therefore, SRC is not a "perfect solution" to eliminate problems caused by jitter:

The clean internal PLL and clock is an attempt to fight the physical time jitter. The advantage of this method is the fact that the DATA IS LEFT INTACT.

The SRC solution provides a great clean clock, but IT ALTERS THE DATA in order to make it “match the new clean clock”.

The artifacts of both approaches (clean PLL based and SRC) are very different.

The better the SRC, the closer we get to perfection. The better the PLL, the closer we get to perfection.      

My point is: the assumption (or marketing claim) that SRC based DA’s offer a complete solution to jitter, is not correct. There is a price to be paid – a SRC in series!

BR
Dan Lavry
Title: Re: Proper word clock implementation
Post by: Roland Storch on November 10, 2004, 04:56:42 pm
danlavry wrote on Wed, 10 November 2004 19:26



Therefore, SRC is not a "perfect solution" to eliminate problems caused by jitter:

The clean internal PLL and clock is an attempt to fight the physical time jitter. The advantage of this method is the fact that the DATA IS LEFT INTACT.

The SRC solution provides a great clean clock, but IT ALTERS THE DATA in order to make it ?match the new clean clock?.

The artifacts of both approaches (clean PLL based and SRC) are very different.

The better the SRC, the closer we get to perfection. The better the PLL, the closer we get to perfection.      

My point is: the assumption (or marketing claim) that SRC based DA?s offer a complete solution to jitter, is not correct. There is a price to be paid ? a SRC in series!

BR
Dan Lavry



Thanks for this explanation. That means there is no perfect DAC which works synchronious to an incoming digital signal.

Wouldn?t than a bigger buffer be the best solution? I mean if I took a harddisk to record the incomming data from a digital source and play that back later I could have a perfect internal clock for my DA converter fed with data from the HD.

If I needed less "delay" couldn?t I just take more buffer size? I wouldn?t have a really synchronized DAC but if I only needed the best playback I could determine the incoming frequency, lets say 48k, then set the internal clock of the DAC to 48k and play. So there would be a small drift, maybe some milliseconds in an hour, but for playback only I had the best solution.

This would not work in studio where the signal has to be synchronious. But for playback only for my ears without further manipulation of the signal, i.e. any player for only hearing a digital source, wouldn?t?that be a perfect (because decoupled) DAC?
Title: Re: Proper word clock implementation
Post by: Bob Olhsson on November 10, 2004, 06:13:19 pm
The bottom-line of all this stuff is the actual quantity and the spectral character of the artifacts produced by each particular PLL or SRC implementation.
Title: Re: Proper word clock implementation
Post by: Nika Aldrich on November 10, 2004, 07:04:15 pm
Ahh, I'm lovinig the direction of this thread.  It's great to see some of this stuff addressed and understood in public.  To answer the questions:

Roland Storch wrote on Wed, 10 November 2004 16:56


Thanks for this explanation. That means there is no perfect DAC which works synchronious to an incoming digital signal.


Well we have to be careful about that.  Sure, it may not work "perfect" to the analog signal, but who cares about perfection?  We're talking about human beings and human limitations.  What we care about is that the variations in the eventual clock signal used for reproduction (the jitter at the D/A) is low enough and of a character enough that it manifests itself as distortion that is too low in amplitude or outside the audible spectrum.  If we can create a system that produces low enough jitter related distortion that we can't hear it then why do we care about "perfect" synchronization?

Quote:

Wouldn?t than a bigger buffer be the best solution? I mean if I took a harddisk to record the incomming data from a digital source and play that back later I could have a perfect internal clock for my DA converter fed with data from the HD.


Sure, so the tradeoff is latency and cost of the larger buffer versus jitter.  Compromises are obviously made along the way.  If you have a HUGE buffer you have to have a lot of local memory and you induce potentially very long latency, but can reduce the jitter to inconsequential, no?  If you only have 1 sample of buffer then you're going to have to reconcile pretty quickly, and that reconciliation causes jitter, yes?

Quote:

This would not work in studio where the signal has to be synchronious. But for playback only for my ears without further manipulation of the signal, i.e. any player for only hearing a digital source, wouldn?t?that be a perfect (because decoupled) DAC?


Sure, and while Dan may not toot his own horn, that is essentially what his gold series converters do.  They have something like 10 seconds worth of buffering in them.  OK, it's not really 10 seconds worth of buffering.  It's 10 seconds worth of buffering when you know the maximum amount the clock will drift.

That is important.  Let me repeat.  If you can know a small detail about the clock - say how much the maximum amount is that it can drift - then you can make a system that caters to that.  No longer do we have to have an infinitely large buffer.  We can instead have only enough of a buffer to last a certain amount of time, and we have a lot more flexibility about how we design that PLL, no?  I mean, if it's only going to drift by x amount then our PLL can be that slow in recalibrating itself.  

I'm not sure if I'm talking at your level or not.   Do you know the axle/wheel analogy for the PLL?

Nika
Title: Re: Proper word clock implementation
Post by: Andy Peters on November 10, 2004, 08:29:10 pm
Just thinking out loud here.  It seems to me that the rather obvious solution is a small elastic buffer.  I wouldn't be surprised if this is already implemented in various places.  Or maybe I'm just missing the point ...

The problem, as I see it, is really that the incoming serial audio data stream is asynchronous to the converter's clocks.  This is because we have to recover the clock and data from the AES/EBU or S/PDIF signal and a recovered clock will always have some amount of jitter.  This jitter will clearly be worse than a crystal.

The classical way of crossing clock domains in a digital system is to use a FIFO. The write side and the read side are totally decoupled and the clocks and read and write enables are separate.  All that is required is that control logic on both sides monitor the FIFO status.  The writing side must monitor a full (or almost-full) flag and hold off writing until their is enough space.  The reading side must monitor an empty (or almost empty) flag to ensure that it doesn't read from an empty buffer.

In the case of an isochronous system like our DAC, we can't hold off writes, as the new incoming samples will simply fall on the floor; nor can we hold off reads, lest we leave holes in our playback.  However, we can further constrain things such that we know that the read and write rates are basically the same. There's no sample rate detection, but that's the point: we use a switch to select the proper sample clock crystal.   This means that the FIFO should never completely fill nor should it ever completely empty (unless we stop writing to it, of course).

The FIFO itself needn't be large, maybe 16 samples ought to do it.  Of course reducing the FIFO size decreases latency.  One way to do it would be to shift the incoming data bits of each sample into a serial-to-parallel shift register and then write that parallel word to the FIFO.  As each sample is shifted in, it's written to the FIFO.

The read side is the reciprocal; when the reader determines there's enough (half full?) samples in the FIFO, it turns on the read state machine and pops a word from the FIFO and shifts it out. The reader continues until it sees an empty FIFO (the indicating that the transmitter stopped sending new samples), at which point it outputs all zeros (simple mute).  The point of all this is that the read side is clocked by the "clean" clocks.

Yeah, it's all just a handful of lines of Verilog ...

--a
Title: Re: Proper word clock implementation
Post by: Nika Aldrich on November 10, 2004, 09:40:23 pm
Andy,

Yes, you're doing great.  So lets look at this.

Andy Peters wrote on Wed, 10 November 2004 20:29

Just thinking out loud here.  It seems to me that the rather obvious solution is a small elastic buffer.  I wouldn't be surprised if this is already implemented in various places.  


Yup, exactly, but that doesn't necessarily solve all of the problems.  How much buffering?  And what do you do when you get buffer overload or underload?

Quote:

The problem, as I see it, is really that the incoming serial audio data stream is asynchronous to the converter's clocks.  This is because we have to recover the clock and data from the AES/EBU or S/PDIF signal and a recovered clock will always have some amount of jitter.  This jitter will clearly be worse than a crystal.


Yup!

Quote:

The classical way of crossing clock domains in a digital system is to use a FIFO. The write side and the read side are totally decoupled and the clocks and read and write enables are separate.  All that is required is that control logic on both sides monitor the FIFO status.  The writing side must monitor a full (or almost-full) flag and hold off writing until their is enough space.  The reading side must monitor an empty (or almost empty) flag to ensure that it doesn't read from an empty buffer.

In the case of an isochronous system like our DAC, we can't hold off writes, as the new incoming samples will simply fall on the floor; nor can we hold off reads, lest we leave holes in our playback.  However, we can further constrain things such that we know that the read and write rates are basically the same. There's no sample rate detection, but that's the point: we use a switch to select the proper sample clock crystal.   This means that the FIFO should never completely fill nor should it ever completely empty (unless we stop writing to it, of course).

The FIFO itself needn't be large, maybe 16 samples ought to do it.  Of course reducing the FIFO size decreases latency.  One way to do it would be to shift the incoming data bits of each sample into a serial-to-parallel shift register and then write that parallel word to the FIFO.  As each sample is shifted in, it's written to the FIFO.

The read side is the reciprocal; when the reader determines there's enough (half full?) samples in the FIFO, it turns on the read state machine and pops a word from the FIFO and shifts it out. The reader continues until it sees an empty FIFO (the indicating that the transmitter stopped sending new samples), at which point it outputs all zeros (simple mute).  The point of all this is that the read side is clocked by the "clean" clocks.


OK, so how does that reconcile The following problem:

The uber-stable interior clock is clocking at 44.1kHz, but being a clock that has a tolerance of around 150ppm, perhaps it is actually 44.09999992 kHz.  The data coming in from the AES stream is also coming in at 44.1kHz, but it is also a crystal device with a given tolerance of around 150ppm.  Its clock happens to think that 44.1000002kHz is 44.1kHz.

We're about to play a 1 hour CD.  

In other words, can we actually ever truly decouple the clocks on these digital devices, given that we have tolerances to deal with?  The purpose of the clock reconciling here is not necessarily to reconcile a stable clock against a jittery one.  We also have to deal with clocks that have two completely different senses of time, and somehow bring them together.  In other words, we HAVE to couple the two clocks in some way, somehow.  Your FIFO idea, however, is certainly the first step in solving the problem, and Dan's designs do indeed incorporate a FIFO buffer in them.

Nika
Title: Re: Proper word clock implementation
Post by: danlavry on November 10, 2004, 09:42:19 pm
“Just thinking out loud here.  It seems to me that the rather obvious solution is a small elastic buffer.  I wouldn't be surprised if this is already implemented in various places.  Or maybe I'm just missing the point ..”

I normally would not mention my concept but your post was begging for a direct response. If you go to www.lavryengineering.com under the support section, find a pdf file for the DA924 product manual. On page 11 there is a short description of my jitter removal, trade marked CrystalLock(TM).

It is a buffer based concept. Even if one has 10 nsec jitter, be it random or a systematic 1KHz sine wave modulation. The DA clock “does not move”.

CrystalLock(TM) makes other jitter removal schemes seem like “stone age” technology and I have been making it since 1997  

Dan Lavry
Title: Re: Proper word clock implementation
Post by: Roland Storch on November 11, 2004, 04:43:20 am
Nika Aldrich wrote on Thu, 11 November 2004 00:04


Sure, and while Dan may not toot his own horn, that is essentially what his gold series converters do.  They have something like 10 seconds worth of buffering in them.  OK, it's not really 10 seconds worth of buffering.  It's 10 seconds worth of buffering when you know the maximum amount the clock will drift.

That is important.  Let me repeat.  If you can know a small detail about the clock - say how much the maximum amount is that it can drift - then you can make a system that caters to that.  No longer do we have to have an infinitely large buffer.  We can instead have only enough of a buffer to last a certain amount of time, and we have a lot more flexibility about how we design that PLL, no?  I mean, if it's only going to drift by x amount then our PLL can be that slow in recalibrating itself.  

Nika



Thanks Nika. That and the above you wrote was very helpful.
Title: Re: Proper word clock implementation
Post by: Nika Aldrich on November 11, 2004, 09:35:12 am
Roland,

Thanks.  I think that three things help bring this conversation around:

1.  The axle/wheel analogy.  If you don't know that one let me know and I'll explain.

2.  An understanding of how jitter manifests itself on a signal.

3.  An understanding of the ear's limitations of hearing that result.

With those three things we can build a clocking circuit that can reconcile two different clocks and we can do it in such a way as to prevent the ear from hearing it.

If you need further clarification on those issues let me know.

Nika.
Title: Re: Proper word clock implementation
Post by: Roland Storch on November 11, 2004, 06:32:06 pm
I can imagine what the wheel/axle analogy means but please explain to be certain.

About nr. 2 I think I know enough, but I never know.

Nr. is very interesting for me.

So I would really appreciate if you told me more - and other participants may appreciate it too.
Title: Re: Proper word clock implementation
Post by: Andy Peters on November 11, 2004, 08:02:05 pm
Nika Aldrich wrote on Wed, 10 November 2004 19:40


Andy Peters wrote on Wed, 10 November 2004 20:29

Just thinking out loud here.  It seems to me that the rather obvious solution is a small elastic buffer.  I wouldn't be surprised if this is already implemented in various places.  


Yup, exactly, but that doesn't necessarily solve all of the problems.  How much buffering?  And what do you do when you get buffer overload or underload?


As I said, you set it up such that the internal clock (that actually clocks the DAC) and the external clock (derived from the AES/EBU input data) are nominally the same, then you're filling and emptying the buffer at the same rate.  There's no danger of underrun or overrun.  If the buffer is, say, 16 samples deep, then if you "mostly" fill it before starting to read, you'll always have samples in the buffer.

Quote:

OK, so how does that reconcile The following problem:

The uber-stable interior clock is clocking at 44.1kHz, but being a clock that has a tolerance of around 150ppm, perhaps it is actually 44.09999992 kHz.  The data coming in from the AES stream is also coming in at 44.1kHz, but it is also a crystal device with a given tolerance of around 150ppm.  Its clock happens to think that 44.1000002kHz is 44.1kHz.

We're about to play a 1 hour CD.  

In other words, can we actually ever truly decouple the clocks on these digital devices, given that we have tolerances to deal with?


You can do no better than the internal clock's tolerance, but that tolerance is not affected by anything on the other side of the FIFO.

Quote:

The purpose of the clock reconciling here is not necessarily to reconcile a stable clock against a jittery one.  We also have to deal with clocks that have two completely different senses of time, and somehow bring them together.  In other words, we HAVE to couple the two clocks in some way, somehow.  Your FIFO idea, however, is certainly the first step in solving the problem, and Dan's designs do indeed incorporate a FIFO buffer in them.


Think about this again.  The FIFO read side is clocked by the DAC clock, which is completely independent of the recovered AES/EBU clock.   The DAC clock tolerance has nothing to do with the input-side tolerance.  The write side could be jittering all over the place but it has no bearing on the other side of the wall.  As long as the elastic buffer has enough depth, there's no problem.  I dunno how else to explain it.

Here's another example of how we can decouple clock domains.  It's the typical PCI sound card.  Standard PCI is 32 bits wide and clocked at 33 MHz.  Of course that's much much faster than Red Book audio, and is completely uncorrelated with audio.  

(Note: Contrived example not involving DMA engines!) During playback, the PC takes a buffer full of audio data from its memory and writes it as fast as possible over PCI into a FIFO on the sound card.  The PC then goes off and does something else as the sound card's "engine" reads audio data from the FIFO at a rate dictated by the sampling frequency.  The host must monitor the state of the FIFO (or wait for an interrupt); when it gets "almost empty" it bursts the next batch of data over PCI to the FIFO.

Again, as long as the buffer never completely empties, the playback engine doesn't have any problems, as its clocked by its own oscillator.  The playback engine doesn't even know that it's being fed data at a much higher speed.

--a
Title: Re: Proper word clock implementation
Post by: bobkatz on November 12, 2004, 12:23:36 am
Terry Demol wrote on Tue, 09 November 2004 20:42



Bob,

What dacs are you going to use for evaluation? Benchmark
is out as Apogee has clearly stated that the effect of BB
does not apply to ASRC based dacs.




Terry, Dan would rather that we not change the focus of his forum to listening tests. This leaves a gap, but, well, pick another forum here and we'll consider moving the thread there. So I consider your question, "off topic" for Dan's forum.
Title: Re: Proper word clock implementation
Post by: Jørn Bonne on November 12, 2004, 04:42:29 am
Good idea, Bob!

But before we do, I think we should extend our apologies to Max and Lucas, who were subjected to an unacceptable hammering, when they tried, quite rightly, to clear up misunderstandings in earlier posts regarding Apogee products. The attacks on them were out of line with one of Dan's ground rules: "Please communicate in a polite and courteous manner"

I, for one, much preferred Max and Lucas's civilized approach to communication, in their attempts to offer a "second opinion", over the yelling and insults from some of the other posters. Why all the anger?

There seem to be disagreement about basic principles here. For the benefit of the readers, we would ideally need additional "opinions" from other experts working to solve the problems of conversion (designers etc). But I doubt they will be forthcoming, the atmosphere in this thread being what it is.

Moving the thread might lead to a more open, unbiased discussion.

Kind regards,

J
Title: Re: Proper word clock implementation
Post by: Nika Aldrich on November 12, 2004, 08:51:43 am
Roland Storch wrote on Thu, 11 November 2004 18:32

I can imagine what the wheel/axle analogy means but please explain to be certain.


Think of two wheels connected by an axle.  On the right wheel put a hash mark at 12:00.  On the left wheel put a hash mark at 6:00.  Now, cut a section out of the axle and replace it with a spring.  As you turn the right wheel the spring builds up tension.  Eventually the left wheel starts to spin.  It eventually reaches the same speed as the right wheel.

Now, slow the right wheel down.  The left wheel does the same, delayed a little in time.  Speed the right wheel up and slowly the left wheel catches up, though it is a delayed, slow reaction.  While the right wheel is turning, hit it and bump it a few times to give it some erratic behavior and that erratic behavior will largely be buffered from the left wheel.  Right?

This is a PLL circuit. The right wheel is the incoming clock.  The left wheel is the outgoing clock and the PLL is the spring in the middle.  The hash marks represent synch - if the hashmark on the left wheel laps or gets passed by the hash mark on the right wheel then the wheels fall momentarily out of synch - one clock has exceeded the other clock.

Now, think about what would happen in this analogy if we shortened the spring or made it more taught?  What would happen if we lengthened the spring or made it looser?  How would these behaviors manifest themselves when I speed up or slow down the right wheel - or give it these occasional, random jerks that cause its "jittery" behavior?  How will the looser or taughter spring manifest itself with regards to the PLL trying to keep the clocks from lapping each other under these conditions?

What are our limitations?  What if I then told you that the right clock would never have erratic behavior more than x amount.  What does that do to our limitations?

There's a primer.  Let me know your thoughts and perhaps we'll be able to figure out how to resolve these clocking issues.

Nika
Title: Re: Proper word clock implementation
Post by: Nika Aldrich on November 12, 2004, 09:04:40 am
Andy Peters wrote on Thu, 11 November 2004 20:02

Think about this again.  The FIFO read side is clocked by the DAC clock, which is completely independent of the recovered AES/EBU clock.   The DAC clock tolerance has nothing to do with the input-side tolerance.  The write side could be jittering all over the place but it has no bearing on the other side of the wall.  As long as the elastic buffer has enough depth, there's no problem.  I dunno how else to explain it.

Here's another example of how we can decouple clock domains.  It's the typical PCI sound card.  Standard PCI is 32 bits wide and clocked at 33 MHz.  Of course that's much much faster than Red Book audio, and is completely uncorrelated with audio.  

...

Again, as long as the buffer never completely empties, the playback engine doesn't have any problems, as its clocked by its own oscillator.  The playback engine doesn't even know that it's being fed data at a much higher speed.

--a


Andy,

You may have thought this all through and I may just be missing a link, but I still don't understand how your circuit deals with the situation of two clocks that think that 44.1k is different.

Let's start with a scenario where the two clocks are out of synch by one 44100th of a second.  The external clock keeps shoving the info into the buffer and the internal clock keeps reading it from the buffer, but after 1 second there's 1 less sample in the buffer than there was to start with.  After two seconds there're 2 fewer samples in the buffer.  After every second we lose a sample in the buffer.  How are you going to reconcile this situation?  Do you have an infinitely large buffer?  Or do you recalibrate the clocks so that they both see time as roughly the same?  How do you do this?

Then play with a different example.  In this example the clocks are out of synch by one 44100th of a second, again, but this time, the clock in the DAC heats up after 5 minutes, and upon heating up it changes its sense of time, and it falls back INTO alignment with the external clock.  Whatever way you reconciled the situation above, how does this situation respond to that fix?

Nika
Title: Re: Proper word clock implementation
Post by: danlavry on November 12, 2004, 01:24:26 pm
bonne wrote on Fri, 12 November 2004 09:42

Good idea, Bob!

But before we do, I think we should extend our apologies to Max and Lucas, who were subjected to an unacceptable hammering, when they tried, quite rightly, to clear up misunderstandings in earlier posts regarding Apogee products. The attacks on them were out of line with one of Dan's ground rules: "Please communicate in a polite and courteous manner"

I, for one, much preferred Max and Lucas's civilized approach to communication, in their attempts to offer a "second opinion", over the yelling and insults from some of the other posters. Why all the anger?

There seem to be disagreement about basic principles here. For the benefit of the readers, we would ideally need additional "opinions" from other experts working to solve the problems of conversion (designers etc). But I doubt they will be forthcoming, the atmosphere in this thread being what it is.

Moving the thread might lead to a more open, unbiased discussion.

Kind regards,

J
Title: Re: Proper word clock implementation
Post by: jcoyle on November 12, 2004, 02:26:53 pm
bonne wrote on Fri, 12 November 2004 09:42

Good idea, Bob!

But before we do, I think we should extend our apologies to Max and Lucas, who were subjected to an unacceptable hammering, when they tried, quite rightly, to clear up misunderstandings in earlier posts regarding Apogee products. The attacks on them were out of line with one of Dan's ground rules: "Please communicate in a polite and courteous manner"

I, for one, much preferred Max and Lucas's civilized approach to communication, in their attempts to offer a "second opinion", over the yelling and insults from some of the other posters. Why all the anger?

There seem to be disagreement about basic principles here. For the benefit of the readers, we would ideally need additional "opinions" from other experts working to solve the problems of conversion (designers etc). But I doubt they will be forthcoming, the atmosphere in this thread being what it is.

Moving the thread might lead to a more open, unbiased discussion.

Kind regards,

J
Title: Re: Proper word clock implementation
Post by: Duardo on November 12, 2004, 10:12:15 pm
Quote:

This was a thread discussing whether an external clock will give better jitter performance versus the internal clock. And Dan made the statement that technology wise ... INTERNAL clock will give less jitter than external clock in a well designed converter. Apogee's Big Ben was never even mentioned.

Now Let's look at the entrance of Apogee into the thread...


You're right, Big Ben was never mentioned specifically, but on the first page of the thread Bob Katz said this:

Quote:

I think it's a good idea you're using the Apogee's as the master clock. They're certainly going to do better on internal sync than locking to any of the other current devices in your system. And only a measurement and listening test of the Apogee's would reveal if they would do better when fed some external "high-end" clock. And if they did do better on external sync, that would be a denigration of the Apogees, not a praise of the external clock.


Don't you think that that statement merits a response, both because Apogee feels that their clock will improve their converter's performance, and because Bob says that it's a denigration of their converters if they do?

Quote:

Here Max states, not only is Dan Wrong, but he is living in the 20th technology-wise and clinging to theories that don't hold up, because the Big Ben proves them wrong. That is NOT CORDIAL OR CIVILIZED, especially when DAN NEVER MENTIONED APOGEE.


Again, Bob said "Manufacturers, it seems are the most clueless"...and, as a manufacturer, probably of the most talked-about master clock on the market right now, do you not think that statement warranted a response from Apogee either, regardless of whether Bob meant it as a slight against Apogee or not?

It's been made clear that this is not the place to discuss listening tests and so on, but I for one am quite interested in hearing what Bob has to say once he's put the Big Ben through its paces.  He seems to have a fairly open mind about it.

I understand that Max may have wandered into territory that Dan doesn't want in this forum, but I do agree that they could have been treated with more respect as well.  

-Duardo
Title: Re: Proper word clock implementation
Post by: Sam Lord on November 13, 2004, 04:17:42 pm
Dan,
Thank you for riding herd over this very long and informative thread, which has spanned the subjects of ADC and DAC clocking, internal clocks compared with external, and various clock circuits.  I hope not to tread over covered ground, but I have two questions:

1) The case for internal clocking of ADCs made by yourself, Bob Katz (here and in his mastering book), and *most* engineers is pretty clear.  However, whenever you have multiple channels, as all ADCs I've seen do, the clock needs to be distributed internally as evenly as possible, no?  So in an 8-channel ADC box, do you use one clock wired to all channels with the best possible shielding, and perhaps a trigger (for a clean square wave) placed at each individual ADC chip?  Or do you have separate clocks at each channel, each synched to a designated master with a PLL circuit?  It seems to me that the first approach is best, but I'd like an expert's opinion.

2)  If an external clock box is very, very good, such as the dCS 998 claims to be (+-1 ppm shipped), used with good, short connections, how good must an internal clock be to give better encoding accuracy?  Also, when you are clocking many channels, does the electrical noise from an internal clock at some point favor the change to an external clocking enclosure?

3)  This might belong in another thread (sorry), but here goes: what is the best multichannel digital transmission medium now available in pro audio gear?  I was wondering about AES47 vs MADI (especially optical) vs firewire and others.  I will be buying soon, and firewire is claiming ever-more adherents.  Any thoughts?    
Thanks, Sam          
Title: Re: Proper word clock implementation
Post by: Sam Lord on November 13, 2004, 04:23:55 pm
Oops...
[/quote]
Also, when you are clocking many channels, does the electrical noise from an internal clock at some point favor the change to an external clocking enclosure?
[/quote]

Sorry, I meant "does noise from one beefy clock or multiple individual (per channel) clocks at some point favor the change to an external clocking enclosure?
Thanks again.
Title: Re: Proper word clock implementation
Post by: danlavry on November 13, 2004, 05:22:25 pm
Sam Lord wrote on Sat, 13 November 2004 21:17

Dan,
Thank you for riding herd over this very long and informative thread, which has spanned the subjects of ADC and DAC clocking, internal clocks compared with external, and various clock circuits.  I hope not to tread over covered ground, but I have two questions:

1) The case for internal clocking of ADCs made by yourself, Bob Katz (here and in his mastering book), and *most* engineers is pretty clear.  However, whenever you have multiple channels, as all ADCs I've seen do, the clock needs to be distributed internally as evenly as possible, no?  So in an 8-channel ADC box, do you use one clock wired to all channels with the best possible shielding, and perhaps a trigger (for a clean square wave) placed at each individual ADC chip?  Or do you have separate clocks at each channel, each synched to a designated master with a PLL circuit?  It seems to me that the first approach is best, but I'd like an expert's opinion.

2)  If an external clock box is very, very good, such as the dCS 998 claims to be (+-1 ppm shipped), used with good, short connections, how good must an internal clock be to give better encoding accuracy?  Also, when you are clocking many channels, does the electrical noise from an internal clock at some point favor the change to an external clocking enclosure?

3)  This might belong in another thread (sorry), but here goes: what is the best multichannel digital transmission medium now available in pro audio gear?  I was wondering about AES47 vs MADI (especially optical) vs firewire and others.  I will be buying soon, and firewire is claiming ever-more adherents.  Any thoughts?    
Thanks, Sam          


The idea is to:
Get minimum jitter at all ADC’s
Have all AD's clock together

I get best results by far with a central clock, distributed to all ADC. Each ADC may have an individual clock buffer, no more than that. A single buffer will out perform even the best “individual PLL per circuit”. Of course one needs to pay attention to keeping the signals clean.  

”2) If an external clock box is very, very good, such as the dCS 998 claims to be (+-1 ppm shipped), used with good, short connections, how good must an internal clock be to give better encoding accuracy? Also, when you are clocking many channels, does the electrical noise from an internal clock at some point favor the change to an external clocking enclosure?”

I am not familiar with the DCS unit. I am surprised to see a +/-1ppm accuracy, I did not think it matters. Accuracy is not jitter, and jitter is what matters.

Regarding the “noise from clocking” – it should not be confused with “just any” digital activity around sensitive analog areas. Why? An AD always has an analog section that is tightly coupled to the digital clocks. Certainly with AD’s there are frequency ranges where the slightest noise couples right into the AD and causes a lot of trouble (contamination). But there are also specific frequencies where the ADC rejects the coupled energy. Those frequencies are at multiples of the sampling rate. The AD clocks are placed at those frequencies (by design). The distance between the AD master clock, the SCLK and the LR clocks and the analog on an IC is nearing the dimensions of nano technology. But the interference is completely rejected because it falls on 0Hz and multiples of Nyquist, thus can be later filtered without impact on the audio range.

So the answer is: No, the noise from internal clock is not much of an issue.  

”3) This might belong in another thread (sorry), but here goes: what is the best multichannel digital transmission medium now available in pro audio gear? I was wondering about AES47 vs MADI (especially optical) vs firewire and others. I will be buying soon, and firewire is claiming evermore adherents. Any thoughts?
Thanks, Sam”

I am not sure. It is a moving target. There is a new firewire IC I need to check out…

BR
Dan Lavry

Title: Re: Proper word clock implementation
Post by: Sam Lord on November 16, 2004, 09:00:49 pm
Thanks Dan, your answers were very helpful.  Sorry for the slow reply; no need to reply unless you so wish.

[quote title=danlavry wrote on Sat, 13 November 2004 17:22]

[Accuracy is not jitter, and jitter is what matters.]

Oh that's right, jitter is variation in the clock period, but accuracy refers to the number of clock cycles per time interval, I think.  I remember Stereophile actually listening to music from a modest CD player way back, and just putting an FM amp on the *digital* output, and hearing music because of the correlated jitter.

>>Regarding the “noise from clocking”..... the interference is completely rejected because it falls on 0Hz and multiples of Nyquist, thus can be later filtered without impact on the audio range.<<

Oh, now I see.  Thanks again, Sam
Title: Re: Proper word clock implementation
Post by: Andy Peters on November 19, 2004, 06:53:11 pm
Nika Aldrich wrote on Fri, 12 November 2004 07:04

You may have thought this all through and I may just be missing a link, but I still don't understand how your circuit deals with the situation of two clocks that think that 44.1k is different.

Let's start with a scenario where the two clocks are out of synch by one 44100th of a second.  The external clock keeps shoving the info into the buffer and the internal clock keeps reading it from the buffer, but after 1 second there's 1 less sample in the buffer than there was to start with.  After two seconds there're 2 fewer samples in the buffer.  After every second we lose a sample in the buffer.  How are you going to reconcile this situation?  Do you have an infinitely large buffer?  Or do you recalibrate the clocks so that they both see time as roughly the same?  How do you do this?


Ah, sorry for the delay (busy!) but I do see your point!

-a
Title: Re: Proper word clock implementation
Post by: danlavry on December 15, 2004, 01:08:31 pm
Nika said:

“I posted a lengthy list of questions to help me iron out the ongoing dispute on the "clocks" issue. I understand Dan was kind enough to reply at length to those questions. I received a phone call when I was on the road today advising me of the response and asking me to read Dan's comments. I came home and the thread is gone.

What happened? Where is it? Why?

Nika”



This is the clock thread!

In the thread “Something I do not Understand” you wrote. “I’m trying to get you to show how. You’re the engineer. It’s all math. Math doesn’t lie, and engineers don’t make mistakes, right?” I took much time to write a good explanation and when you discovered I was right you did not acknowledge my effort or my position.

Instead, you immediately started another Clock thread demanding answers from Lavry and another company, in what you called an on-going dispute, stating that you and Bob need to get to the bottom of things. Your message tried to create a new format of “Dan answer this” and "Apogee answer that”.  As far as I am concerned there is no “on-going dispute”. I already stated technical facts about the clock that no one can or did dispute, and that is all I will do for now.

At the same time, on a thread “Understanding Dan’s 192Khz Arguments” you virtually repeated what I said and asked me to comment on it.

Dear Nika, I have a life and cannot be your personal tutor. Your level of activity here is welcome if you can bring new ideas instead of creating a circuitous rehash. I do not have time for such interactions. Also, it is inappropriate for you to pose as an authority, translator or moderator, while at the same time using direct sales techniques to sell your book.

One to one personal communications are now necessary so that I will not have to embarrass you. Please contact me directly.

Regards,
Dan Lavry


Title: Re: Proper word clock implementation
Post by: Nika Aldrich on December 15, 2004, 01:38:02 pm
danlavry wrote on Wed, 15 December 2004 13:08

This is the clock thread!


Great.  Will you please repost my previous questions.  

Quote:

In the thread ?Something I do not Understand? you wrote. ?I?m trying to get you to show how. You?re the engineer. It?s all math. Math doesn?t lie, and engineers don?t make mistakes, right?? I took much time to write a good explanation and when you discovered I was right you did not acknowledge my effort or my position.


No.  I found your answers to be unsatisfactory.  I did not think that they addressed at all the specific question I asked.  I found my answer another way - in a way that specifically addressed my questions, and explained the answer to me in a way that was helpful.  Rather than hash out the discrepancies between what you posted and what my question was I thought I should let it drop.  I appreciate your time in answering, however.  I do understand your post and certainly agree, but don't find it to be related to what I asked about.  I didn't see the need to belabor the issue.

Quote:

Instead, you immediately started another Clock thread demanding answers from Lavry and another company, in what you called an on-going dispute, stating that you and Bob need to get to the bottom of things.


I called it an ongoing dispute because apparently it is.  I did not demand answers.  I merely asked questions to help the bulk of us reconcile disputed and discontinuous and incomplete information - information that is not reconciling between the various parties involved.

Quote:

At the same time, on a thread ?Understanding Dan?s 192Khz Arguments? you virtually repeated what I said and asked me to comment on it.


People were asking for clarification and more specific examples on a point you had raised (and one that I had raised several times before over the past few years).  I tried to answer their question but provided the courtesy to you of not trying to speak on your behalf erroneously.

Quote:

Your level of activity here is welcome if you can bring new ideas instead of creating a circuitous rehash.


I believe that NONE of the questions I posted last night were a "continuous rehash."  If the answers to those questions can be found somewhere on this thread or in your other writings please direct me to them.  I don't see how it is "rehash" at all.

Quote:

Also, it is inappropriate for you to pose as an authority, translator or moderator, while at the same time using direct sales techniques to sell your book.


RE: posing as an authority, doesn't anyone who answers a question pose as an authority in some capacity?  How can you claim that noone should pose as an authority on an open forum?  As for a translator, you're the boss.  I disagree.  Not everyone understands everything the same way all the time.  Sometimes a different approach is needed.  We do a lot of re-interpretting information on these forums.  As for moderator - I don't pretend to be one.  I gave up that responsibility on this site many months ago.

Yes, I am using direct sales techniques to sell my book.  I don't see the consistency in your statement, however.  

Nika
Title: Re: Proper word clock implementation
Post by: danlavry on December 16, 2004, 07:40:13 pm
“Dan,

1. I have heard now from multiple places that you have claimed that the Big Ben "has no crystal clock inside of it," and that it is "synthesized."”


I never stated it, and in fact if you go and read that clock thread, I assumed the best case – that they do have an internal crystal, only at higher frequencies. One of my points was that a fundamental harmonic crystal is less jittery than higher frequency crystal. But that was not my main point. Here you are making me rehash what I already said.

I am not against external clocks when they are needed. I already stated that whenever one can use a reasonable quality internal clock, located inside, near the AD, they will have less jitter than using external clock, sending it through a transmitter, cable, receiver, and phase lock loop circuitry pushing and pulling a VCXO crystal that is more jittery than a fixed one...  Do I need to rehash it again?

I already explained that a “clock box” can not send a signal to reduce jitter in another box, without knowing what is taking place at that other box. The idea of being able to correct something without having knowledge of what it is bogus! I stated that EVEN IF the clock box were able to send ANY signal to correct the jitter inside the AD (AND IT DOES NOT!), it would not know what signal will correct the jitter, because the clock box DOES NOT KNOW what is wrong in the AD box ahead of it.
That is so because it is a one way communication- No feedback. The clock box sends THE SAME signal to any AD box. It does not change the type of signal for different AD’s. I already said all that. Do I need to rehash that?

But let us pretend that there is magic in the air, and the clock box knows exactly what to do, and what is taking place at the end of that cable leading to the AD box (and that is already VOODO ELECTRONICS). Given that the clock attempts to send a perfect square wave, perfect timing, no jitter… where will it put the message that contains the instruction to fix the jitter in the AD? The information to make things better is not sent via amplitude modulation. It is not sent via frequency modulation. It is sent via no modulation. Modulating the clock signal is adding jitter.

So putting it all together, you send a message via no modulation (thus no message) to fix something you have no idea about (that varies from AD to AD) and get less jitter?

Or is it that you send a message via modulation, thus adding jitter, though you do know which message to send, and the message will exactly cancel the jitter due to unknown factors, including random jitter?

Is it worthwhile bringing such possibilities in a technical forum? But then you said:

“3. Is it possible that the Big Ben puts out a clock signal with "tuned" jitter at certain frequencies such that clocking from it MIGHT yield better results than an internal clock in certain situations? I know you have said that if the Big Ben is the clock master then it has no feedback from the downstream devices with which to create an "improved" clock over the device's internal clock. Is it not possible, however, that Apogee knows enough about both the internal crystal and the PLL of a given A/D converter that they can design a clock system which, when coupled with the receiving PLL, produces jitter at more desirable frequencies than the internal clock in the A/D converter?”

No!!!

The path from clock input to AD clock input is VERY INDIRECT. It is not like you can “grab a clock edge” and “move it”. The PLL itself is a very “soft rubber band”. Also, no one has knowladge of random behaviour before it happend. After it happened it is too late to fix!  


This part of my message is not a rehash. I welcome occasional questions from people that do not know the fundamentals of electronics, or wish to get into an area that I can help with. But you have posted nearly one message for each 2 of mine in this forum (where I am the moderator). Most of them are not questions, but answers. You said:

“We do a lot of re-interpretting information on these forums. As for moderator - I don't pretend to be one. I gave up that responsibility on this site many months ago”.

Nika,
I do not know who “WE” are. The audio police? A special audio elite group?
My personal criteria for re interpreting is:

1. Anyone that wishes to take the role of re-interpretting should have some technical chops in the area of discussion.  

2. re-interpretting  is not the same as changing the signature at the end of a virtually repeated message. Such practice has an element of claiming undue credit.

Also, the reason many people come here is to learn. Yet, you stated “I found your answers to be unsatisfactory” where someone else response to the SAME THREAD was “That is the clearest description of stochastics relating to quantitization I have ever read, great job!”. There are a lot of people that appreciate the forum, and that is good enough for me.

”4. It is my impression that you are wholly opposed to the presence of Big Ben in the industry. From what my clients tell me there are several situations wherein it improves the clock signals in ways that make sense to me.”

First, this last point does not belong here. This forum is not where I come between you and what you said to the client. I would hope that after reading my clock posts one would have enough sense to at least realize that an external clock jitter is inferior to internal operation. I hope that people realize that 192KHz is a crock. I hope they understand that dynamic performance is 6dB/bit and there is no real such thing as real 24 bit performance. I hope they don’t pay 1000$/ foot for some gold cable. That and more may be my main motivation for running the forum, but I am not going to get between a salesman and his clients.

Regarding the statement about “several situations wherein it improves the clock signals in ways that make sense to me.”  This is an opportunity to contribute. Explain what situations and how and why is works, and what makes sense to you. By all means! I promise to be open minded to anything that is not complete bull.

More specifically to your question:
I am not in the practice of categorically opposing specific gear. I am against  hype and misleading the customers with a bunch of bull. So yes, I am pretty down on the Big Ben. What happened to Ardvark? They had a good clock and could not stay alive. Is it because people bought into that low jitter crock marketing bull story about “external clock will make it sound better”? I can not prove that is what did it to Ardvark , but I suspect so.

Dan Lavry  

Title: Re: Proper word clock implementation
Post by: Lucas van der Mee on December 16, 2004, 09:08:06 pm
Dan ,
You think it is ok to down our products without knowing how they work , spec out and/or perform. You use your platform for pure marketing against us, while you claim in the same response that is not on topic.

The reason why people prefer the Big Ben over Aardvark is very simple. We perform better specs wise, we offer more features and our box is better value.
You have yet to design a product that can beat the Big Ben, so I understand your frustration. But I do not understand your abuse of the PSW platform to release this anger. I consider that to be a sign of very low quality.


Lucas van der Mee,
Sr. Design Engineer
Apogee Electronics
Title: Re: Proper word clock implementation
Post by: bobkatz on December 16, 2004, 09:24:01 pm
Lucas van der Mee wrote on Thu, 16 December 2004 21:08

Dan ,
You think it is ok to down our products without knowing how they work , spec out and/or perform. You use your platform for pure marketing against us, while you claim in the same response that is not on topic.

The reason why people prefer the Big Ben over Aardvark is very simple. We perform better specs wise, we offer more features and our box is better value.
You have yet to design a product that can beat the Big Ben, so I understand your frustration. But I do not understand your abuse of the PSW platform to release this anger. I consider that to be a sign of very low quality.


Lucas van der Mee,
Sr. Design Engineer
Apogee Electronics



I was supposed to get a sample of the Big Ben to test. What happened?
Title: Re: Proper word clock implementation
Post by: bobkatz on December 16, 2004, 09:32:20 pm
Dan Lavry has been extraordinarily correct technically in all his responses about clocking and how it works, the principles of internal and external clocking, including one recently to Nika today. Since you cannot defeat the laws of physics, and therefore it should be Apogee's onus to demonstrate, via blind listening tests and published technical measurements how an external clock could possibly make a properly-performing converter perform "better".

I personally do not see Dan using this forum to promote one person's products or denigrate anyone else's. If he says he is "down on the Big Ben" it is not an attack on a product but on the unscientific methods by which that product has been promoted.  I do see Dan poking holes in other people's claims to beat the laws of physics, and he does it very well.

Please, let's not have another round of subjective arguments unsupported by technical measurements or facts. Please, let's not have another round of fruitless repeats of the same claims and counterclaims.

I have offered to perform objective tests and measurements and subjective blind tests. Send me a Big Ben for testing and let the tests begin.

BK
Title: Re: Proper word clock implementation
Post by: Level on December 16, 2004, 10:10:39 pm
That would be a nice one to put to bed Bob. Implementation of the clock, you could give the perspective. I use my internal clock and I just don't get all these clock issues, although I would love to be proven wrong, if my internal is giving problems I do not hear, nor anyone else has noticed to date.
Title: Re: Proper word clock implementation
Post by: Nika Aldrich on December 16, 2004, 11:12:22 pm
Dan,

Thank you for taking the time out of your busy schedule to reply.  Your post was a composite response from more than one post on more than one topic.  I am going to try to keep this technical and will reply to the non-technical, non-clock related issues offline or in another thread.  You are welcomed to post them back here if you wish, but I believe that that line of discussion detracts from the issue of clocks and jitter that we are slowly hashing out here.

danlavry wrote on Thu, 16 December 2004 19:40

I never stated it (that big ben has no crystal clock), and in fact if you go and read that clock thread, I assumed the best case ? that they do have an internal crystal, only at higher frequencies. One of my points was that a fundamental harmonic crystal is less jittery than higher frequency crystal. But that was not my main point. Here you are making me rehash what I already said.


OK, sorry, I don't mean to rehash.  I heard a quote attributed to you and it was apparently incorrect.  I do believe that Big Ben has some sort of crystal inside of it.  Moving on.

Quote:

I am not against external clocks when they are needed. I already stated that whenever one can use a reasonable quality internal clock, located inside, near the AD, they will have less jitter than using external clock, sending it through a transmitter, cable, receiver, and phase lock loop circuitry pushing and pulling a VCXO crystal that is more jittery than a fixed one...  Do I need to rehash it again?


No.  I agree with this in principle - I believe I have a chapter in my book dedicated to this concept.  I read Bob's book where he draws the same conclusion.  I have read your papers wherein you draw the same conclusion.  I think we are all pretty clear that, all things being equal, an internal clock running at the oversampled rate is better than the complex transmission path through which an external clock must go.  This is not to say that there aren't exceptions to that rule.  A box could have, for example, a poor quality internal crystal that is highly susceptible to jitter (poor grounding, power supply, whatever) yet a really stable interior PLL.  I would think this situation would be rare, but it is the type of situation that could yield an improved jitter spec from an external box.

Another example would be a box that has no internal PLL and a poor quality internal clock, but which accepts clock signals at the oversampled rate.  The first time I was ever exposed to improvements in signal quality using an external clock it was in this situation and the difference was very noticeable.

These are two situations wherein an external clock would yield an improvement beyond the internal clock in an A/D converter.  There are assuredly others.  But again, in principle, and all things being equal, in theory this should not happen.  The big question right now - especially looking at the market response - is whether or not this situation is a "theoretical" situation or one in practice.  Of course, the difference between theory and practice is that, in theory, they are the same...

Quote:

I already explained that a ?clock box? can not send a signal to reduce jitter in another box, without knowing what is taking place at that other box. The idea of being able to correct something without having knowledge of what it is bogus! I stated that EVEN IF the clock box were able to send ANY signal to correct the jitter inside the AD (AND IT DOES NOT!), it would not know what signal will correct the jitter, because the clock box DOES NOT KNOW what is wrong in the AD box ahead of it.
That is so because it is a one way communication- No feedback. The clock box sends THE SAME signal to any AD box. It does not change the type of signal for different AD?s. I already said all that. Do I need to rehash that?


OK.  I'm following this concept and always have followed this concept.  Allow me to make a dumb analogy:

In your D/A converter design, using Crystal Lock technology (I believe you call it) you incorporate what I think is a rather ingenious scheme.  You essentially decouple the internal clock from the external clock, only occasionally (every 10 seconds?) reconciling the two.  This is aided by a FIFO buffer that makes sure that we never lose a sample in a 10 second period.  When you reconcile the two clocks you make a change to the internal clock to accommodate the external clock's variance.  This yields very high jitter but at extremely low frequencies (.1Hz?) which yields sidebands so close to the fundamental that they are effectively inaudible.  

One might say "But you can't decouple the two clocks from each other - what if the external clock falls so far out of sync that you get gross error between the two - outside the capabilities of the buffer."   The answer is that you, Dan, know enough about clocks in use in the industry that you know exactly how much variance would be commonplace - I believe about 150 ppm?  I forget the specifics.  Knowing that crucial piece of information you can design that circuit such that the external clock and the internal clock will never fall so far out of sync that the buffer will overrun or underrun and you will never lose a sample.  It is that key piece of information - the maximum amount of variance you'll find in a common clock signal - that allows you to design this circuit.  If you didn't have that you'd have to design a much tighter PLL - and you do for purposes of slaving to SMPTE, etc.

OK, I might have a few details messed up in there.  Please forgive them.  What I am driving at is this:  Is it possible that Apogee (or Aardvark, or dCS, or some other company that makes an external clock) might know a crucial piece of information about circuit design of clocks or PLLs in typical devices (or specific devices) in this industry that would allow them to modify their external circuitry accordingly to take advantage of a "weakness" in the internal design?  This isn't an example of some sort of "feedback" that indicates something to the external box.  This is an example of knowing ahead of time what the weaknesses are so that they can be accommodated for certain, specific boxes that may be the target of the manufacturer's market?

Allow me to give a hokey example.  Perhaps there is some A/D converter on the market that, for whatever reason I don't know, performs better with a higher amplitude clock signal than typically comes off of the average crystal circuit.  However, I happen to know that certain PLLs generally put out a hotter signal.  As a designer, knowing this information in advance, and knowing that those A/D chips and that PLL design are prevalent in boxes in the specific market that I'd like to address, I could devise an external clock that would cause this specific circuit to sound better than on its internal clock.  

Now that's a totally made up example as you know, but I hope you understand my point.  Is there NO advance knowledge about clocks, PLLs, circuitry and the conventional box in this industry that a manufacturer could use in order to trump the conventional wisdom that internal is better?

I removed the next substantial portion of the message and will reply elsewhere so as to keep this on topic regarding clocks

Quote:

?4. It is my impression that you are wholly opposed to the presence of Big Ben in the industry. From what my clients tell me there are several situations wherein it improves the clock signals in ways that make sense to me.?


Quote:

First, this last point does not belong here. This forum is not where I come between you and what you said to the client.


Not what I said to the client.  What the client said to me.

Quote:

I would hope that after reading my clock posts one would have enough sense to at least realize that an external clock jitter is inferior to internal operation. I hope that people realize that 192KHz is a crock. I hope they understand that dynamic performance is 6dB/bit and there is no real such thing as real 24 bit performance. I hope they don?t pay 1000$/ foot for some gold cable. That and more may be my main motivation for running the forum, but I am not going to get between a salesman and his clients.


I am not a salesman.  I left Sweetwater some time ago.

Quote:

Regarding the statement about ?several situations wherein it improves the clock signals in ways that make sense to me.?  This is an opportunity to contribute. Explain what situations and how and why is works, and what makes sense to you. By all means! I promise to be open minded to anything that is not complete bull.


Big Ben has a lot of features.  One is that it is a clock distro - and a good one at that.  It therefore improves clock signals over daisy chaining through many boxes and in situations wherein T'ing off is inappropriate.

There are some boxes in this industry (the 888/24 from Digidesign, for example) that have very poor internal clock circuits and can be dramatically improved from an external device running superclock.

Big Ben will soon be able to generate wordclock locked to black.  A higher quality box than others on the market that can do that would be welcomed.  

These situations all make sense to me - especially the first at this point.  I am hearing a lot of positive feedback from people using Big Ben for its many features.  Some of these people are touting it for its as-of-yet-unreconcileable better performance than an internal clock, but not all.  

Quote:

More specifically to your question:
I am not in the practice of categorically opposing specific gear. I am against  hype and misleading the customers with a bunch of bull. So yes, I am pretty down on the Big Ben.


From what I have seen (and I don't see everything) Apogee has not been marketing Big Ben as a "fix your internal clock" solution.  It is their users that are touting it that way.  I could be wrong.

Quote:

What happened to Ardvark? They had a good clock and could not stay alive. Is it because people bought into that low jitter crock marketing bull story about ?external clock will make it sound better?? I can not prove that is what did it to Ardvark , but I suspect so.


I believe Aardvark's problems started long ago.  Perhaps that's fodder for a new thread, but their PC computer solutions did not do as expected.  Further, competition on the market and improved designs that obfuscated the need for Aardvark's solution eroded their market.  Those are just my suspicions.

Again, Dan, thank you for your time.  I think you can see that I mean well, am not trying to waste your time, and am trying to keep this technical, per your wishes.

Nika
Title: Re: Proper word clock implementation
Post by: Nika Aldrich on December 16, 2004, 11:38:55 pm
bobkatz wrote on Thu, 16 December 2004 21:32

Dan Lavry has been extraordinarily correct technically in all his responses about clocking and how it works, the principles of internal and external clocking, including one recently to Nika today. Since you cannot defeat the laws of physics, and therefore it should be Apogee's onus to demonstrate, via blind listening tests and published technical measurements how an external clock could possibly make a properly-performing converter perform "better".


Bob,

I hope you do not mind me posting a few segments from this archive here.  The following is a segment from a dialogue that took place from November 1 through November 7 of 2001 between David Josephson, Julian Dunn, yourself, and many others on this very topic.  The evidence that it would be possible to best the performance of an internal clock with an external clock was compelling enough that you changed something in your book:  Responding to Ben Gilsdorf's reason that an internal clock might be subject to more problems than it's accompanying PLL you write: "...And so do I... I'll have to revise my book....

BK"

This is the dialogue that led up to that:

Bob Katz:
>  I've started a running argument in the Digital Performer maillist. My
>  basic contention is that converters will inevitably perform better
>  (jitter-wise) on internal crystal sync unless the designer of the
>  converter was brain-dead or made some severe design compromises. No
>  one denies the contention that a phased lock loop can never exceed
>  the performance of a perfect crystal oscillator, the best it can do
>  is equal that performance. And typically, be worse, at least
>  measurably.
>
>  However, many critical listeners report better sonic performance from
>  converters such as the Digidesign 888 under stable external WC sync.
>  I find it very hard to believe that the internal crystal oscillator
>  of the 888 is so bad that it can't do better on internal! Perhaps
>  this is just a matter of "the lesser of two evils".

David Josephson:
Bob, you of all people should be careful about ignoring critical
listener reports in favor of supposed theory!  A crystal oscillator is
no magic device; the crystal is simply a very high Q mechanically
tuned circuit that passes signal predominantly at one frequency. Put
it in a feedback loop around an amplifier and it oscillates, same as
any other tuned circuit. But the bandwidth is not infinitely narrow,
the Q is not infintely high. Design of a low phase noise crystal
oscillator is not terribly difficult but it isn't as simple as connecting
it around some CMOS inverters like you see in most digital implementations.
Different crystal cuts have different bandwidths too; sometimes at the
expense of frequency stability over temperature. I can very easily
believe that a good phase locked loop would perform better in terms of
phase noise in the audio band than an ordinary crystal oscillator
would.

Bob Katz:
>  "In general, a typical converter will likely perform better on
>  internal sync than external".

David Josephson:
In general I think that's probably so. But the reader should be
cautioned that if it seems that it sounds better on external
sync, believe your ears.

Bob Katz:
>  It just has always stood to reason.. it makes so much engineering sense.

David Josephson:
As someone who has designed and built many crystal oscillators over
the years, mostly for radio applications, I will refrain from commenting
on whether it makes so much engineering sense.


later:

Bob Katz:
>  Then let's talk practical engineering at the price/performance levels
>  of a Pro Tools 888.
>  Now I don't know shinola about what's inside an 888, so I could be
>off base...
>
>  If, as opposed to the crystal-around-inverter-trick, Digidesign were
>  to buy an off-the shelf DIP-socket type rectangular metal can master
>  crystal oscillator, feed it from a clean, bypassed power supply,
>  grounding, etc. This is "engineering by Xerox" or at least buying
>  someone else's module, but it is doable at this level of dollars. But

David Josephson:
A quick look at off-the-shelf crystal oscillator modules in the Digikey
catalog reveals only one with a phase noise or jitter spec, and it's 30
psec typical, 50 maximum. It would be safe to assume that the more
generic parts are worse than that.

Bob Katz:
>  to my knowledge, you can't buy a good PLL off-shelf... it has to be
>  weaned and wined and dined and tweaked and double-PLLed until it's
>  blue in the face.

David Josephson:
If you are in the business of producing such equipment, it is customary
to have on your staff at least one competent analog designer with some
background in PLL design. Techniques for combining off-the-shelf VCOs,
phase detectors and reference oscillators to produce a predictable level
of PLL performance have been well documented for at least 30 years.

Bob Katz:
>  Anyway, the question, based on the above is: Which is more likely to
>  perform better, an ordinary PLL designed with a few hours of
>  engineering, or the off-the-shelf crystal oscillator module... which,
>  internal, or external, is likely to perform better?

David Josephson:
While I agree that in my experience, it is more likely that internal
clock will produce lower jitter at a given converter, there are enough
variables to make a blanket statement that it "will sound better" unwise.
There is also the question of the spectrum of the jitter; while the
PLL will almost certainly have higher jitter at low frequencies, a
crystal oscillator may well have significant phase noise (which becomes
jitter) at high frequencies in the audio band, which may mean accentuated
jitter artifacts at certain frequencies of program material (this is
possible with PLLs too).

later contribution by Julian Dunn:

>Another explanation is that the phase alignment under external WC sync
>reduces some crosstalk effect.  If the comparison of "sonic" qualities
>is made with the WC connected and a switch selecting between internal
>and WC sync then on internal there would be the potential for crosstalk
>from the WC input.


There was more.  I picked what I thought was probably critical to your revision on this topic in your book.  Of course, I made the same point in my own book, but I think you and I both allowed for the notion that in some conditions boxes like Big Ben COULD provide better results - or at least one should be open to it and do their own listening tests.  This notion is directly contradictory to the point made so emphatically by Dan when he writes, regarding whether or not Big Ben could improve on an internal clock, "No."

Cheers!
Nika
Title: Re: Proper word clock implementation
Post by: bobkatz on December 17, 2004, 01:59:17 pm
I am perfectly aware of all the nuances, Nika. Which is why more than once I have said, "A poorly-designed converter (one with a poor internal clock) can conceivably sound better on external sync".

Then you may recall, in this thread, the guys from Apogee replied something to the effect that "many respectable engineers using what are recognized as excellent converters feel that the Big Ben improves their sound".

This sounds exactly like a marketing attempt to look a the tail side of the penny instead of the head.

Bob (and any properly-schooled engineer) says:

"By definition, if a converter sounds better on external clock, then it must be defective. At the very least its internal clock must be poorly designed, and its PLL may also be poorly designed".

I believe that the marketing geniuses at Apogee have turned this fact all around into marketing hype. "Put our expensive bandaid on your bad converters to make them sound better".

I will attempt to settle these issues and many more with definitive measurements and blind listening tests. Apogee is supposed to send me a Big Ben. This was at least a month ago and I haven't seen it. Of course I can see their reluctance to send a unit to a "biased" tester, but if I'm biased, it's thoroughly on the side of science. Fortunately, I have ears, too!

BK
Title: Re: Proper word clock implementation
Post by: Nika Aldrich on December 17, 2004, 02:07:25 pm
Bob,

I understand.  I guess I don't see their reports of what other people say as a marketing department gone amuck.  They aren't claiming that the box does one thing or another.  They are claiming that other people have stated that it does one thing or another.  I remember Aardvark claiming "plug this in and it will make your stuff sound better."  That's a bold marketing claim, in my opinion.  Claiming that some reputed engineers found it improved their signal chain is perhaps incomplete, but not overly hyped from my perspective.  Perhaps that's a nuanced position?  Perhaps it's a fine line?  I'm not sure.  Again I am led to believe that it's Apogee's users that are marketing its sound quality so well - not Apogee.

I hope Apogee sends you one.  To your credit, your tests have always been scientifically conducted and complete with thorough results and whatever caveats were necessary.

Cheers!
Nika
Title: Re: Proper word clock implementation
Post by: danlavry on December 17, 2004, 03:51:15 pm
Nika Aldrich wrote on Fri, 17 December 2004 04:38

bobkatz wrote on Thu, 16 December 2004 21:32

Dan Lavry has been extraordinarily correct technically in all his responses about clocking and how it works....


Bob,

I hope you do not mind me posting a few segments from this archive here....Cheers!
Nika


This is the dialogue that led up to that:

David Josephson:
Bob, you of all people should be careful about ignoring critical
listener reports in favor of supposed theory! A crystal oscillator is no magic device; the crystal is simply a very high Q mechanically tuned circuit that passes signal predominantly at one frequency. Put it in a feedback loop around an amplifier and it oscillates, same as any other tuned circuit. But the bandwidth is not infinitely narrow,
the Q is not infintely high….


I do not have a strong disagreement with Mr. Josephson. I find the way he states matters to be a lot “softer” than mine. I respect his method of leaving a lot of room to insure that nothing wrong is being said.

However, my view of crystals is a very definite.  So I will explain.  David J. said that a crystal a  simply a very high Q mechanically tuned circuit, which may seem to the casual observer as no more then some “piece of  metal” like a tuning fork. In fact, the whole concept of the crystal is its very orderly arrangement of the molecules themselves. It is tuned on a molecular level, and it has 3 axis’s: mechanical, electrical and optical. Think of a 3 dimensions “bee hive” type of molecular organization.

Indeed, it does not have infinite Q, but I would go beyond saying “is simply a very high Q” device.  I would say it is an INCREDIBLY HIGH Q. To quantify it:
A “standard tuned circuit” with Q=100 is very high Q in my book. A crystal Q ranges from 10^4 to 10^7 (from 10000 to 10000000). See the book “Design of Crystal and Other Harmonic Oscillators” by Benjamin Parzen. My point about the fundamental mode oscillators providing less phase noise can be found at the book “Crystal Oscillator Circuits” by Robert Matthys.

The complete equivalent circuit of a crystal is complex, but the simplified version of an inductor, series resistor, series cap and a parallel cap yields a good first approximation.

The equivalent inductor is HUGE. It can be as high as Henries(!). For a reference, 1mH (mili Henry) is a big inductor, lots of core material, and many many windings. 1 Henry is 1000 time bigger. High inductance at our MHz frequencies is a problem for many reasons.
   
The equivalent crystal series capacitor is as low as a few pF. For a reference, if I recall, 2 squares plates 1 square mil each (1/1000 of an inch), separated by 1 mil yields about 1pF.

Can you imagine trying to build a tune circuit with such values from such huge inductor and tiny cap?  The crystal provides us just that.

The Q goes up with increased inductance and decreased capacitance (resistance is the third factor). As an analog designer, I do not see any way to get near such a high Q using anything other than a crystal using inductors and caps. A crystal device really stands out there when compared to all other tuned circuits.

“There was more. I picked what I thought was probably critical to your revision on this topic in your book. Of course, I made the same point in my own book, but I think you and I both allowed for the notion that in some conditions boxes like Big Ben COULD provide better results - or at least one should be open to it and do their own listening tests. This notion is directly contradictory to the point made so emphatically by Dan when he writes, regarding whether or not Big Ben could improve on an internal clock, "No."

You seem to be looking for that a possible scenario where a PLL driven by external clock is better than internal. You will have to find a design with a terrible internal clock and a fantastic PLL. But after you find it you could still not say that the external clock improved the jitter. All you can say is that that particular internal clock sucks.

The jitter you end up with when using the external connection is the INHERIT PLL circuit jitter PLUS more jitter due to the external clock, PLUS the interconnects, PLUS the PLL and the rest. What I call the “INHERIT PLL circuit jitter” is the jitter of the PLL circuit itself with no additional contributions. As if it is driven by zero jitter clock theoreticaly perfect clock source. The best way that I know to test it is to drive it with as low as possible jitter fixed crystal, connected directly to the PLL input...

Using an external clock ADDS to the inherit jitter. It does not reduce the inherit jitter. Driving a PLL circuit with a clock does not reduce it’s jitter. The question is how much does it adds to it. Lets all be clear about it. Many people got the impression from advertisements that Big Ben will do just that, reduce the jitter.

Mr. David J’s suggestion is that there is a possibility, though not likely, that a PLL based circuit will have less jitter than a poorly designed fixed crystal. My question to him is: A PLL requires an oscillator. What kind of oscillator will yield the lowest jitter? An RC and a comparator? An LC with a Q of 100? Or is it a pullable crystal, which is almost as good as a fixed one? (A fixed internal crystal is of course the proposed best way to go). Also, as you must know, it is important to have a low jitter oscillator as a building block for the PLL circuit. The PLL loop filter itself cannot remove all the oscillator jitter.

Dan Lavry


Title: Re: Proper word clock implementation
Post by: danlavry on December 17, 2004, 08:31:36 pm
Nika Aldrich wrote on Fri, 17 December 2004 19:07

Bob,

I understand.  I guess I don't see their reports of what other people say as a marketing department gone amuck.  They aren't claiming that the box does one thing or another.  They are claiming that other people have stated that it does one thing or another....
Cheers!
Nika



Do you really believe what you said? Are you saying that they have nothing to do with statements such as:

"Happily a solution is possible so long as the device in question can synchronize to an external clock." ???

Also guess who said the following on this thread? In was Lucas

“In other words, when you HAVE to clock to an external , jittery, source, you will find that your results will be much better if you have the Big Ben do the clocking first and clock your device to the Big Ben. This is objective, measurable and verifiable.”

First, why not clock to a clean source such as a crystal? If you do, you will not need to buy a box to clean the jitter, and a good crystal oscillator will beat a cleaned jittery clock. Big Ben cannot claim to do better than provide say a word clock 0101 pattern, hopefully with very low jitter.  A lot of words about cleaning jitter, sounding better, adaptive filtering and high end technology do confuse the customer. Heck, they tried it on me and everyone else on this web. And what do they say when it does not make sense?

Lucas also said:
“Then there is the much more difficult to explain phenomenon of converters starting to sound “better” when clocked to the Big Ben as a master clock only; the C777 running on internal. That is something we found empirically and like I said in my previous posting, we researched and optimized.”

What can I say? That statement from an engineer? For some reason it keeps bringing the joke into my head… the punch line was “cross the field where the cows graze and don’t step in the bull….”.

Apogee employees did come here trying to tell people that what I said is false.(ie: about internal clock being better than external clock.) Was it not the reason they came here?

Dan Lavry

Title: Re: Proper word clock implementation
Post by: danlavry on December 17, 2004, 09:15:21 pm
Lucas van der Mee wrote on Fri, 17 December 2004 02:08

Dan ,
You think it is ok to down our products without knowing how they work , spec out and/or perform. You use your platform for pure marketing against us, while you claim in the same response that is not on topic.

The reason why people prefer the Big Ben over Aardvark is very simple. We perform better specs wise, we offer more features and our box is better value.
You have yet to design a product that can beat the Big Ben, so I understand your frustration. But I do not understand your abuse of the PSW platform to release this anger. I consider that to be a sign of very low quality.


Lucas van der Mee,
Sr. Design Engineer
Apogee Electronics




Lucas,
First, I did not intend to restart this argument. In fact, I deleted Nika’s attempt to re-stir the pot. Nika insisted I comment. He complained to the people running the PSN about my deleting his message. Perhaps I should not have replied to his comments, but now that you are here……..well here we go again!

These are 2 of the questions asked previously on  11 6 2004. They are still not answered.

“THE BOTTOM LINE:”

I.  QUESTION:  WHICH SOLUTION PROVIDES LESS JITTER AT THE AD LOCATION?

1.   A REASONABLE FIXED CRYSTAL
2.   BIG BEN DRIVING A CLOCK INTO A CHASSIS EQUIPPED WITH A PLL VIA A 10 FOOT CABLE?

II.  QUESTION: CAN ONE EVER CANCEL OR REDUCE A RANDOM NOISE AT ONE END OF A CABLE WITHOUT KNOWING WHAT KIND OF JITTER ACTIVITY IS OCCURRING AT THE OTHER END?
(can one cancel, remove or reduce the receiver and PLL noise, some of it unknown, some totally random, by driving it with a clean low jitter clock? Driving it with any clock?)


So now you are back once more still without technical input.  You say:

“You think it is ok to down our products without knowing how they work, spec our and/or perform.”

I also know you cannot fly by waving your arms, which is equivalent to the claim that you can have better jitter performance by driving an AD externally.  And yes, lets stay technical:  YOU SAID SPEC OUT:  Thanks for reminding us all. What are the JITTER SPECS?  We are particularly interested in data supporting the claim regarding reduction of the jitter inside the AD when Big Ben is hooked as external source.

“You use your platform for pure marketing against us, while you claim in the same response that is not on topic.”

First, you guys came in here displeased with my general statements that internal clock is a preferred method.  I guess it went against your marketing propaganda, and you tried to intimidate with “your engineering of the 21st century” and the rest of the bull.  You are claiming it was started as pure marketing against you EVEN AFTER BEING SET STRAIGHT by some people on the thread. (Selective memory?)

Second, as you said, I do not make a crock, uh clock, so I am not in competition with you on that product. I feel clean. Do you?

“The reason people prefer the Big Ben over Aardvark is very simple.  We perform better specs wise, we offer more features and our box is better value.”

Hopefully you will desist from spinning “value and features” and talk like an engineer.  Please remember this thread had nothing to do with values and features. The subject we are talking about is JITTER.  

But now that you are here lets get into the engineering. Many people bought into your claims that driving their AD with Big Ben will reduce their AD clock jitter. We are not talking about whether people like what they hear when they use your clock. We are speaking as engineers about actual physical engineering principles that are SOLID. Your claims go against those SOLID ENGINEERING PRINCIPLES.

One of the visitors previously on this threat asked if you were going to offer an official apology to people around the world who believed the misinformation spewed about this product and purchased it.

“You have yet to design a product that can beat the Big Ben, so I understand your frustration.”

My dear boy trying to insult me is silly especially since I was the one who designed the first Apogee “low jitter clock” and created the concept. I may design another one in the future. Thanks for the idea.

“But I do not understand your abuse of the PSW platform to release this anger. I consider that to be a sign of very low quality.”

No technical comments? Only more personal attacks?

  ------------------------------------------------------------ ---------------------------------------------------------
Let us see your “high quality” comments in the form of a technical defense of your design.

We begin with the following questions:

Question:  Does an average AD operating with reasonable internal crystal clock have less jitter when driven by Big Ben? YES OR NO

If you say NO, the group will need to determine what steps your company should take to remedy the wrong that was done.

If you say YES, Then:

LETS HAVE YOUR TECHNICAL COMMENTS ANSWERING MY TECHNICAL COMMENTS (I and II above) (Marketing and personal attacks qualify as “low quality” responses. If you choose to disappear as you did last time without answering, well that is certainly your choice).

One can discuss concepts and principles without divulging what you call your “secrets”.  I can help you along with the following comments in the form of general principles.

EXAMPLE:(Box A) cannot correct for jitter in another (Box B) without knowing what takes place at (Box B) is as universal and general as gravity! This is no secret.

EXAMPLE: An ideal clock with least jitter is based on having all cycles identical, thus there is no room for ANY modulation of the clock. This is another universal and general principle.

I look forward to hearing a professional design engineer’s response from you Lucas.

Dan Lavry


Title: Re: Proper word clock implementation
Post by: bobkatz on December 18, 2004, 08:37:40 am
Level wrote on Thu, 16 December 2004 22:10

That would be a nice one to put to bed Bob. Implementation of the clock, you could give the perspective. I use my internal clock and I just don't get all these clock issues, although I would love to be proven wrong, if my internal is giving problems I do not hear, nor anyone else has noticed to date.


Another way to look at this is "the battle of the marketing against the science", or "can $1800 (retail) of external components really beat $40 of internal"?

For (in a particular test) if an external clock really beats an internal one, then the manufacturer of the converter under test must really have cut some cheap corners, OR at the time of manufacture, his technical chops were not as good as they oughta!It takes a tremendous amount of knowledge of analog, digital, RF and systems design to make a great-performing converter. Which is why my personal list of "great" converters is very small, indeed.

I know a converter designer who is modest and does know what he is doing. He also knows the limitations of his design expertise. So he freely admits that to design a great PLL would take far more time and money and consulting work than he could afford, so instead, he made sure that he had a well-performing internal clock in his converter, and exploit that as an advantage. I thoroughly agree to that philosophy.

I also know another converter designer who built a converter with "no compromise" in mind. It took one man-year to design the clocking circuits, both internal and the superior PLL. How long do you think it will take for them to make up their R&D costs?

Additionally, BOTH of these converter manufacturers have been entirely honest in their marketing statements of the capabilities and limitations of their gear.

BK
Title: Re: Proper word clock implementation
Post by: Nika Aldrich on December 22, 2004, 12:50:18 am
Sorry guys, I've been on vacation - still am.  I'll have to chime in when I get some time.

Cheers!
Nika
Title: Re: Proper word clock implementation
Post by: doug hazelrigg on December 22, 2004, 04:54:59 pm
My I/O gear: Apogee 200 via SPDIF into M-Audio Audiophile 192. My question is: should I always be using the Apogee as the clock source even after A/D conversion and tracking is done, in other words, when mixing? Or should I switch to the card's internal clock. I ask because I had assumed the clock from the Apogee was "better" than whatever the soundcard used

Thanks for allowing me to but in the long and informative thread
Title: Re: Proper word clock implementation
Post by: Eliott James on December 22, 2004, 05:15:20 pm
Please forgive my non technical observation but it does address the financial constraints placed on the designers of more comman converters, and the $ constraints of the recordist. They go hand in hand.

Many smaller studio, and yes, the project guys, don't use the best converters. They can't afford multiple channels of ins and outs of great converters. When you can clock these less expensive converters to a better clock, the whole rig sounds better. I have done this myself clocking a MOTU rig with a Mytek 2 channel AD converter. It does sound better that way. I think that is what Apogee is addressing in their marketing.

I don't think anyone is fooled into thinking the MOTU (or any other lessor brand) rigs of the world have great converters, and you could even say they are "defective" in that their clock and analog stages are not the best, but I think "defective" is a strong word. Lessor quality, yes. Less expensive, yes.

If I could I would have nothing but the best. Buy I can't. Is it wrong for Apogee to offer a better clock to make the whole system sound better? I don't think so. This is real world, not the world of absolute best.

BTW, I don't use and am not affiliated with Apogee, and I've never thought much of their equipment personally, but then I've not heard their stuff lately. Maybe it's improved. I like the Mytek gear. Good stuff.
Title: Re: Proper word clock implementation
Post by: bobkatz on December 22, 2004, 07:08:53 pm
Eliott James wrote on Wed, 22 December 2004 17:15



Many smaller studio, and yes, the project guys, don't use the best converters. They can't afford multiple channels of ins and outs of great converters. When you can clock these less expensive converters to a better clock, the whole rig sounds better. I have done this myself clocking a MOTU rig with a Mytek 2 channel AD converter. It does sound better that way. I think that is what Apogee is addressing in their marketing.




I agree. Though from the claims expressed in these threads, you'd think that the Big Ben can improve any converter on earth. Apogee backed down a hair from that claim by revising their statement in saying that D/A converters with ASRC circuits built in cannot be improved by Big Ben.

We can extrapolate from that, since ASRCs consist of a VERY good DIGITAL PLL coupled with variable low pass filters----that any converter with a real good PLL would not benefit from the Big Ben.

What are the economics of the MOTU converters versus the Big Ben addon, anyway? To me it seems expensive enough to warrant an upgrade to a better set of converters anyway. By the way, I'm not knocking MOTU, I think they offer great value for the money. But at some point you have to look at the economics; would you put a Cadillac engine in a Yugo? How much do you have to spend in bandaids before it would have been better to upgrade the converter in the first place? What are the dollars we're talking about here?
Title: Re: Proper word clock implementation
Post by: Eliott James on December 22, 2004, 09:42:39 pm
Well, 8 in of Mytek is $2500, 8 out is $2500, plus computer firewire interface. You're at $5500 to $6000.

8 in of Lavry Blue is $4300, 8 out is $3850. Plus firewire computer interface. Your're at $8650 to $9250.

MOTU 828 MKIII (8 in and out) is about $950. Add Big Ben at, what, $1400. Firewire interface is on the MOTU.  You're at $2350.

MOTU 828 MKIII is about $950. Clock to Mytek 2 channel AD at $975. Firewire interface is on the MOTU.  You're at $1925.

$1925 for pretty good sound with MOTU/Mytek clock, plus many other options in/out on the MOTU, plus have two extra very good converters with the Mytek.
$2350 for pretty good sound (presumed) with MOTU/Big Ben, plus many other options in/out on the MOTU.
$6000 for very good sound with Mytek.
$9250 for presumed great sound (I haven't heard them personally) for Lavry Blue.

Those extra thousands (around $4075) between the MOTU/Mytek rig and the 8 in/out Mytek setup goes to better mics, mic pres, monitors or whatever the small studio needs that will probably have a greater impact on the final sound they're after than the clarity of the converters, depending of course on the type of music they're recording.

This is the choice many have to make.

*Prices are based on an online retailer. YMMV.
My experience is a MOTU rig clocked to a Mytek 2 channel AD will noticably improve the sound of the MOTU converters, but they will not sound quite as clear and defined as the Mytek converters themselves. Can't vouch for the Big Ben.
I don't use the MOTU unit now, but did at one time.
Title: Re: Proper word clock implementation
Post by: Level on December 22, 2004, 11:38:30 pm
Great point Eliott!

Some cats are just rolling in the dough, have 200K in outboards of different kinds and still will not pay an acoustic engineer a meager sum of 3000.00 to come in and provide proper sound control for the rooms.

It IS a situation of priorities. Throwing dollars at issues has never been the best remedy. Room/Monitors/calibration and clean signal path is the one half. Knowing proper useage and judgement is the other.

I had a 3 million dollar facility I was the Chief engineer of 10 years ago, what I have now smokes it all and little of what I have now is truly from a technological development.
Title: Re: Proper word clock implementation
Post by: danlavry on December 23, 2004, 06:53:46 pm
Eliott James wrote on Thu, 23 December 2004 02:42

Well, 8 in of Mytek is $2500, 8 out is $2500, plus computer firewire interface. You're at...$1925 for pretty good sound with....This is the choice many have to make....



There are many ways to configure your studio, and yes mics are important and so are AD’s and speakers. Different people balance it differently. Personally, I subscribe to the notion of having very good tools, for a carpenter, mechanic, EE, for mastering, recording… There is a lot of range between spending a 1 million dollar for one mixing counsel and a few grand… What should investment priorities be? This is not the place to argue specifics while referring to specific gear and prices.

But I agree with you that one should spend their money wisely. One way to prioritize is to understand the technical facts, not just react to hype, advertisements, specific unit recommendations.

This thread is about clocks and jitter is an important clock issue. So IF you decide you need external clock, what then? Let’s isolate the clock as a STAND ALONE decision.

What do you want from an external clock? Say you need a few frequencies (such as 44.1KHz, 48KH, 88.2KHz and 96KHz) as a steady low jitter source. What is better than a box with 4 (or more) good crystal oscillators and some line drivers (cable drivers?). It is a cheap, good solution based on the right technology, and it is the cheapest way to go.

Big Ben claims some wonderful secret jitter filtering. You do not get to filter a good crystal based clock. You can receive a jittery clock and filter out the jitter, then send it as a clock. Is that what you need? A jitter filtering box, acting as a slave PLL to a poor clock? Or do you need a steady stand alone low jitter clock box. It is an important distinction to make. In most cases, a house sync is a stand alone master clock.

A few people said that the external clock improved their poor AD’s. I am not at all convinced of it, accept in very rare cases. As a rule, MOST of the jitter problem is ahead of the external clock (cable, receiver, PLL…). Say you have a real bad internal clock, say 1nesc (1000psec). But the same designer put in a 50psec random jitter PLL circuit for external lock (including cable, receiver…). If you drive that circuit with external 1psec jitter, the combined jitter outcome is 50.01psec. If your external clock jitter is 10psec, the combined outcome is 50.99psec – less than 1psec difference. If the external clock jitter is 25psec, the outcome is 55.9psec – less than 6psec difference! As long as the external clock is less than 1/2  the internal jitter, you are almost completely limited by the internal PLL, not the external clock.

Do you need to receive a dirty clock and clean it? Or do you need a fixed clock (a few known frequencies). The statement “we filter jitter” is too general. The question is where do you filter it. Filtering it at the INPUT (received jittery clock) does not matter to most house sync applications where you do not need to start with a jittery clock. One way or another, you are looking for say a word clock with repeated 01 cycle pattern where each cycle is identical to each other cycle. That means steady voltages, fast rise and fall times and low jitter. The same holds for AES but with a different pattern. No one is sending secret messages forward. I would not call it a marketing hype if they pointed out clearly that the jitter applies when the box operates as an intermediary PLL, trying to “look like” a stand alone crystal while tracking yet another clock. But they got carried away claiming that their clock as a stand alone box, has some magic ability to impact what is ahead.

I am amazed at how much talk and focus has been given lately to clocks. Yes, you may need an external clock, but it is not the “next step in improving conversion” like so many people were convinced of. Conversion is complex and has many aspects to it. A clock is only one such aspect, and my 50psec PLL jitter example shows that an external clock need only be two or three times better than the PLL, and you are at the point of diminishing returns.

Regards
Dan Lavry
Title: Re: Proper word clock implementation
Post by: bobkatz on December 23, 2004, 10:46:26 pm
Thanks, Elliot, for the dollars to donuts comparison.

I'd spend the root money on great converters first. If your car doesn't have a good motor, it doesn't matter how much premium gas you put in the tank, the performance will suck. I disagree that great mikes and preamps still sound good through mediocre converters. Some people have actually gotten used to that "hazy" sound, but when they hear the transparency improvement with a great converter, it changes their whole day.

A client changed his analog mixdown from direct into a Masterlink using its converters to a high end external converter running at 2496. The sound quality of his mixes went up tremendously. I say, don't scrimp on converters!

BK
Title: Re: Proper word clock implementation
Post by: Eliott James on December 24, 2004, 02:37:33 pm
I agree with you, Bob. If you can find the $, it makes sense to go for the best.

On a whole different level though, it's like you wrote about the Lavry Gold DA. When you've got those $ in your budget.....
Smile
Title: Re: Proper word clock implementation
Post by: Lucas van der Mee on December 27, 2004, 01:40:41 pm
Bob and Dan,

You are both “interpreting” our words and saying that we claim things we don’t. So let’s get it straight one more time.
We do CLAIM:
1.   The Big Ben is an excellent low jitter solution if you need a master clock
2.   If you have to clock to a very jittery source, you’ll get better results most of the time by having the Big Ben cleaning it up first.
We haven’t claimed anything else.

We did REPORT that users feel that their converters sound better when they clock to the Big Ben. These reports are multiple and consistent. That is where you guys become all upset about, because it seems to be against all we know about conversion, etc.
So was I, when I got the first reports, but unlike Dan I do not say: “that cannot be”. I do trust the critical thinking and listening of my customers and investigated the results. The fact is, through our own investigations we have discovered that there is indeed an audible improvement in most cases when using Big Ben as a master clock. We have not CLAIMED that this is due to an external clock lowering the jitter. This is your own presumption and it is wrong.

Bob,
With regards to your comments about ASRC based clocking: In contrast to your claims, SRC based converters usually do NOT have a great PLL. That is the whole purpose of using the SRC. You can use a cheap chip-based PLL on the input side of the SRC and a crystal on the output side which also clocks the DA chip. Simple, but it means that the SRC is ALWAYS in the audio path. So you never get the original data and always get the SRC artifacts at the DA converter chip.

Lucas van der Mee
Sr. Design Engineer
Apogee Electronics
Title: Re: Proper word clock implementation
Post by: bobkatz on December 27, 2004, 07:18:34 pm
Lucas van der Mee wrote on Mon, 27 December 2004 13:40

Bob and Dan,

You are both ?interpreting? our words and saying that we claim things we don?t. So let?s get it straight one more time.
We do CLAIM:
1.   The Big Ben is an excellent low jitter solution if you need a master clock
2.   If you have to clock to a very jittery source, you?ll get better results most of the time by having the Big Ben cleaning it up first.
We haven?t claimed anything else.




Aww gee! You took all the fun out of it, Lucas! Well, regardless, it doesn't matter how much water has gone under the, dam, let's just limit the claims to that and test for that. Words are slippery and I personally believe that Apogee's printed advertisements and some previous statements by Apogee employees in this thread have gone somewhat beyond those claims.

Quote:



Bob,
With regards to your comments about ASRC based clocking: In contrast to your claims, SRC based converters usually do NOT have a great PLL.




Did I claim that? Not to my knowledge. Of course you are right. The analog PLL in an ASRC-based converter can be anything from a shitty to an ordinary to an extraordinary model. Perhaps you got the idea that I was claiming otherwise from my statement describing the ASRC as containing a superb Digital PLL, which of course both the Analog Devices and TI chips contain. However, it is the dual stage reduction of a great front end (analog) PLL combined with the Digital PLL (and good power supply, grounding, and layout, of course!) that makes an ASRC-based DAC sound as good as it does. I have NEVER claimed otherwise, unless someone has misinterpreted my words or taken them out of context.

Quote:



That is the whole purpose of using the SRC. You can use a cheap chip-based PLL on the input side of the SRC and a crystal on the output side which also clocks the DA chip. Simple, but it means that the SRC is ALWAYS in the audio path. So you never get the original data and always get the SRC artifacts at the DA converter chip.




Absolutely, you can't get something for nothing. The AD 1896's predecessor did not have sufficient resolution to be considered "super hi-fi", but things have changed drastically. The distortion products produced by the 1896 and the TI equivalent are orders of magnitude lower and the SRC artifacts are all-but inaudible.

Well-designed ASRC-based D/A converters employing the latest ASRC chips have raised the quality bar so much that it requires a very-expensive, well-made design, in my opinion, to top them sonically.

BK
Title: Re: Proper word clock implementation
Post by: Nika Aldrich on December 31, 2004, 11:45:33 am
Guys,

I've been on vacation for the past 10 days and have intentionally avoided the internet for that time.  I have no idea of all that may have transpired, but have read through the last several posts and am relieved that it seems we're finally coming to some conclusions and agreement on some matters.  Moreso, it appears we are clearing up a lot of miscommunication and opportunity for further miscommunication.  Because it appears that this forum is going to be turned off in the next 24 hours I am not going to go any further into reading everything and trying to respond as that seems like it would be wasted effort.

Thanks to all that participated and thanks to all that worked to encourage cooler heads to prevail.

Nika
Title: Re: Proper word clock implementation
Post by: danlavry on December 31, 2004, 04:24:37 pm
Nika Aldrich wrote on Fri, 31 December 2004 16:45

Guys,
.... it appears we are clearing up a lot of miscommunication and opportunity for further miscommunication.  Because it appears that this forum is going to be turned off in the next 24 hours...Thanks to all that participated and thanks to all that worked to encourage cooler heads to prevail.
Nika


Nika,

I am sorry to see that you are back disseminating misinformation again. Due to an outpouring of an amazing number of intelligent people, I will be a permanent moderator on this site.

Dan Lavry
http://www.lavryengineering.com
“In a time of deceit, telling the truth is a revolutionary act.”
Title: Re: Proper word clock implementation
Post by: bobkatz on December 31, 2004, 06:03:53 pm
danlavry wrote on Fri, 31 December 2004 16:24



Due to an outpouring of an amazing number of intelligent people, I will be a permanent moderator on this site.





Congratulations, Dan! You are an incredible resource. Leshona tova et al.


BK
Title: Re: Proper word clock implementation
Post by: Bob Olhsson on December 31, 2004, 08:06:37 pm
That's great news!
Title: Re: Proper word clock implementation
Post by: jfrigo on January 01, 2005, 03:36:19 am
We're fortunate to have you as a resource here, Dan. Glad you'll be sticking around.
Title: Re: Proper word clock implementation
Post by: Nika Aldrich on January 01, 2005, 03:41:35 am
danlavry wrote on Fri, 31 December 2004 16:24

Nika Aldrich wrote on Fri, 31 December 2004 16:45

Guys,
.... it appears we are clearing up a lot of miscommunication and opportunity for further miscommunication.  Because it appears that this forum is going to be turned off in the next 24 hours...Thanks to all that participated and thanks to all that worked to encourage cooler heads to prevail.
Nika


Nika,

I am sorry to see that you are back disseminating misinformation again. Due to an outpouring of an amazing number of intelligent people, I will be a permanent moderator on this site.

Dan Lavry



Well that just tells you how fast the industry changes!  My 11 day old information is apparently no longer valid after coming back from vacation.  It's great to have you as a resource and as a permanent moderator.  I'll see you in your permanent forum.

Sorry for the misinformation!

Nika
Title: Re: Proper word clock implementation
Post by: danlavry on January 01, 2005, 07:32:23 pm
Lucas van der Mee wrote on Mon, 27 December 2004 18:40

Bob and Dan,

You are both “interpreting” our words...We haven’t claimed anything else...."Lucas van der Mee
Sr. Design Engineer
Apogee Electronics



Lucas,
Once again, no technical comments about the clock, instead you posted a denial.
Regarding your first point:

“l. The Big Ben is an excellent low jitter solution if you need a master clock.”

Your literature makes people think they do need one when they don’t. But aside from that I am reminded of a scene from Monty Python. A knight on the ground with limbs cut off from a battle, lay there still cajoling the other guy present to come and fight.  Here you are after all that has transpired, still trying to sell your clock on a forum dedicated to engineering issues.

Your second statement
“2. If you have to clock to a very jittery source, you’ll get better results most of the time by having the Big Ben cleaning it up first.”

How much is “very”. Also This statement hedges your previous statements.

NEXT:
“We have not CLAIMED that this is due to an external clock lowering the jitter. This is your own presumption and it is wrong.”

If I state that a door is painted white, and you say I am wrong, you are in fact saying the door is NOT painted white without having to say the words. (The door is not painted white.) On Oct, 27, 04 an Apogee rep came here taking issue with my statement that the best method to achieve lower AD jitter is to use a good internal crystal whenever possible instead of an external clock (See my previous postings)

You did, in fact, take a stand that was OPPOSITE of what you are now saying. Regarding your current sentence,

ALSO 2: "We haven’t claimed anything else.”

OK. Yes, “Let’s get it straight”. Apogee is not claiming that Big Ben will improve ADs as a clock source, compared to using AD under internal crystal operation!! It does not claim to clean jitter when used as a master clock.. Only when it resends a “very jittery source.”

AND AGAIN Your back door insult to me. “but unlike Dan I do not say: “that cannot be”. I do trust the critical thinking and listening of my customers and investigated the results.”

What results? Researched It? How investigated? What measurements? Graphs? Published where? We are not about listening tests here. But then again, I previously STATED why you could not publish results. Your claim to impact the sound ahead is an impossibility. I can investigate how I can fly like a bird. A gullible person may confuse investigation with actual results coming from an investigation.

So Lucas, what do you say to all the people that interpreted your “A Cure For the Jitters” coupled with the claims of sonic improvements of ADs when driven by Big Ben as a master clock (You do not claim it any more) as a sure way to enhance conversion (relative to internal clocking?) Do you take no responsibility for it? Do you blame the dealers?

Apogee has done an excellent job of overstating the ability of this piece of gear. Your company’s marketing line, “A Cure for the Jitters” coupled with claims of sonic improvements when the clock is used as a master clock, is at best misleading. Many do not understand the many types of jitter created during recording and transmission, where the problems are and are not and what the remedies are.

It has been my goal on this thread to try and enlighten people about the clocking concept and to raise the level of understanding. You could have help the enlightenment process. Instead we get hype and “leaders with 21st century (secret) technology". And yes, taking a cheap shot by calling me an old 20th century theoretical guy. Well you are way out of your league.

The company you work for has gone a long way with advertising and a reputation made on the quality of my designs now called “legendary and legacy” on the Apogee website. I am the authority on jitter and have been in this business from tubes to ICs. Leave the lame statements home and send the measurement results with a clear description of the test setup.

Dan Lavry
www.lavryengineering.com





Title: Re: Proper word clock implementation
Post by: ammitsboel on January 01, 2005, 09:15:21 pm
bobkatz wrote on Fri, 24 December 2004 03:46

A client changed his analog mixdown from direct into a Masterlink using its converters to a high end external converter running at 2496. The sound quality of his mixes went up tremendously. I say, don't scrimp on converters!

BK


OK... so running 2496 increases sound quality?

Sorry to interrupt, but this is just wrong.


Best regards
Title: Re: Proper word clock implementation
Post by: bobkatz on January 01, 2005, 10:09:05 pm
ammitsboel wrote on Sat, 01 January 2005 21:15

bobkatz wrote on Fri, 24 December 2004 03:46

A client changed his analog mixdown from direct into a Masterlink using its converters to a high end external converter running at 2496. The sound quality of his mixes went up tremendously. I say, don't scrimp on converters!

BK


OK... so running 2496 increases sound quality?

Sorry to interrupt, but this is just wrong.


Best regards




I'm so impressed by your deductive abilities, Henrik. The high quality external converter beat the quality of the internal converter of the Masterlink, at any sample rate, actually.
Title: Re: Proper word clock implementation
Post by: ammitsboel on January 02, 2005, 07:05:27 pm
bobkatz wrote on Sun, 02 January 2005 03:09

ammitsboel wrote on Sat, 01 January 2005 21:15

bobkatz wrote on Fri, 24 December 2004 03:46

A client changed his analog mixdown from direct into a Masterlink using its converters to a high end external converter running at 2496. The sound quality of his mixes went up tremendously. I say, don't scrimp on converters!

BK


OK... so running 2496 increases sound quality?

Sorry to interrupt, but this is just wrong.


Best regards




I'm so impressed by your deductive abilities, Henrik. The high quality external converter beat the quality of the internal converter of the Masterlink, at any sample rate, actually.


OK, now that cleared it all up Smile
Since you have had experience with the Masterlink, do you consider the internal converters better or perhaps equal to the ones in the panasonic DAT machines(when comparing in 16/44.1)?

Let's keep the deduction discussion out of this thread... although I would love to pick back at your deduction scheme some other time Smile

Best Regards  
Title: Re: Proper word clock implementation
Post by: bobkatz on February 20, 2005, 04:35:30 pm
[quote title=ammitsboel wrote on Sun, 02 January 2005 19:05][quote title=bobkatz wrote on Sun, 02 January 2005 03:09]
ammitsboel wrote on Sat, 01 January 2005 21:15

bobkatz wrote on Fri, 24 December 2004 03:46

A client changed his analog mixdown from direct into a Masterlink using its converters to a high end external converter running at 2496. The sound quality of his mixes went up tremendously. I say, don't scrimp on converters!

BK


OK, now that cleared it all up Smile
Since you have had experience with the Masterlink, do you consider the internal converters better or perhaps equal to the ones in the panasonic DAT machines(when comparing in 16/44.1)?




Actually I was pretty impressed by the converters in the Masterlink. A bit grainy, a tetch bright, but not bad at all for an in the box converter. But there are of course better outboard units and when my client, who had been recording 9624 into the Masterlink converters, "converted" to the Benchmark A/D the sound became, well, better!

It's been too many years since i listened to a Pana DAT machine converter. But things have advanced tremendously in the converter ballpark for me to believe the Masterlink is anything but way superior to a Pana converter.

Another "in the box" converter that is way better than its price and price/performance ratio is very high, is the A/D in the Waves L2. Very nicely done. An A external converter, for sure. Leaving only room for the A plus external boxes.
Title: Re: Proper word clock implementation
Post by: rsdio on March 15, 2005, 04:48:54 am
Andy Peters wrote on Thu, 11 November 2004 17:02

Here's another example of how we can decouple clock domains.  It's the typical PCI sound card.  Standard PCI is 32 bits wide and clocked at 33 MHz.  Of course that's much much faster than Red Book audio, and is completely uncorrelated with audio.  

(Note: Contrived example not involving DMA engines!) During playback, the PC takes a buffer full of audio data from its memory and writes it as fast as possible over PCI into a FIFO on the sound card.  The PC then goes off and does something else as the sound card's "engine" reads audio data from the FIFO at a rate dictated by the sampling frequency.  The host must monitor the state of the FIFO (or wait for an interrupt); when it gets "almost empty" it bursts the next batch of data over PCI to the FIFO.

Again, as long as the buffer never completely empties, the playback engine doesn't have any problems, as its clocked by its own oscillator.  The playback engine doesn't even know that it's being fed data at a much higher speed.


Andy, you've overlooked an important difference between your computer PCI example and an outboard D/A: There is feedback in the computer, but not in the outboard D/A device.

In a computer, the disk does not read itself, nor does memory automatically transfer itself.  There is software which monitors the FIFO and accesses disk or memory on demand.  This is called a "pull" model of data flow.  It is easy to decouple the output clock in a "pull" model.

With outboard D/A, driven from a CD transport or any other digital audio source, there is no feedback.  The D/A cannot ask for data if it is starved, and missing even a single sample would be bad.  This is a "push" model.  The digital audio is sent out at the assumed clock rate.  All the D/A can do is lock to the incoming clock rate and attempt to de-jitter, or lock to its own clock and risk FIFO overflow or FIFO underflow.  Only an infinite FIFO size can correctly handle the latter, in the absence of some kind of feedback to the device sending the digital audio in the form of flow control.
Title: Re: Proper word clock implementation
Post by: crm0922 on March 19, 2005, 02:07:42 am
Fascinating discussion...

Did Apogee ever *directly* answer the questions as Dan posed earlier?


Quote:

I. QUESTION: WHICH SOLUTION PROVIDES LESS JITTER AT THE AD LOCATION?

1. A REASONABLE FIXED CRYSTAL
2. BIG BEN DRIVING A CLOCK INTO A CHASSIS EQUIPPED WITH A PLL VIA A 10 FOOT CABLE?

II. QUESTION: CAN ONE EVER CANCEL OR REDUCE A RANDOM NOISE AT ONE END OF A CABLE WITHOUT KNOWING WHAT KIND OF JITTER ACTIVITY IS OCCURRING AT THE OTHER END?
(can one cancel, remove or reduce the receiver and PLL noise, some of it unknown, some totally random, by driving it with a clean low jitter clock? Driving it with any clock?)


I think it is necessary that they clear the air about these questions, as their comments in this thread, as well as those in their marketing materials, make it seem as though the answers would be in direct conflict with scientific fact.

This is, at the very least, a disservice to their customers and to potential customers.  I have a friend who just recently contacted me and suggested he might purchase a Big Ben to "improve" his RME converters.  He must have gotten the idea that the Big Ben would result in improved clocking from somewhere.

I was just hoping to see some more flame throwing, actually.

Just kidding.

Chris
Title: Re: Proper word clock implementation
Post by: jimbo-baby on March 20, 2005, 07:28:27 am
i really feel for dan here...it's extremely moral and brave to defy the 192kHz thing as a converter designer, while it would obviously be easier and probably more profitable to "keep up with the joneses" and compete with the other manufacturers. not only that, but he'd also avoid all the hastles of arguing for hours and hours with people who know alot less than him, not only on that subject, but also on others such as this clocking one, on a forum which i'm sure he intended for technical discussion, and not the kind of bad tv soap style feuding which is going on here. i know people are entitled to their opinion but this is supposed to be a place where opinion is properly backed up. if you can't back up your statements, why argue with someone who can? i would love to see a proper reply from the apogee guy, but i've got the feeling i won't. i'd like to thank dan for being good enough to share his knowledge on this forum, and can only wish that his level of integrity would rub off on other audio designers/companies.

jimbo
Title: Re: Proper word clock implementation
Post by: crm0922 on March 20, 2005, 03:15:06 pm
jimbo-baby wrote on Sun, 20 March 2005 07:28

i really feel for dan here...it's extremely moral and brave to defy the 192kHz thing as a converter designer, while it would obviously be easier and probably more profitable to "keep up with the joneses" and compete with the other manufacturers. not only that, but he'd also avoid all the hastles of arguing for hours and hours with people who know alot less than him, not only on that subject, but also on others such as this clocking one, on a forum which i'm sure he intended for technical discussion, and not the kind of bad tv soap style feuding which is going on here. i know people are entitled to their opinion but this is supposed to be a place where opinion is properly backed up. if you can't back up your statements, why argue with someone who can? i would love to see a proper reply from the apogee guy, but i've got the feeling i won't. i'd like to thank dan for being good enough to share his knowledge on this forum, and can only wish that his level of integrity would rub off on other audio designers/companies.

jimbo


Agreed.  It seems that apogee spent a lot of time complaining about Dan's rhetoric and little time defending their claims here and those in their advertisements.

One of their ads has an engineer claiming the first thing you should do when you buy a DAW is add a Big Ben to improve the sound.  If this is technically impossibile, if jitter is the cause of clock-related bad sound, then that certainly seems like borderline false advertising.

I would like to see this defended if Apogee is ever to be considered a serious pro audio innovator.

Chris
Title: Re: Proper word clock implementation
Post by: Touchwood Studios on March 21, 2005, 10:53:27 am

I use an external clock in my studio for clock all my different pieces together. The only differnce in what I hear is that there are no random clicks and pops (bad clocking).
Title: Re: Proper word clock implementation
Post by: Max on March 21, 2005, 01:06:35 pm
crm0922 wrote on Sun, 20 March 2005 20:15

jimbo-baby wrote on Sun, 20 March 2005 07:28

i really feel for dan here...it's extremely moral and brave to defy the 192kHz thing as a converter designer, while it would obviously be easier and probably more profitable to "keep up with the joneses" and compete with the other manufacturers. not only that, but he'd also avoid all the hastles of arguing for hours and hours with people who know alot less than him, not only on that subject, but also on others such as this clocking one, on a forum which i'm sure he intended for technical discussion, and not the kind of bad tv soap style feuding which is going on here. i know people are entitled to their opinion but this is supposed to be a place where opinion is properly backed up. if you can't back up your statements, why argue with someone who can? i would love to see a proper reply from the apogee guy, but i've got the feeling i won't. i'd like to thank dan for being good enough to share his knowledge on this forum, and can only wish that his level of integrity would rub off on other audio designers/companies.

jimbo


Agreed.  It seems that apogee spent a lot of time complaining about Dan's rhetoric and little time defending their claims here and those in their advertisements.

One of their ads has an engineer claiming the first thing you should do when you buy a DAW is add a Big Ben to improve the sound.  If this is technically impossibile, if jitter is the cause of clock-related bad sound, then that certainly seems like borderline false advertising.

I would like to see this defended if Apogee is ever to be considered a serious pro audio innovator.

Chris



If you read through the entire thread from the beginning, you will understand that this all began with some remarks about Apogee by Bob Katz in reference to some of Dan's statements. It was only after these initial disparaging remarks toward Apogee were made that we felt compelled to respond.  Our position on this has always been based on results and data compiled through listener experience and we question the rigidity of the dogma espoused here based on said experience. Dan had deleted a number of our responses in this regard, so we have opted not to continue this discussion in his forum. If you really do care about this issue and not the rhetoric being thrown around by a couple of folks with absolutely no experience on the product, do the listening.

As for the ad, understand that these are unpaid endorsements in the user's own words. Obviously, his and many other engineer's experience differs from some of the opinions on this forum, hence our comments in this very thread.

Regarding serious pro audio innovation, Apogee's record speaks for itself, and I would suggest doing a bit of research on the subject before making such statements.
Title: Re: Proper word clock implementation
Post by: PookyNMR on March 21, 2005, 02:34:50 pm
Max wrote on Mon, 21 March 2005 11:06



Our position on this has always been based on results and data compiled through listener experience and we question the rigidity of the dogma espoused here based on said experience. Dan had deleted a number of our responses in this regard, so we have opted not to continue this discussion in his forum. If you really do care about this issue and not the rhetoric being thrown around by a couple of folks with absolutely no experience on the product, do the listening.


I think the argument was that listening tests are far too subjective.  Plus, Dan wants this forum do be about data and facts (as per his statement at the top of the forum), not tainted by the subjectivity of listening tests, which in the end tells little more than the individuals perferences.  

What was wanted was hard data / measurements.  None were provided.  All that was provided was "listen for yourself."  Sounds to many like a cop-out, trying to hide bad numbers.
Title: Re: Proper word clock implementation
Post by: howlback on March 21, 2005, 03:27:00 pm
PookyNMR wrote on Mon, 21 March 2005 14:34

I think the argument was that listening tests are far too subjective...
which in the end tells little more than the individuals perferences.  


Listening tests are subjective, but may provide much more information than just preference.   Depending on the testing and analysis, subjective testing may render hard data about perception.

I don't know how this applies to this word clock thread, but perceptual data can be just as "hard" as data from physical measurements.

 
Title: Re: Proper word clock implementation
Post by: Max on March 21, 2005, 04:16:29 pm
PookyNMR wrote on Mon, 21 March 2005 19:34

I think the argument was that listening tests are far too subjective.  Plus, Dan wants this forum do be about data and facts (as per his statement at the top of the forum), not tainted by the subjectivity of listening tests, which in the end tells little more than the individuals perferences.  

What was wanted was hard data / measurements.  None were provided.  All that was provided was "listen for yourself."  Sounds to many like a cop-out, trying to hide bad numbers.


Again, please read the thread (and Nathan, try and read it without the fervent anti-Apogee bias that pervades all of your posts, on this forum and others).

The assertion was that if an external clock can improve the sound quality of a single converter, there must be something wrong with the converter's clock design.

I quote:

If you hear differences when you change clocks "controlling" your converters, then you have a defective converter design! A well-designed converter should contain internal phase locked loops whose performance reduces any incoming jitter artifacts to inaudibility. An external clock is a bandaid for a "cure" which can only be done properly within a good converter design. In fact, any converter which does not perform equally as good or BETTER on internal clock than external is also defective.

BK


Followed by Dan's comment:

Yes indeed! Well said.


We dispute that, based on the only thing that really matters, how it sounds. The fact that many well thought of converters in the field are improved sonically and perceived to be more accurate with the addition of Big Ben in the opinions of their users over and over again is evidence enough to dispute the assertion as being factual. Measurements and numbers have nothing to do with this.

So again, the accusation is that somehow any converter that benefits sonically with Big Ben as the clock source must be poorly designed. We said that means a heck of a lot of converters are poorly designed, including converters other than Apogees that many claim to be great. In fact, much of what Bob Katz says later proves my point.

The AD 8000 is about the only Apogee converter I'm "comfortable" with in terms of its sound quality (personal opinion). I tested it on internal and external clock with a good clock, for jitter artifacts, and it appears to perform about the same either way, which is a good sign.

There is no doubt among any listener that has used the AD-8000 with Big Ben that Big Ben improves the sound quality of the AD-8000 significantly. Additionally, our current hybrid PLL design (used in the Rosetta 800 and Rosetta 200) is greatly improved over the PLL in the AD-8000, which we consider to be an average clock design by today's standards. So based on this, One has to consider the possibility that an external clock can improve the sound quality of a converter with a good internal PLL design. Finally, the Rosetta 800 and Rosetta 200 are very high quality clock sources in and of themselves, much better than the AD-8000 and as good or better than any other internal PLL on the market save for the C777, yet even they are improved sonically with Big Ben as the master.

None of this has to do with how things get measured, it has to do with sound. If that is "out of bounds" or "too subjective" for this forum, than the original assertion should have been out of bounds as well, for it referenced "hearing differences" i.e. "sound quality", not mathematics.
Title: Re: Proper word clock implementation
Post by: chap on March 21, 2005, 04:18:34 pm
I have great respect for engineers.  They are a unique, fanatical
breed and make the gear that many of us use for a living.
That said, put me in a room with any engineer and 40 tracks of music. and I will emerge with music. I can't speak for the engineer.
 My point is that listening does count.  If it does not, you have failed your mission.
I get paid to listen to music (and make it) 5-7 days a week.
I wouldn't pretend to have a deep understanding of every piece of gear but I'm happy to use it every day.  I can't build a guitar but I get paid to play them.  I would never buy one without listening to it first.  I am one of the people who has reported positive results to Apogee (unsolicited and certainly not paid for).
I have used all of the converters mentioned here and like the X series because they work and sound right to my ear.
My TC6000 sounds better clocked to Big Ben and it already sounded good.  To be dismissive of the importance of listening is like a paint maker being dismissive towards painters.

It would seem as if this thread has turned into a schoolyard
'dozens' session.  Just keep making us good gear.
peace
chap
Title: Re: Proper word clock implementation
Post by: danlavry on March 21, 2005, 08:03:07 pm
Max wrote on Mon, 21 March 2005 18:06

crm0922 wrote on Sun, 20 March 2005 20:15

jimbo-baby wrote on Sun, 20 March 2005 07:28



i would love to see a proper reply from the apogee guy, but i've got the feeling i won't. i'd like to thank dan for being good enough to share his knowledge on this forum, and can only wish that his level of integrity would rub off on other audio designers/companies.

jimbo


Agreed.  It seems that apogee spent a lot of time complaining about Dan's rhetoric and little time defending their claims here and those in their advertisements.

I would like to see this defended if Apogee is ever to be considered a serious pro audio innovator.

Chris



If you read through the entire thread from the beginning, you will understand...

Max




Max,

I am sure many of the visitors to this forum view the "Daily Show" hosted by John Stewart. One of the recent guests was Harry Frankfurt, a Princeton professor and the author of the book “On Bullshit”.

I bought the book and read it. It contrasts bullshit with lying. ..." for the bullshitter, he is neither on the side of the true nor on the side of the false. His eye is not on the facts at all as the eyes of the honest man and of the liar are, except insofar as facts may be pertinent to his interest in getting away with what he says."... The liar knows that what he is saying in not true.

You said: “Dan had deleted a number of our responses in this regard, so we have opted not to continue this discussion in his forum”.

I deleted some of your repetitive salesman, non-technical responses but none of Lucas's. In fact, while you said “we have opted not to continue this discussion in his forum” Lucas did respond and I have the statement from his last response below.

On Monday Dec 27 2004, an Apogee Sr. Design Engineer Lucas said on this thread:

“Bob and Dan,

We do CLAIM:
1. The Big Ben is an excellent low jitter solution if you need a master clock
2. If you have to clock to a very jittery source, you’ll get better results most of the time by having the Big Ben cleaning it up first.
We haven’t claimed anything else.”


The Apogee guys did not answer any of the 3 questions below (and other questions). Instead they made the above 2 claims.

My unanswered questions are:

I. QUESTION: WHICH SOLUTION PROVIDES LESS JITTER AT THE AD LOCATION?
A. A REASONABLE FIXED CRYSTAL
B. BIG BEN DRIVING A CLOCK INTO A CHASSIS EQUIPPED WITH A PLL VIA A 10 FOOT CABLE?

II. QUESTION: CAN ONE EVER CANCEL OR REDUCE A RANDOM NOISE AT ONE END OF A CABLE WITHOUT KNOWING WHAT KIND OF JITTER ACTIVITY IS OCCURRING AT THE OTHER END?
(Or stated another say: Can one cancel, remove or reduce the receiver and PLL noise, some of it unknown, some totally random, by driving it with a clean low jitter clock? Driving it with any clock?)

III. QUESTION: CAN ONE MODULATE A LOW JITTER CLOCK WITHOUT ADDING JITTER?
A low jitter clock requires the utmost in Word Clock cycle to cycle REPEATABILITY . (Alternatively, for AES signal it requires the frame pattern to be repeatable). That requirement disables and negates any possible signal modulation. With no modulation one can not send any additional information to change the behavior of the “box” ahead (such as an AD).

The Big Ben connection to the AD is one directional. It sends a clock yet receives no signal back. Therefore, the Big Ben would have to figure out what is happening in a different box “all by itself” without any signals communicating that information. Once you overcome that impossibility, the Big Ben would have to figure out (among other things) how to send information to the AD to alter its behavior. That too is an impossibility, unless you introduce modulation which is jitter.

So you do not know what message to send, but you send it anyway by means of “not sending a message”.


My technical questions trump any claim to secret or proprietary methods for achieving the impossible. My questions exemplify solid and fundamental electronic engineering and physics principles. It was not surprising that Lucas and you completely backed away from your PREVIOUS posture, which was initially posited as Apogee’s secret technology of the 21st century against the old engineering of the last century.

You are now saying "Our position on this has always been based on results and data compiled through listener experience…”

What happened to your statement: “until we developed technology that would make a difference”. What happened to what Lucas said: “Just like you, I was convinced that internal clocking was the best thing to do. But here is what I found…. So yes we are very proud of our work and can imagine you don’t like to see someone succeed in something you thought was impossible.”

This goes beyond bullshit.

Lucas, the current Apogee designer could not show technically that his gear would do what Apogee claimed because in fact the claims were impossible.  

What happened after Lucas presented his very watered down 2 claims?
Your advertisements continue to espouse subjective listening tests.

Does this advertising practice CONTRADICTS the statement “We haven’t claimed anything else?”
After saying that you do not claim  Big Ben will improve AD performance, Apogee advertising material pushes that same point using customer's testimonials  to say it for you.
Suppose I want to sell you snake oil, but do not want  to defend it (I know I cannot). I go and find a few people that believe the powers of my snake oil, and I provide advertising channels for them to say so… Is that OK? Audio has a lot of “snake oil”, because there is such a lack of accountability.

The Dan Lavry designs (AD500, DA1000, UV22 with Jerry) that Apogee manufactured and marketed could stand on their own without bullshit.

Dan Lavry
Lavry Engineering, Inc.
"In a time of deceit telling the truth is a revolutionary act."


Title: Re: Proper word clock implementation
Post by: PookyNMR on March 21, 2005, 10:53:29 pm
Max wrote on Mon, 21 March 2005 14:16


The assertion was that if an external clock can improve the sound quality of a single converter, there must be something wrong with the converter's clock design.

<snip>

We dispute that, based on the only thing that really matters, how it sounds. The fact that many well thought of converters in the field are improved sonically and perceived to be more accurate with the addition of Big Ben in the opinions of their users over and over again is evidence enough to dispute the assertion as being factual. Measurements and numbers have nothing to do with this.

None of this has to do with how things get measured, it has to do with sound. If that is "out of bounds" or "too subjective" for this forum, than the original assertion should have been out of bounds as well, for it referenced "hearing differences" i.e. "sound quality", not mathematics.


Please, please explain to me how a difference can be heard but not measured?  If there is a difference (an improvement), why could it not be measured?  Help me understand.

Max wrote on Mon, 21 March 2005 14:16

Again, please read the thread (and Nathan, try and read it without the fervent anti-Apogee bias that pervades all of your posts, on this forum and others).


I have more or less related my experiences with your equipment.  Forgive me if my emotions of disapointment tainted my posts.  I have had a number of equipment failures (in gear that I demoed for purchase and in another facility).  And, ironically, I have also had your units not do so well in listening test with other units.

I will, however, from now on give you guys the benefit of the doubt because Fletcher gave his word on his experience of your QC.

I appologize if I've been an ass to you or your company.  Can we still be friends??

[edited]
Title: Re: Proper word clock implementation
Post by: crm0922 on March 21, 2005, 11:45:28 pm
Max wrote on Mon, 21 March 2005 13:06


If you read through the entire thread from the beginning, you will understand that this all began with some remarks about Apogee by Bob Katz in reference to some of Dan's statements.

...




Hi Max.  I don't care about the origin of all the comments made, and yes, I did read the entire thread.  What I asked for was an answer to the two important questions Dan asked (again, on this page in fact).  I am a consumer, and I deserve those answers, even if you don't think Dan does.

Quote:


As for the ad, understand that these are unpaid endorsements in the user's own words. Obviously, his and many other engineer's experience differs from some of the opinions on this forum, hence our comments in this very thread.



I'm note sure it matters whether they are paid or unpaid endorsements.  Publishing these statements is an endorsement in and of itself of the engineer's opinions by Apogee.

Quote:


Regarding serious pro audio innovation, Apogee's record speaks for itself, and I would suggest doing a bit of research on the subject before making such statements.


I guess I asked for this (re: flamethrowing), but do you honestly think it is a good idea to insult your customers?  I am not asking for anything other than the answers to Dan's short and simple questions, only from a customer/consumer point of view.

Regarding "serious innovation", a designer of pretty highly regarded converters has questioned your clock design in Big Ben and the higher end converters in your line.  He seems to view much of the hubub about Big Ben as marketing directed towards people that are stuck on the idea that an external clock will "clean up" their audio.

Why should I believe that your company is, *today*, making products that are serious innovations?  Because you've sold more converters than your competition?  

You don't know who I am, and, as far as you are concerned, I may own a couple full racks of Apogee gear and want to know if I've wasted my money, and where.  And don't tell me to use my ears, I can't objectively compare your converter designs with *every* other converter in the world, with and without Big Ben, etc.  It is too time consuming, and not often very revealing.  We almost always have to pick from less than 5 choices with any type of gear, so some of us may want some answers on a technical level before we buy (or sell) some Apogee equipment.

Is everyone enjoying these canned worms I've let back on the loose again?  I really didn't want to start another flame war.  I think Apogee can answer Dan's question without giving away their nuclear secrets, I just hope they come around and actually do it.

Chris
Title: Re: Proper word clock implementation
Post by: bananahill on March 22, 2005, 08:07:15 am
The big argument here seems to be numbers vs. listening.   I realize this forum is supposed to be about data, but come on, the final result has to be listening or else the numbers are meaningless.  Music is a subjective experience and we are using the converters to record music ( and/or sound effects, which also should have clarity, depth, definition, etc.)

Something stated earlier about premium gas in a bad engine...
If the car doesn't get you anywhere, the car has no value.
The numbers can be outstanding, yet the sound may not.
The numbers can be less than ideal, yet sound great.
Does anyone here deny that a Neve 1073 is a thing of beauty?
Does anyone hear claim that same 1073 is numerically transparent?

How about a double blind listening test with the above parties present?
compare converters...compare Big Ben in or out of the system.

how about a jazz quintet?  drums, bass, piano, trumpet, sax.
all acoustic, all complex instruments.

I will offer my engineering services and analog gear anywhere in Los Angeles.
I will do my best to enlist world class musicians to perform.
I will offer my Pro Tools system.  Each manufacturer will supply converters and/or clock.
I will encourage other engineers to be present and verify impartiality.

Put your money where your mouth is.

How about it?  any takers?

Results to be posted on this site.  Perhaps an article in Mix or EQ?

Since the term Bullshit has already been brought into this forum, I will continue its use.  
Let's cut the bullshit and do some listening.
Without listening, it's all wanking about the tree in the forest and Schroedinger's cat.
Title: Re: Proper word clock implementation
Post by: crm0922 on March 22, 2005, 09:17:20 am
bananahill wrote on Tue, 22 March 2005 08:07

The big argument here seems to be numbers vs. listening.   I realize this forum is supposed to be about data, but come on, the final result has to be listening or else the numbers are meaningless.  


I will channel Dan for a moment and point out that there are many other forums about listening, this one is for technical discussions.  Dan has made the case that there is no technical possibility that a converter clocked externally will experience less jitter in the incoming clock stream than one clocked internally (to a decently implemented internal clock).

Apogee has asserted many things, but the statement I think some of us are looking for, which mixes the technical realities with some people's perception of "better audio" is something like this:

"A converter clocked to an external Big Ben unit experiences more jitter than it would clocked internally (given a reasonable internal clock) but sounds better to some for unknown and unquantifiable reasons."

The next logical conclusion is, of course, that more jitter results in better sound, or at least does not result in poorer sound.

Hmmm.

Chris

PS - I do not have an anti-Apogee bias at all, but this discussion has made me think about a few things, and I'd really like to see this followed up, preferably civilly.  Smile

Title: Re: Proper word clock implementation
Post by: crm0922 on March 22, 2005, 09:27:20 am
Max wrote on Mon, 21 March 2005 16:16


We dispute that, based on the only thing that really matters, how it sounds. The fact that many well thought of converters in the field are improved sonically and perceived to be more accurate with the addition of Big Ben in the opinions of their users over and over again is evidence enough to dispute the assertion as being factual. Measurements and numbers have nothing to do with this.



I am trying to be civil, but this type of thing makes me go bananas.  I refuse to accept that the laws of physics do not exist, as do many on this forum.

You must concede, Max, that externally clocked converters experience more jitter at the clock input of the attached converter chip.  To deny this suggests that somehow the external converter is able to counteract the jitter introduced by cabling, line drivers, and PLL circuits that are hard wired into the path.

If your product is able to alter the performance of a PLL operating in another chassis, please let us know this, and you needn't explain how, but this particular feat would be quite impressive.

Otherwise, reasonable people are still left to assume that your statements about "better sound" are pure sales-talk.

Thanks, Max, for continuing to respond to the posts in this thread.

Chris
Title: Re: Proper word clock implementation
Post by: Joe Crawford on March 22, 2005, 09:51:27 am
You must look at clock jitter as noting other than “dither in the time domain”… We even shape it to get better low level resolution and bring out the finer details in the audio.  Our Studio Master Clock, through the use of “state-of-the art random noise insertion” (SOA-RNI) significantly improves high frequency presence and detail by adding just the right amount of jitter, exactly where it is needed to be most effective.

Joe Crawford (Do I get that job in marketing now?)
Stony Mountain Studio
Shanks, WV  26761
Title: Re: Proper word clock implementation
Post by: Joe Crawford on March 22, 2005, 09:56:40 am
I forgot to add...  It's only $6,437.27 per port.

Joe Crawford
Title: Re: Proper word clock implementation
Post by: bananahill on March 22, 2005, 10:31:41 am
OK, no listening, just theory...

The cat...
Big Ben (or any external clock) cannot work.  It does not make a system or converter sound better.

The box...
double blind listening test.

put the cat into the box and send it into space... the internet.

when does this cat die?

As long as you never listen, you can believe the cat is still alive.

However if you open the box...


If you don't want to listen, then there is no point to any of this.

I understood the point of this forum is to reach a deeper understanding of how this stuff works.  The theory is all great in the design stage. Once the product is made, you must use it.  To use it, you must listen to the conversion.  
If all you want to do is look at waveforms on a scope, then do not sell something as a musical product!  Sell it as test equipment!

The point of a deeper understanding of our technology should lead to the end target...music sounds better!

Shouldn't a technical discussion include the FUNCTION of the equipment in question?
Title: Re: Proper word clock implementation
Post by: chap on March 22, 2005, 12:10:47 pm
 I Can't believe I'm taking time out of a busy day for this but since the bullshit is flying........
Dan,

Having re-read this entire thread, I do not see the relevance of your "technical questions" to what Apogee is saying. I also do not agree that the message has in any way been watered down. Seems to me you are using a very selective process when drawing conclusions regarding what Apogee is saying. For example, just because their position is based on listener experience, does not impeach the comments about the technology in Big Ben, rather, it supports them.

I think the real bullshit here is trying to have a theoretical argument about something you have no experience with. I and all my friends, golden eared engineers with thousands of hours spent doing this for a living, with award after award hear exactly what Apogee is claiming. Do you think Apogee's "bullshit" is so powerful that the likes of Sterling Sound, myself (3 Emmy's this year, 3 Billboard charting CDs), Manny Marroquin (2 grammys this year)  and the hundreds of other high profile engineers with Big Ben have been hypnotized into believing in a placebo? This is our livelihood! we use what makes us better at our jobs and Big Ben makes the process of recording and mastering better in the ways Apogee described. I understand this does not jive with your personal bias against Apogee (yes, it's that obvious) but you are dong the audio community on this forum a disservice by not being open to the possibility that what we hear is what we hear and not some group hypnosis caused by good marketing.

As for the AD500 and DA1000...they were very good in their day, but nothing compared to even the Mini-Me today, which I understand you had nothing to do with. Those of us that buy Apogee products do so because they sound the best, not because they are marketed the best. My AD8000SE's (I think you approved of them) were saved by Big Ben.  I have used your products and have a great repect for you as a designer but you are not the only designer and you have shown a personal disrespect towoards guys like Lucas who are peers and equals. It only detracts from your message.  There is no snake oil here, just solid, forward thinking engineering. Should this thread be retitled to 'Apogee Bash"?
chap
Title: Re: Proper word clock implementation
Post by: danlavry on March 22, 2005, 12:59:49 pm
chap wrote on Tue, 22 March 2005 17:10

 

Dan,
Having re-read this entire thread, I do not see the relevance of your "technical questions" to what Apogee is saying...

....There is no snake oil here, just solid, forward thinking engineering.

chap


Chap,

Many of your messages in various chat rooms on ProSound web show your commitment to Apogee gear. You are indeed entitled to your opinion. We would like to have some technical input from you to backup your opinion. We are about technological depth here. Your last comment, similar to Max’s seems to be an attempt to marginalize technological discussion in favor of opinion.

It is not my fault that Apogee's unhappy customers come to my site to vent out their frustrations.  I and others have been accused of having an anti Apogee sentiment. The fact is: I have done my best to be fair. After Lucas dropped his previous claims and stated indirectly that they do not claim their clock will improve an AD, I stopped talking about their clock.

I wonder how long and how far you are willing to stand your ground.

In this forum I have a thread about cables. In there are a lot of unbelievable claims about cables, ranging from “magic numbers” to “paralleling a 12AWG cable for the bass and a 14AWG for the highs”.  All have some things in common: They call themselves innovative, and they claim to yield the best sound some post comments from satisfied customers, or rave reviews from a magazine. Of course we all know that repeating how innovative and creative a product is does not make it so.

Example 1:
On the cable thread, there is a guy who claims to have invented a “quantum purifier” that grabs the electrons and makes them go at extra speed, to shake the noise out….
Should anyone even bother to try and listen to his cable? As far as I am concerned, a transaction where one guy sells a Quantum purifier cable to another should result with the buyer in the crazy house and the seller behind bars.  

Example 2:
A MARKETTING COMPANY “X” sells cable that they claim are: ULTRA LOW JITTER CABLE and TEMPERATURE COMPENSATED CABLE. I do not know what temperature compensated cable means but I KNOW that THE CONCEPT OF “ULTRA LOW JITTER CABLE” is not true. Cables don’t have jitter . A cable is a PASSIVE device. It is just a “piece of material”. The level of jitter is determined by the electrical signal driving the cable! There may be other factors (variable load, induced noise and more). One thing for sure: but there is no such a thing as ultra low jitter cable.

“Jitter to cable” is like a “laptop to Moses”
“Jitter to cable” is like a “Pistachio nut to a whale”
“Ultra Low jitter cable” is like “super high speed tree”
“Ultra Low jitter cable” is like “butter to a headphone jack”

Guess who the marketing company X is? Apogee with their “ULTRA LOW JITTER CABLE”.

Folks, we have the double whammy solution. A clock box that was touted as “the cure for the jitters” (which is not being claimed explicitly any longer, only indirectly) and the “ultra low jitter cable”.

At this point, I am adding question # 4 and 5 to Apogee:

4. What the heck is TEMPERATURE COMPENSATED CABLE?


5. HOW DOES A “ULTRA LOW JITTER CABLE WORK”?

a. WHEN YOU DRIVE A ULRAT LOW JITTER CABLE WITH A JITTERY SIGNAL, DOES THE CABLE SUCK THE JITTER OUT?

b. IF THE CABLE IS REALLY LONG, DO YOU GET NEGATIVE JITTER?

c. Is the principle of operation based on quantum jitter suckers, or is it also a top secret of the 21st century that a guy like me does not have experience in.

Please know that in future that statements about personal opinion and marketing will be deleted unless backed up with technical input.

Dan Lavry
Lavry Engineering, Inc.
Title: Re: Proper word clock implementation
Post by: Max on March 22, 2005, 03:06:30 pm
http://recforums.prosoundweb.com/index.php/t/4432/2217/?SQ=c 628846a51673fce83baa8317a10a169
Title: Re: Proper word clock implementation
Post by: crm0922 on March 22, 2005, 03:20:09 pm
chap wrote on Tue, 22 March 2005 12:10


My AD8000SE's (I think you approved of them) were saved by Big Ben.  

...

chap


Everyone seems to be skirting this question, so I'll just keep asking it.

If your AD8000's were "saved" by Big Ben, can you make the statement that added jitter improves the sound of audio?  This is the only statement that makes sense regarding both the laws of physics and what you are hearing.

Chris
Title: Re: Proper word clock implementation
Post by: PookyNMR on March 22, 2005, 04:54:07 pm
I do not doubt anyone that there may be a difference by adding an external clock.  Listening preferences aside, I would just like to know exactly what that difference is.

I still have a question that I can't seem to get an answer for.

If there is a perceivable audible difference why could it not be measured and quantified?
If there is a difference in the jitter of the system, why could it not be measured?

I am confused why the lack of technical data.  While there are some who don't seem to care about technical data, there are others who do.

Peace.
Title: Re: Proper word clock implementation
Post by: Joe Crawford on March 22, 2005, 05:09:57 pm
OK, this thread has finally gotten my dander up.  This is why I went fishin’ a few years back….    

Max – you are a true believer… and in your position you have to be.  I respect you for it.  Your company has a good line of excellent (or almost excellent, this is where it gets subjective) products.  And, if you’re like every other high tech company out there, in order to market them, and meet the next quarter’s sales projections, I would imagine your marketing department sometimes stretches the truth, maybe just a little bit, as far as factual engineering is concerned.  They probably don’t even know it when they do.  This is totally normal in the audio industry as well as most other bleeding edge technology based industries.  It has been a generation or two since sales, marketing, and upper level management in high tech companies were staffed with engineers, rather than specialist in sales, marketing, and management (HP got away with it for years until H & P both retired).  Just like today’s political parties, every one spins it a little, no one tell the consumer the factual truth anymore.  We engineer's just have a problem with spin.

But, if you eventually get curious, everyone calms down, and, I guess, we all recover from this thread, try something.  Take you oldest (and most regarded) development engineer (if you still have one other that new grads, and have managed to keep a lead for a few years, and not the engineering manager, he’s probably corrupted already) out for few drinks some night after work and just have an off-the-record talk with him.  Ask him his actual opinion of the direction the product line is taking and what he thinks about the current products.  If you really get frisky, ask him a few of the questions brought up on this thread. Even run your current marketing liturature buy him.  You might be surprised.  Of course, if you bully him, you will get just the answers you want.  Like everyone else, he probably values his job.

It’s very hard to develop a corporate culture that permits, even encourages, free speech between sales/marketing and engineering.  Their “prime directives” are totally different.  I would guess that out of the dozen or more high tech companies I worked for in the last 40 years, only one or two has had a good rapport between those groups. At least from my point of view, they were the best companies ever I worked for.

Well, so much for this soap-box…

Joe Crawford
Title: Re: Proper word clock implementation
Post by: danlavry on March 22, 2005, 05:28:31 pm
Max,

AGAIN for the record:
On Monday Dec 27 2004, an Apogee Sr. Design Engineer Lucas said on this thread:
We do CLAIM:
1. The Big Ben is an excellent low jitter solution if you need a master clock
2. If you have to clock to a very jittery source, you’ll get better results most of the time by having the Big Ben cleaning it up first.
We haven’t claimed anything else.”



My clear technical comments showing that improved performance is technically impossible have not been answered!

Now to the “low jitter cable issue”

You said:

Finally, let it be clear to the end user that you are deceiving them with another semantic argument when making reference to our cables. Of course cables don't lower jitter, but cables do create jitter, some more than others. Apogee's Wyde-Eye cable creates less jitter artifacts than other cables, hence the phrase "ultra low jitter". We did not call it the "jitter free" cable or even the "jitter reducer", so this blatant attack is a deception.

The blatant deception is selling Ultra low jitter cable. Cables have a lot of specs ranging for bandwidth to resistance to capacitance and much more. But CABLES DO NOT HAVE A JITTER SPECS.

Yes you did not call it "jitter free" or   "jitter reducer". You also did not call it a “rose garden” or a “frying pan”. But you called it an “Ultra low jitter cable”  In trying to regroup you come up with a “rational” that  “cables do create jitter”. That is also nonsense. Covering up BS with more BS.  To my knowledge Apogee is the only one that claims that cables have jitter specs.

Yes I know, you cannot argue about “Ultra Low jitter cable”. No one can. You don’t have any measurements to show, just the regular SNAKE OIL blended with personal attacks, followed by claims of hearing it better.

ULTRA LOW JITTER CABLE IS SNAKE OIL. No way around it! Oops, you forgot to answer the other question about the "Temperature Controlled Cable".

Dan Lavry
Lavry Engineering, Inc.
Title: Re: Proper word clock implementation
Post by: crm0922 on March 22, 2005, 08:31:59 pm
danlavry wrote on Tue, 22 March 2005 17:28

Max,

Yes I know, you cannot argue about “Ultra Low jitter cable”. No one can. You don’t have any measurements to show, just the regular SNAKE OIL blended with personal attacks, followed by claims of hearing it better.

ULTRA LOW JITTER CABLE IS SNAKE OIL. No way around it! Oops, you forgot to answer the other question about the "Temperature Controlled Cable".

Dan Lavry
Lavry Engineering, Inc.



Dan, if I were to use a piece of zip cord to carry word clock signals, would I experience more jitter on the receiving end of my system?  I am asking this question honestly, as the thought that a cable could contribute to jitter issues seems at least reasonable.

If the zip cord is bad news, then would proper impedance and capcitance ratings improve the jitter situation?  If so, could Apogee's cable be built with exacting standards that help resolve such issues?

I sure have learned a lot from this discussion, and I don't mind the snapping back and forth.  This type of discussion will make everyone's products better, since you know there are people there to question it.

Now I'll refrain from posting to this thread 10 times a day for a little while.

Chris
Title: Re: Proper word clock implementation
Post by: dpd on March 22, 2005, 09:43:14 pm
Cables can distort the digital waveforms and close up the 'eye pattern' which will have the effect of increasing the bit errors in any subsequent detector.  Most detectors are going to clip the incoming waveform, detect the clock and make the 1/0 detection at the mid-bit point, precisely to eliminate the effect of jitter - at least the bit synchronizers I have used do this.  (not sure about AES detectors).

But, unless the cable is defective (e.g. intermittent), what mechanism exists in the cable to create jitter?  There are no active, noise-inducing, parts of a cable so about all it can do is distort the waveform.

Am I missing something here?
Title: Re: Proper word clock implementation
Post by: jimbo-baby on March 23, 2005, 01:35:54 am
it occurs to me that the "improvements" in sound reported with use of the big ben may be because of what it adds to the signal rather than what it takes away (i think someone used the term "pretty jitter" to describe this). if that's the case, and some people think it does sound better, it is then being used the same way we use a 1073...to colour the sound in a way we think sounds cool. what is the goal of a converter? most people would say transparency. if you don't care about the number crunching and you just dig whatever sounds more ear pleasing, then go with it, and be satisfied. but don't complain about someone like dan, who has every right to question apogee over it's designs, being such a respected designer himself. he's not just bagging them out, he's asking them to explain how exactly their claims are possible. if they have some amazing new technological breakthrough, then they can patent it, and explain it to all. it'd be fantastic to do a blind listening test, but the results should be measured technically as well. that way everyone can make up their own mind. but you'd better beware if you're using a DA which adds it's own "ear pleasing colour" for in the box mixing or mastering, because then you're kind of missing the point of the flat response of your expensive room, acoustic treatment and monitors.
Title: Re: Proper word clock implementation
Post by: danlavry on March 23, 2005, 01:55:30 pm
crm0922 wrote on Wed, 23 March 2005 01:31

danlavry wrote on Tue, 22 March 2005 17:28

Max,

Yes I know, you cannot argue about “Ultra Low jitter cable”. No one can. You don’t have any measurements to show, just the regular SNAKE OIL blended with personal attacks, followed by claims of hearing it better.

ULTRA LOW JITTER CABLE IS SNAKE OIL. No way around it! Oops, you forgot to answer the other question about the "Temperature Controlled Cable".

Dan Lavry
Lavry Engineering, Inc.



Dan, if I were to use a piece of zip cord to carry word clock signals, would I experience more jitter on the receiving end of my system?  I am asking this question honestly, as the thought that a cable could contribute to jitter issues seems at least reasonable.

If the zip cord is bad news, then would proper impedance and capcitance ratings improve the jitter situation?  If so, could Apogee's cable be built with exacting standards that help resolve such issues?

I sure have learned a lot from this discussion, and I don't mind the snapping back and forth.  This type of discussion will make everyone's products better, since you know there are people there to question it.

Now I'll refrain from posting to this thread 10 times a day for a little while.

Chris


Chris,

First, let me have some fun:

Let us give Max the opportunity to dig a bigger hole!
Say Apogee is right. Say they discovered that there is another physical law to be added to the 5 Maxwell equations. Maybe not, I will be flexible. Say there is some mechanism that makes cables manufacture jitter. That is what Max said is it not?

So in the good old tradition of cable specs, lets come up with a figure for pico second per foot (cables are specified in pF/foot , Ohms/foot, uH/foot…). Say 1psec /foot jitter sounds like a great spec to me (why not, I can pull things out of the air as well as Apogee). So take a zero jitter clock (assuming it is there) and add 100 feet of cable, and you have 100psec jitter.

So that is one more argument leading us to the conclusion that internal clocking is better, and assuming Big Ben has little jitter, 100 feet of cable will do you in. So go for internal clock.

There is still one more problem left. No one is saying what temperature controlled cable is. I thought about it long and hard, trying to figure out what the hack is temperature controlled cable. I figured if I say the wrong thing, I will be accused of deception, but let me give it a try:

There is a hidden oven built around the copper to make sure the electrons are happy. If they get a cold, they may start shaking very fast and when they enter the low jitter PCB trace they will be very jittery. I wonder if the temperature control is solar or nuclear.

That is the best I can come up with, and the sellers of that will not shed any more light on the issue. Until such time, I have the right to mock it. I think the customers have the right to demand an explanation!
   
Now, that I had my morning fun… to your question

If I carry your argument forward, then resistors make jitter, capacitors make jitter, everything does. Of course passive components do not generate jitter, least of all a cable, a solid piece of copper.

A cable bandwidth, capacitance, resistance, inductance, the skin effect, the proximity effect and the rest of the specs obey the same principles that ALL OF ELECTRONICS does – the 5 Maxwell equations, the laws of physics that govern all electricity and magnetism (which always go hand in hand).

A resistor generates noise. That noise is and the mechanism which generates it is all in line with Maxwell’s equations. There is a mechanism at play that is explainable. Say I build a comparator circuit for word clock detection. I will use resistors. To set a detection point (threshold). If I use higher value resistors, they will generate more noise, and the outcome can be more jittery.

Does that open the door to salesmen to sell low value resistors as "low jitter resistors"?

Each part, be it a resistor, IC, cable… has specific characteristics that are INDEPENDENT of its application. A 10 Ohm resistor is ALWAYS 10 Ohms resistor, be it for audio, instrumentation or just in the stock room all by itself. A 1/4 watt resistor is always a 1/4 watt resistor. I know a lot of things about that resistor and they GO WITH THE COMPONENT.  Of course if I dissipate more power than the rating allows,  I will fry it. If I only dissipate 1mW in a 250mW device I have huge margin of safety. It does not mean that I can call it high reliability device. IT IS THE WAY I USED IT that made it high reliability.

It would be appropriate to consider the specs that come with a cable. You may want it to be of certain bandwidth, capacitance, shielded, and so on. That is how you put a system together. But to say that the cable makes jitter?

You can say that a system has low jitter because of a list of factors and cable bandwidth may be one of them. Cable shielding may be another. Much of the outcome is about the interaction of different factors such as cable driver characteristics, load, environment noise, grounding. So you cannot attach attributes to a component characteristics, that varies from case to case, from setup to setup.

I can cut a tree and put it on a high speed jet. Does it become a “high speed tree”?

And again, I really want an explanation regarding the meaning of temperature controlled cable. Apogee will not answer me. They are the “Wal-Mart of converters”.  There are a lot of ear people in the industry, that do not know much technical. But almost every one knows that there is a thing called jitter and “it is a bad thing”. So you sell them a flawed concept about an external box fixing a problems in another box, and that is a crock. But they push it anyway. Even after their engineer backs off the claims AND HE DID RIGHT HERE ON THIS FORUM. ANYONE THAT CAN READ CAN SEE IT FOR THEMSELVES. Then you add some BS about ultra low jitter temperature controlled … it is really bad for the industry to have people believe in BS.

You see, ear people get to decide what they want, what they hear, what they like. In a sense, everyone is “king” of their own turf. But this forum is technical, and that “equality” does not exist here. Some people are technical gurus and others are totally uninformed. One can have a lot of awards in the ear side and it does not make them technical. One of the difficult challenges for me is to learn to accept that a lot of visitors here are just “used to a different style” where they get to decide everything. A guy came here yesterday talking about his awards and many years of work as if it is the most importent thing for a cable or clock discussion. What about my many years of dealing with the ELECTRONIC aspects of the cable? I have no cable awards. In fact, it is the end of the era where “fat cat club” ruled audio by ear, ignoring technical realities, with technical awards given by name recognition (thus advertising money). Where is it all leading?
The method of advertising crock in audio has been traditionally to drum up some well known names and have them say nice things for you in exchange to something  else.  
Technical folks are a lot less willing to “just say anything”. But hearing people don’t look at numbers, and they “bend easier”. You can see the discrepancy between Lucas and Max. Lucas came back to a defensible position. Max will not. Perhaps Lucas wishes that his company would have not said that their cable is low jitter temp controlled.  

So naturally, all the snake oil in audio comes back to drumming up folks willing to say things about “how it sounds”. You make them believe you are about substance and they will back you up. Some have so much confidence that they do not even bother with a blind test. They have done so for many years, being respected, even reinforced for such behavior.

And now I am here, calling it BS. Yes Max, the equipment salesman from Sweetwater, now at Apogee, has attacked  me and I will respond to it in the next few days. I am still here, at almost 60 years of age, after designing lots of gear for my own and various companies and making a very good living, with very appreciative and happy customers.

I have 3 new products waiting for me to finish them if only I could get away from this forum for a while.

Regards
Dan Lavry
www.lavryengineering.com




Title: Re: Proper word clock implementation
Post by: Logichead on March 23, 2005, 05:44:10 pm
TOWARDS COMMON SPECIFICATIONS FOR DIGITAL AUDIO INTERFACE JITTER

In the for what it's worth department, google the title above and you will find a white paper on jitter. In section 3.8 it appears to say that jitter is induced by lower bandwidth cable. This would mean high bandwidth cable would be less likely to induce jitter. Isn't that the point of the Apogee digital cable? (BTW, they include measurements.)

So which manufacturer is selling the snake oil? Pot, kettle, black?

More importantly, when these threads become personal they lose their validity. It is with regret and with respect that I make the following request:

Dan - you need to stop using this forum to sell your products and yourself, or  resign as moderator. It is inappropriate for you to use your position to attack your competitors and former employers. You (and any other manufacturer) have too much to gain here to be unbiased.

The rest of you - use your friggin' ears to examine what the numbers tell you. There was a time when everyone knew that it was impossible to fly.

Best regards....H









Title: Re: Proper word clock implementation
Post by: chap on March 23, 2005, 08:16:39 pm
Dan,
I've stated that I respect your work and your products.
I've also experimented enough to know that I do prefer Apogee's X series.  I applaud your desire to have a technichal discussion
but when you blatently dismiss listening as an aspect of the discussion, you do a discredit to your own work.

Again, until the paint is on the canvas, it's just a can of paint.
Maybe one paint specs out better but you have dismissed the painters who would buy your paint.  You can delete this if you want
but to me, it further reveals that these anti Apogee threads are as much about your 'feelings' as they are about technichal aspects.

I continue to respect your work. I have used your paper on 192kz
sample rates in a college course I teach.  I agree with you and I thought it was good work.  I don't understand why you would demonize a competitor.  I could say Bill Frissell isn't a good player but that would make me look like a jerk (he's a great player).  His abilities don't diminish mine.  

Aren't there other fish to fry than guys who work hard on innovative products and deliver on them?

All respect,
chap
Title: Re: Proper word clock implementation
Post by: chap on March 23, 2005, 08:20:33 pm
ditto that logic.

If it helps Dan, I'm only a few years younger than you and would
advise that you go and make those products.  It's great to be passionate
about something (I am) but try using it it a more positive way.
This thread is not helping you.
peace,
chap
Title: Re: Proper word clock implementation
Post by: danlavry on March 23, 2005, 09:36:41 pm
Logichead wrote on Wed, 23 March 2005 22:44

TOWARDS COMMON SPECIFICATIONS FOR DIGITAL AUDIO INTERFACE JITTER

In the for what it's worth department, google the title above and you will find a white paper on jitter. In section 3.8 it appears to say that jitter is induced by lower bandwidth cable. This would mean high bandwidth cable would be less likely to induce jitter. Isn't that the point of the Apogee digital cable? (BTW, they include measurements.)

So which manufacturer is selling the snake oil? Pot, kettle, black?

More importantly, when these threads become personal they lose their validity. It is with regret and with respect that I make the following request:

Dan - you need to stop using this forum to sell your products and yourself, or  resign as moderator. It is inappropriate for you to use your position to attack your competitors and former employers. You (and any other manufacturer) have too much to gain here to be unbiased.

The rest of you - use your friggin' ears to examine what the numbers tell you. There was a time when everyone knew that it was impossible to fly.

Best regards....H





First, what julian is calling a “cable induced” jitter that is a mechanism that was well explained and discovered by Hawksford. The Hawksford paper is called "Is the AES/EBU standard Flawed" It involves an active digital audio signal that changes with the music such as AES (clock plus data). Not a signal that is repetitive such as in the case of WORD CLOCK.

Apogee is selling a word clock cable and calling it a low jitter cable. You are bringing up an application that has nothing to do with word clock, nor does it have to do with any periodic signal such as sync to AES black.

You are transferring results from a different application, where one cable characteristic (high frequency cutoff) interacts with a transformer lower bandwidth to cause a signal droop that will change a comparator trip point. All of that only happens when the signal is of a specific type.  

Your example does not apply here at all. The way a cable, a transformer and the rest of the circuit reacts to an AES signal with data does not have anything to do with word clock.

This is in fact a good way to explain why cables don’t make jitter. There is a big difference between a component characteristic and the way a component behaves in a circuit when it is interacting with other components AND given signals.

That is why the concept of cables and jitter are miles apart.

The characteristics that belong to a cable description are constant for all applications, such as capacitance per foot, resistance and so on.


In the paper you quoted the application is different. It is about transfer of a different signal and it takes interaction of ALL 3 elements to make that jitter:

A. A changing signal
B. Rise time limitation due to bandwidth
C. DC blocking high pass

A given cable that may be a problem for transfer of digital audio to a DA can be great for word clock.

Cables do not make jitter. Resistors don’t make jitter. Capacitors do not either. A combination of parts in a given application can cause jitter, and changing resistor values may alter the jitter, but that does not make a specific resistor value a low jitter resistor. The same exact resistor may be the worst value in another application.

I would not mind an cable specification of bandwidth, capacitance, inductance attenuation per length, frequency…  All of those are cable characteristics.
Jitter is not a cable characteristic.

Say that Julian were to use an AC coupling made of a blocking Cap value of  0.1uF. Say a change to 1uF yielded better results. Does it make a 1uF cap a low jitter cap?  

A sales guy may claim with a VERY LONG STRETCH that a 1uF cap is low jitter for this circuit. But the improvement would not be there when you go for a different application. That’s is why you do not call a 1uF capacitor a low jitter capacitor. No one will call a 1uF capacitor a low jitter capacitor just because it improved jitter in one specific circuit. So no one should call a cable a low jitter cable just because it helped achieve lower jitter in one circuit.

Don’t forget to add to it the fact that we are talking about a different circuit, a word clock link not an AES data link, which is the link discussed in Hawksford and Julian’s papers.

Dan Lavry
www.lavryenginering.com
Title: Re: Proper word clock implementation
Post by: Logichead on March 23, 2005, 11:46:04 pm
Example 2:
A MARKETTING COMPANY ?X? sells cable that they claim are: ULTRA LOW JITTER CABLE and TEMPERATURE COMPENSATED CABLE. I do not know what temperature compensated cable means but I KNOW that THE CONCEPT OF ?ULTRA LOW JITTER CABLE? is not true. Cables don?t have jitter . A cable is a PASSIVE device. It is just a ?piece of material?. The level of jitter is determined by the electrical signal driving the cable! There may be other factors (variable load, induced noise and more). One thing for sure: but there is no such a thing as ultra low jitter cable.

?Jitter to cable? is like a ?laptop to Moses?
?Jitter to cable? is like a ?Pistachio nut to a whale?
?Ultra Low jitter cable? is like ?super high speed tree?
?Ultra Low jitter cable? is like ?butter to a headphone jack?


Please know that in future that statements about personal opinion and marketing will be deleted unless backed up with technical input.

Dan Lavry
Lavry Engineering, Inc.
[/quote]


There does indeed appear to be "cable induced jitter" as noted in the white paper. You seem to agree, in contrast to your previous statements, seen above. To quote the paper, "this model shows how cable induced jitter (is) strongly dependant on the bandwidth of the link." The example quoted started with this statement:

"If a jitter-free interface signal is transmitted down a cable then the data pattern in the signal will modulate the signal zero-crossing timings.  This modulation is such that patterns of zeroes produce more delay to the transition timings than patterns of ones, so as the data varies the timings modulate in sympathy."

And later on reached this conclusion:

"If we consider a short chain of digital audio devices, where each device is locked to the previous one, we have several contributions to the jitter at the end of the chain.  Each device will add its own intrinsic jitter, and each interconnecting cable will make some contribution with cable-induced pattern-dependent jitter."

Notice also, in the above statements, you did not limit your statements to word clock cables only. (Yes, the thread is labeled "proper word clock implementation" but the subject has ranged far and wide since the thread's begining.)

It remains inappropriate for you - a manufacturer - to moderate a thread which villifies, either directly or indirectly, other manufacturers. Your bias is showing.

Best regards....H







Title: Re: Proper word clock implementation
Post by: Nika Aldrich on March 24, 2005, 05:43:50 am
Is this thread really about semantics?  Cables don't directly induce jitter - the jitter comes from EMI, RFI, other types of noise.  Good cables merely reject that jitter better.  In this capacity, bad cables are said to "induce" jitter because it is accepted that some degree of noise is a part of the environment, so cables that allow this noise to affect the signal are said to "induce" that jitter.  This seems like pretty basic stuff to me?

As far as the filtering of cables, I assume it has been discussed that cables act as filters and can thus cause modulation of the square wave pulses and can therefore create jitter that was otherwise not present.  Lowering the cutoff frequency of the filter can increase the modulation.  The less square the waveform the greater likelihood that the cable's inherent filtering properties exacerbate the jitter present from noise sources as well.  For these reasons as well, good quality cabling helps reduce jitter manifested through the process.

I think that the filtering properties of a cable do indeed "induce" jitter, but even if we reject this for some reason and say that cables don't "induce" jitter, but they can be said to "reject" jitter, and if you accept that that jitter from noise is an inherent part of the environment one could certainly decide to describe the use of poor cables as "inducing the manifestation of this problem."  It seems petty to try to pick apart these semantics.

As for Apogee's "low jitter cable" (or "ULTRA low jitter cable" or whatever) I think the implication is that it has high jitter rejection, has better inherent filtering properties, and thus exudes lower amounts of jitter?  

Something seems petty about debating the semantics at play here, especially when we all know what was meant, and when AES whitepapers have been written on the significance of using good quality cabling so as to provide lower jitter to digital devices.

Nika
Title: Re: Proper word clock implementation
Post by: bananahill on March 24, 2005, 08:04:13 am
Dan,

I have to agree with Logichead.  You are out of line here.
You have every right to express your opinions and findings and argue these points.
The conflict is your role as moderator.  From the tone of many of the posts, no one is moderating this thread.  You are a very involved participant.

I think you should remove yourself as moderator and continue to post.
If you want to remain moderator, you need to do this in an objective and unbiased manner.
Title: Re: Proper word clock implementation
Post by: Logichead on March 24, 2005, 08:16:27 am
Nika Aldrich wrote on Thu, 24 March 2005 10:43

Is this thread really about semantics?  Cables don't directly induce jitter - the jitter comes from EMI, RFI, other types of noise.  Good cables merely reject that jitter better.  In this capacity, bad cables are said to "induce" jitter because it is accepted that some degree of noise is a part of the environment, so cables that allow this noise to affect the signal are said to "induce" that jitter.  This seems like pretty basic stuff to me?

As far as the filtering of cables, I assume it has been discussed that cables act as filters and can thus cause modulation of the square wave pulses and can therefore create jitter that was otherwise not present.  Lowering the cutoff frequency of the filter can increase the modulation.  The less square the waveform the greater likelihood that the cable's inherent filtering properties exacerbate the jitter present from noise sources as well.  For these reasons as well, good quality cabling helps reduce jitter manifested through the process.

I think that the filtering properties of a cable do indeed "induce" jitter, but even if we reject this for some reason and say that cables don't "induce" jitter, but they can be said to "reject" jitter, and if you accept that that jitter from noise is an inherent part of the environment one could certainly decide to describe the use of poor cables as "inducing the manifestation of this problem."  It seems petty to try to pick apart these semantics.

As for Apogee's "low jitter cable" (or "ULTRA low jitter cable" or whatever) I think the implication is that it has high jitter rejection, has better inherent filtering properties, and thus exudes lower amounts of jitter?  

Something seems petty about debating the semantics at play here, especially when we all know what was meant, and when AES whitepapers have been written on the significance of using good quality cabling so as to provide lower jitter to digital devices.

Nika


No doubt there are semantics involved here, as is often the case on a forum. That's not what I find important.

First, if you read that white paper (the pdf is included a few posts up) it appears that low bandwidth cable actually adds jitter to jitter-free digital audio. I am only interpreting what I read, but I'm pretty sure I have it right. This was news to me, as I suspect it is to you.

Next, Dan has been using this forum inappropriately, just read his posts if you are uncertain. He has directly singled out Apogee, his former employer and a current competitor for his most severe criticism. Fine, criticise away, but not as the moderator.

Dan, again - you are damaging your reputation as well as Pro Sound Web's. Imagine your response if it were someone at Apogee moderating this thread and saying these things about you.

IMO Apogee should contact the operators of the forum and complain. If only to save Dan from himself, this entire thread should be deleted.

Best regards....H

Title: Re: Proper word clock implementation
Post by: Nika Aldrich on March 24, 2005, 09:10:26 am
Logichead wrote on Thu, 24 March 2005 08:16

First, if you read that white paper (the pdf is included a few posts up) it appears that low bandwidth cable actually adds jitter to jitter-free digital audio. I am only interpreting what I read, but I'm pretty sure I have it right. This was news to me, as I suspect it is to you.


If you're talking about the ability for the filtering characteristic of cables to provide a modulation effect on the square wave data run through the cable this isn't exactly new stuff - it's pretty basic waveform theory.  The question has to do with the practical results of that - what type of bandwidth is necessary in order to "induce" this jitter on a bad cable, especially when dealing with 44.1kS/s pulses?  I suppose I could run a test in Matlab and see.

So I concede Dan's point in advance of actually knowing - that even in this pathological case, jitter isn't induced for some reason.  Still, there can be little denying that some cables reject jitter better than others.  

Nika
Title: Re: Proper word clock implementation
Post by: bobkatz on March 24, 2005, 09:36:20 am
danlavry wrote on Wed, 23 March 2005 21:36



Cables do not make jitter.




It's also the language that is being used in the promotion of the product. The fact is that a cable must have the right impedance, and ideally maintain that impedance when flexed or bent in order to reduce reflections in the circuit and have a small eye-pattern. As I understand it, these reflections can confuse a receiver and cause time-based problems. Whether that is being a "low jitter cable" is a matter of how you interpret the language. I'd certainly want to buy a cable with the right impedance, with the stabilizer wires inside to maintain that impedance. Those are important characteristics. But it makes for dull marketing language!

BK
Title: Re: Proper word clock implementation
Post by: danlavry on March 24, 2005, 09:37:30 am
I thought more about what I said last night, and I just want to be more specific than I was.

I find the term “cable induced jitter” to be a poor choice of words. Julian did the math well, and he performed his experiments and collected data. I guess he was focused on the math derivation, and on getting the massage across that AES signals are impacted by a cable.

I would say that the jitter in an AES link is impacted by the cable, impacted by the transformer or capacitor. Impacted, not induced. The word induced suggests that the cable does something actively. But the cable does not act. It reacts. There is no capacitor induced jitter either for the same reason – the capacitor is passive. As a passive device, a cable reacts differently to different input signals and the reaction depends on interaction with other components.

The original discoverer of the effect of jitter on a digital audio link was Hawksford. He explained the mechanics very well. The signal has droop due to low frequency coupling. The droop is related to the time duration of each logic state. With data changing on an AES link, the droop time duration changes also which causes a “wobble”. The waveform voltage goes up and down tracking the digital logic states. Given that the signal transitions between states has a finite rise time (it is not vertical) the time where a comparator reference (threshold) is crossed by the wobbly signal also wobbles. That is the jitter we are taking about. Data link jitter.

I find the term cable induced jitter to be off the mark. The cable does not induce jitter. That is not the way I heard Hawksford explanation to be, nor is it my understanding of how the mechanism is. The cable does not do anything by itself. It does not induce anything it is passive. It “just sits there”. Just like other passive parts (resistors, caps, diodes, empty circuit boards, metal chassis and more). Is a metal chassis a low jitter chassis?  Supose it that help shield a circuit from interference that cause jitter in one application, and it increases it in another case. The chassis has attributes (length, width, material data…)  but you can not say it is an low jitter chassis. It is just a passive piece of metal. It is passive material and it does not “make waves” by itself. Just like a cable.

I assume there was no harm intended by Julian, just a poor choice of a word. But someone is using that for commercial promotion of a product. And we are not taking rocket science here, it is just a piece of passive copper with some insulation. Why does someone home in on the one “off phrase” in an article? Is it lack of depth and technical competence? Readiness to just say anything? A misinterpretation of a poor choice of words.

Does someone that says that cables make jitter and sell a low jitter cable be required to explain how cables make jitter? Just pointing at “so and so said this or that” does not make it so. Where is the mechanism for cables to induce anything? It is not there. If you believe it and sell it, measure it. Guess what- whatever jitter numbers you measure in an AES data link will change with different transformers and even with sample rate. Cables are reactive. They are put into a structure and become a part of an interaction.

A piece of cable on the ground, connected to nothing does not “make signals”, and jitter without a signal makes no sense. A cable has characteristics. There are mechanical (weight length…) and electrical (capacitance, resistance…) so you can say “low capacitance cable” or 10 foot cable. Those characteristics go with the cable wherever it is in all applications.

It is improper to take a passive device into a specific application with specific conditions and assign to it attributes that belong to that specific case. The number 2 can mean 2 billion dollars or 2 cents. Shell we call the number 2 very significant? It is just there passively. It is not inducing money. Like a cable, the number 2 is “just there” until you give it some use, some context. It is passive, like a cable. The word is passive.  

And suppose that I am dead wrong (I am not). Then the cable induced jitter explanation is a reason to stay away from external clocking in favor of internal clocking. External clock box always requires cables. If Apogee thought that a cable makes jitter they should encourage the use of internal clock, not an external Big Ben.  

Regards
Dan Lavry
www.lavryengineering.com
Title: Re: Proper word clock implementation
Post by: Logichead on March 24, 2005, 09:52:37 am
Nika Aldrich wrote on Thu, 24 March 2005 14:10

Logichead wrote on Thu, 24 March 2005 08:16

First, if you read that white paper (the pdf is included a few posts up) it appears that low bandwidth cable actually adds jitter to jitter-free digital audio. I am only interpreting what I read, but I'm pretty sure I have it right. This was news to me, as I suspect it is to you.


If you're talking about the ability for the filtering characteristic of cables to provide a modulation effect on the square wave data run through the cable this isn't exactly new stuff - it's pretty basic waveform theory.  The question has to do with the practical results of that - what type of bandwidth is necessary in order to "induce" this jitter on a bad cable, especially when dealing with 44.1kS/s pulses?  I suppose I could run a test in Matlab and see.

So I concede Dan's point in advance of actually knowing - that even in this pathological case, jitter isn't induced for some reason.  Still, there can be little denying that some cables reject jitter better than others.  

Nika


Thanks for your reply Nikka. I also appreciate the reasoned tone of your comments. The practical implications of cable induced jitter appear to be small. There are indeed limits on what the numbers tell us! While you appear to say there is essentially no such thing (in practical terms) as cable induce jitter, the white paper seems to say otherwise. It is an infinitismal point we are dealing with here, no need for histrionics. I can see both conclusions.

As you say, "there can be little denying that some cables reject jitter better than others." That is all Apogee appears to claim, at least on their website.

I don't want to read too much into it, but your silence on Dan's actions as participant and moderator leaves me wondering how you feel about a manufacturer using a forum as Dan has.

Best regards....H






Title: Re: Proper word clock implementation
Post by: Nika Aldrich on March 24, 2005, 10:01:40 am
Eh gad, that was the longest, most vindictive and most nuanced apology I've seen in a long time.

Nika
Title: Re: Proper word clock implementation
Post by: Logichead on March 24, 2005, 10:37:00 am
Nika Aldrich wrote on Thu, 24 March 2005 15:01

Eh gad, that was the longest, most vindictive and most nuanced apology I've seen in a long time.

Nika


What he said.

Best regards....H
Title: Re: Proper word clock implementation
Post by: danlavry on March 24, 2005, 11:13:14 am
Logichead wrote on Thu, 24 March 2005 14:52

Nika Aldrich wrote on Thu, 24 March 2005 14:10

Logichead wrote on Thu, 24 March 2005 08:16

First, if you read that white paper....


...necessary in order to "induce" this jitter...

Nika


While you appear to say there is essentially no such thing (in practical terms) as cable induce jitter, the white paper seems to say otherwise.
Best regards....H




You obviously did not read a word I said. The white paper is misusing a single word "induced" instead of "react" and Max the salesman from Apogee thinks it is sufficient reason to sell a cable. A separate and valid point: The white paper does not apply to word clocks, thus word clock low jitter cable is out. Third, if Max were correct (he is not) then it means that external clocking is even worse then I pointed out because of the added jitter (it calls for a cable)...

You argue without registering the content of what is being said!!!

You are not staying with it. You are not participating in a technical dialog. You are DIVERTING THE ISSUE to Dan as a moderator.

You asked Nika to "rate me" as a moderator.

I am asking all the participants and viewers to rate you on your level of understanding and your style of participation here.  

Dan Lavry
www.lavryengineering.com
Title: Re: Proper word clock implementation
Post by: Logichead on March 24, 2005, 11:36:30 am
What our levels of technical understanding are is not the issue - unless we're going to be taking entrance exams before posting. However much this may have begun as a technical discussion, your actions as moderator have taken matters elsewhere.

Do the right thing, Dan.

Best regards....H
Title: Re: Proper word clock implementation
Post by: danlavry on March 24, 2005, 11:43:21 am
I am trying to stop another run away train. That "low jitter cable buzz" is starting to spread. It is just one more of those faulty "audio concepts" that in my opinion should be stopped. I present technical input and ask questions. There should be no need for insults but I will not be walked over. Cables don't induce signals and the audio world should learn this.

I saw cable  advertisement  by Gepco, where they saying: “They (the cables) feature minimal attenuation, low jitter and”....

Meanwhile, the misconceptions about external clocking have become wide spread, and now the ultra low jitter cable notion is starting to spread as well.  

I do not view Apogee as a competitor. My opinion of their gear is that it is consumer quality judging from the signals it generates.  Period.  I do view Apogee as having a lot of influence and marketing power, much of it from the quality of the electronics that I designed while there in the past and I question their claims for technical leadership. Avoidance and diversion is not the behavior of a technical leader.

Apogee refuses to answer the posted technical questions. They do not answer others and they do not answer me.  I am not the problem. It appears that I am the only chance to get things straight.  Apogee should not send a salesman here to argue with a high powered engineer. That is the reason the conversation turns personal and off subject.
 
What is coming up next? 384fs? I will not design gear above 96KHz regardless of the monetary consequences to my company. My stand is right and honorable.

We have here 3-4 Apogee supporters - “regulars” that instead of insisting that Apogee  engage in a technical dialog and answer questions, choose to go after me for doing a poor job of moderating. One guy seems to think that quoting a poor use of language (Cable induced jitter) is sufficient to make the principles of physics (Maxwell's equations - the foundation of electricity and magnetism) go away.

Cables do not generate signals!!! Cables don’t induce anything!

Where is a single technical comment from Apogee explaining what “temperature controlled cable” means?   One cannot put a passive device in a circuit and have it “wake up” and start doing active things like inducing anything. Cables do not manufacture signals or jitter.


One more time: Apogee engineers. What is a temperature controlled cable? Is any passive device capable of inducing a signal? Does an explanation of a cable interaction with other components under AES (with data) have anything to do with Word clock cable? If so what?

Regards
Dan Lavry

Title: Re: Proper word clock implementation
Post by: trevord on March 24, 2005, 01:56:22 pm
I am a long time lurker here
but i have to add my 2 cents on this one

first let me say i agree with dan lavry.

let me add a different take on his explanation
1) there are many jitters in engineering. clock jitter can be defined as the varying position of the edges of the clock
for example a 10ns clock would be
0 - rise edge
5 - fall edge
10 - rise
15 - fall

a jittery clock would be
0 - rise edge
5 - fall
11 - rise
14 - fall

this results in all the bad effects of jitter in the word clock for digital audio

this is a function of the clock source - if a jittery clock were transmitted thru a perfect medium it would still be a jittery clock

now for cables
assuming a perfect clock - a low jitter clock SOURCE
what happens at the end of the cable as far as clock jitter is concerned has more to do with
the DRIVER of the signal
the TERMINATION logic at the end
and
the EMI of the space around the cable
THAN the cable itself.

if your cable has some horrendous characteristic it is in relation to some standard it is expected to be used in

for example if your cable has horrible capacitance specs - it can be taken care by an accommodating design at the driving and terminating ends of the cable

saying a cable is low-jitter is laughable without mentioning what is expected at either end of the cable
the most perfect word-clock co-ax would produce the most noisy signal at the end with the wrong termination

a claim can be made that these cables would have the least effect if a perfect word-clock is perfectly driven and perfectly terminated
but this is dan's point - all the "perfection" concerns things that are NOT the cable!!!

it is also a circular marketing argument
it can be  said once "standard termination and driver specs" are met, these cables are better than the average
but if "standard termination and driver specs" are met then an average cable is adequate - by the definition of the standard.

i am surprised at the opposition to dan's observation in a professional engineering forum

trevor

Title: Re: Proper word clock implementation
Post by: Eliott James on March 24, 2005, 06:05:02 pm
Oil and water. "Ear" people and "Tech" people. The two seldom meet in understanding, although both are equally right and wrong on a consistant basis about the same topics.
Title: Re: Proper word clock implementation
Post by: danlavry on March 24, 2005, 06:17:05 pm
It seems to me that many people here do not appreciate the important distinction between the AES mechanism that introduces jitter and the Word clock transmission that does not have that mechanism. So I will explain (credit for the discovery of the mechanism belongs to the very brilliant professor Hawksford):

The plot below is a “scope picture” of 4 waves. The top one (in red) shows a sequence of “highs” and “lows”, but the duration of each state is not equal. The wave stays high for “extra long time” between about 250 and 575, it stays low for “extra long time” between 700 and 850. The point is: it is not a periodic wave!

index.php/fa/885/0/

The blue plot represent the same wave but with slower rise and fall times. The slowing down of rise and fall represent the combined limitations of the upper frequency capability of the transmission link. That limitation is due to components such as the signal driver, driver output transformer, cable and more.

The black trace below shows what happens to the blue trace (rise time limited signal) when passing it through a transformer or capacitive coupling. Transformers are common for AES. Capacitive coupling is more common for SPDIF. Note that the AC coupled signal droops further down when the input has a longer high state, and drift further up when the input has a longer low state. The outcome is a “wobbly” signal which does depend on the duration of the low and high state as is the case with AES signal.

The last trace, the purple one, is a receiver comparator threshold. Note that the transition times are when the black wave crosses the purple wave. Those points can be early or late because of the time dependent droop (due to variation of the low high states). For example the negative transition around 780 is very early. The signal (black trace) barely moved and the comparator reacts. That variable data dependent timing interacts with BOTH rise time a droop to generate that data link jitter.      

See next post for the word clock case

Regards
Dan Lavry
www.lavryengineering.com
Title: Re: Proper word clock implementation
Post by: danlavry on March 24, 2005, 06:28:03 pm
The presentation for word clock is very similar to the one I posted in the last thread. But word clock is simpler because the state alternate between low and high at a periodic manner (low, high, low, high....)

index.php/fa/886/0/

The red trace is the input.
The blue trace represents the rise time limitations (driver, cable...)
The black trace represents the low frequency limitation (transformer or capacitive coupling.
The purple is the comparator threshold.

Due to the periodic nature of the word clock signal, all cycles are identical, and there is no signal wobble therefor the mechanism that applies to jitter generation with AES signals DOES NOT APPLY HERE.

So quoting Julian's paper here (with or without the mis- statement about cable induce jitter) is OUT OF PLACE.

Regards
Dan Lavry
www.lavryengineering.com
Title: Re: Proper word clock implementation
Post by: Nika Aldrich on March 24, 2005, 07:51:03 pm
Dan,

You keep talking about how Apogee is out of line with their marketing that cables can "induce jitter" and that there's, therefore are better because they are "low jitter cables."  You keep discussing the word "induce" as a straw man argument, discussing how cables can't "induce" and therefore Apogee's argument is bad.

Please tell me, where does Apogee discuss cables "inducing" jitter in their marketing text?  

What I read in Apogee's marketing of "low jitter cables" is that their cables provide greater jitter rejection.  Why is this reading erroneous?

Nika
Title: Re: Proper word clock implementation
Post by: danlavry on March 24, 2005, 09:42:37 pm
Nika Aldrich wrote on Fri, 25 March 2005 00:51

Dan,

You keep talking about how Apogee is out of line with their marketing that cables can "induce jitter" and that there's, therefore are better because they are "low jitter cables."  You keep discussing the word "induce" as a straw man argument, discussing how cables can't "induce" and therefore Apogee's argument is bad.

Please tell me, where does Apogee discuss cables "inducing" jitter in their marketing text?  

What I read in Apogee's marketing of "low jitter cables" is that their cables provide greater jitter rejection.  Why is this reading erroneous?

Nika


Nika,

This time I have done the footwork looking for quotes through previously written posts. Next time please do it yourself. And so here is the outcome of my footwork below:

Max said:

Finally, let it be clear to the end user that you are deceiving them with another semantic argument when making reference to our cables. Of course cables don't lower jitter, but cables do create jitter, some more than others. Apogee's Wyde-Eye cable creates less jitter artifacts than other cables. We did not call it the "jitter free" cable or even the "jitter reducer", so this blatant attack is a deception.


I say:

You see, Max said “cables do create jitter”. He then decided not to continue but the “Apogee regulars” picked up the argument by pointing to a particular section in Julian’s paper. It was Logichead 3-23-05. That is where Julian uses the phrase Cable induced jitter in context to AES data link.

Of course right after I explained that cables don’t make jitter, he came back with the quote from Julian’s paper as if it is some kind of a bible, beyond Maxwell’s laws of Physics….

So Max said that my comments were deceitful because cables do create jitter. The few “pro Apogee guys, no matter what”  found a poor phrase “Cable induced litter” and of course some still do not get that there is no such thing as cables generating jitter, that cables do not induce jitter, and that Julian paper was not referring to what I am talking about which is ultra low jitter word clock cable, where higher bandwidth may not always be a plus.

There are a lot of people in audio that do not understand jitter but know it is a “concept” and it is “bad”. So one can sell stuff by marketing all sorts of products as “low jitter”.

From the Word Clock Cable literature at the Apogee website one finds this misinformation:

"The 110 ohm cable maintains the exclusive double shield system of gold foil and braiding for improved clarity and RF rejection. Discerning high-performance audio enthusiasts can now enjoy the benefits of this ultra-low jitter and temperature
controlled cable, which has become the leading choice of cable for mastering engineers and the professional audio industry in general since its intoruction in 1993."


Since you and Max worked together in sales at Sweetwater it is possible that you got “numb” to all sort of claims. I am an engineer and of course think of the consumers in a different way than a salesman. Obviously, an innocent person may believe that he/she will have less jitter by buying a cable that is touted to have ultra-low jitter. The words temperature controlled as you well know, sound impressive to a non-technical person. I will show you an example of an innocent victim: Logichead really believed that cables induce jitter. This belief was based on a misuse of one word in an otherwise relatively ok white paper that does not apply to work clock.

Dan Lavry
Lavry Engineering, Inc.

Title: Re: Proper word clock implementation
Post by: Nika Aldrich on March 25, 2005, 07:23:31 am
Dan,

First, to correct your spreading of misinformation, Max, nor anyone else at Apogee, has ever worked in sales at Sweetwater.  

Second, the quote you refer to is clearly referencing 110ohm cable, which is AES/EBU cable, which even YOU say can "induce" jitter according to Hawksford, Dunn, et al.  That quote clearly does not refer to 75ohm wordclock cable, does it?  And even IF the website were referring to 75ohm wordclock cable (which it very clearly is NOT) the Apogee website information would STILL not be incorrect, for the wordclock cable could REJECT more jitter than other cables (from EMI, RFI, etc.)  and thus indeed be a "low jitter" cable.  Yes?

It seems that Max's quote that "cables can create jitter" was correct in regards to AES/EBU cable, and that their website says the same thing.  You seem to be extrapolating from these quotes from Max that ALL cables can create jitter, including wordclock cables, but Max says no such thing.  He also doesn't say that Cat5 cables can induce jitter, that romex can induce jitter, or that aircraft cable can induce jitter.

But let's go a step further.  I will say here that ALL cables can induce jitter - it just depends on what you use the cable for.  If you transmit pure square waves on the cable (word clock pulses, like BNC cables are TYPICALLY used for) then the cable itself will not "induce" jitter.  It can certainly help "reject" jitter, however.  If, on the other hand, you use the cable to transmit more complex digital data, such as SPDIF, AES/EBU, ADAT format, MADI, or whatever else, and you intend to extract the wordclock from that signal then indeed the cable CAN "induce" jitter.  Since 75ohm cable can be used for both wordclock AND for SPDIF, it is indeed fair to say that coax cable CAN "induce" jitter, depending on what it's used for.  Apogee sells their 75ohm cable in various terminations for wordclock AS WELL AS formats that are sensitive to cable-induced jitter, such as MADI, video applications, and SPDIF.

I'm sure that, in light of this, you would agree that Apogee should go back and clarify their website and more correctly say that BOTH their 110ohm and their 75ohm cable are "ultra-low jitter cable," rather than just their 110ohm cable as in the quote you referenced last night?  After all, we want complete and accurate information, yes?  

Nika
Title: Re: Proper word clock implementation
Post by: danlavry on March 25, 2005, 01:04:57 pm
Nika Aldrich wrote on Fri, 25 March 2005 12:23

Dan,

First, to correct your spreading of misinformation, Max, nor anyone else at Apogee, has ever worked in sales at Sweetwater.  

Second, the quote you refer to is clearly referencing 110ohm cable, which is AES/EBU cable, which even YOU say can "induce" jitter according to Hawksford, Dunn, et al.  That quote clearly does not refer to 75ohm wordclock cable, does it?  And even IF the website were referring to 75ohm wordclock cable (which it very clearly is NOT) the Apogee website information would STILL not be incorrect, for the wordclock cable could REJECT more jitter than other cables (from EMI, RFI, etc.)  and thus indeed be a "low jitter" cable.  Yes?

It seems that Max's quote that "cables can create jitter" was correct in regards to AES/EBU cable, and that their website says the same thing.  You seem to be extrapolating from these quotes from Max that ALL cables can create jitter, including wordclock cables, but Max says no such thing.  He also doesn't say that Cat5 cables can induce jitter, that romex can induce jitter, or that aircraft cable can induce jitter.

But let's go a step further.  I will say here that ALL cables can induce jitter - it just depends on what you use the cable for.  If you transmit pure square waves on the cable (word clock pulses, like BNC cables are TYPICALLY used for) then the cable itself will not "induce" jitter.  It can certainly help "reject" jitter, however.  If, on the other hand, you use the cable to transmit more complex digital data, such as SPDIF, AES/EBU, ADAT format, MADI, or whatever else, and you intend to extract the wordclock from that signal then indeed the cable CAN "induce" jitter.  Since 75ohm cable can be used for both wordclock AND for SPDIF, it is indeed fair to say that coax cable CAN "induce" jitter, depending on what it's used for.  Apogee sells their 75ohm cable in various terminations for wordclock AS WELL AS formats that are sensitive to cable-induced jitter, such as MADI, video applications, and SPDIF.

I'm sure that, in light of this, you would agree that Apogee should go back and clarify their website and more correctly say that BOTH their 110ohm and their 75ohm cable are "ultra-low jitter cable," rather than just their 110ohm cable as in the quote you referenced last night?  After all, we want complete and accurate information, yes?  

Nika



Dan,

First, to correct your spreading of misinformation, Max, nor anyone else at Apogee, has ever worked in sales at Sweetwater.

OK. I will take it back. I thought everyone that worked in Sweetwater was in sales or sales support. He was the ProTools expert at tech support was he not? I call it sales, but I can take it back.

In fact, I saw his comments on the Sweetwater forum about big Ben “Call it what you want, but the difference is not subjective”.
   
   http://www.sweetwater.com/forum/showthread.php?s=7601e1c4241    f8c6123aa04414285de93&threadid=6715&perpage=15&p agenumber=4

Not subjective means what? It means objective. That goes against the present marketing which is about subjective listening!

Second, the quote you refer to is clearly referencing 110ohm cable, which is AES/EBU cable, which even YOU say can "induce" jitter according to Hawksford, Dunn, et al.

Wrong. I got that quote from their SPECIFIC WORD CLOCK CABLE description! (Go to Apogee site, click on store and nevigate to Word Clock Cable (not AES cable) or click on the link below)

That quote clearly does not refer to 75ohm wordclock cable, does it?

You just pointed out another error in their marketing. They are selling 110 Ohms cable under their WORD CLOCK.  

   http://www.digitalriver.com/dr/v2/ec_MAIN.Entry10?xid=28122& amp; amp; amp;    amp;PN=1&SP=10023&V1=331437&DSP=&CUR=840& ;amp ;amp ;amp ;PGRP=0&CACHE_ID=0

So is that another BIG ERROR in the marketing department? Or are they selling 110 Ohm cables instead of 75Ohms. to people that use 75Ohm termination? Why did not go and see for yourself? You don’t even see that you are defending an error. Your bias is showing.

And even IF the website were referring to 75ohm wordclock cable (which it very clearly is NOT) the Apogee website information would STILL not be incorrect, for the wordclock cable could REJECT more jitter than other cables (from EMI, RFI, etc.) and thus indeed be a "low jitter" cable. Yes?

No. you are getting off track! I am not going to get into a pissing match about one word or another.

The issue is not about a single phrase such as “ultra low jitter word clock” cable. It is not about a single phrase “temperature controlled cable” (which you selectively ignore!). It is not about some people saying (or not saying) that they like what they hear. Apogee is trying to make it into an issue of the ear against the technology, and it is not that either.

All of this back and forth started as a technical discussion. My recent posted graphs are what technical is about, not all of these words. It was not intended to become a “David vs. Goliath” of audio. Remember our other run-ins where you kept me busy with your daily inputs.

I do my best on this forum to educate about the technical, to give theory and info that will help individuals themselves ask questions have a better understanding of some aspects of audio. So I got into talking about clocks and AD’s.

Apparently what I said “did not sit well” with Apogee. Apogee engineer Lucas stated his position on this forum. I accepted it. Amazingly enough, Apogee did not. Someone’s requested farther dialog recently, and I did not even reply. Then Max came in and said that Apogee did not participate in a dialog, as if Lucas had not said what he did. That was the point where I renewed my comments, including the word clock cable comment.

One does not make a judgment of a product based on one piece at a time. One looks at how the pieces interact together. So let me show you how the later cable comment ties into this thread:

I can assume that Apogee and I agree that less jitter is a good goal. I say so all the time. Their stand is indicated by the fact that they are selling an ultra low jitter cable. We all agree that jitter is bad.  

That by itself provides enough grounds to reject and some crazy suggestions by some (trying to explain the impossible) that adding jitter explains the sonic improvements when driving an AD externally via a cable.

I stated that generally speaking, internal clock is better whenever possible. Apogee said otherwise.

Here is what technology leadership means:

If Max is right and cables make jitter, Apogee should be telling people that buy their box to keep connections short. Apogee should have made some effort to QUANTIFY jitter (learn how much) say 100 foot cable adds to a their clock box BEFORE getting into an argument with me about internal vs. external clocking. Even if my other un answered arguments are wrong (they are not wrong), that one by itself needs a TECHNICAL answer. They chose a marketing answer – ultra low jitter cable.

If Max is wrong and cables do not make jitter (and that is the case), then we have a company that claims technology leadership but is not a technology leader. I used to tolerate it, until Max and Lucas came here storming with impossible claims, trying to “create impressions” that they represent the 21st century and I should be disregarded as a 20th century type. I defended my position well. I was forced to back Apogee “up against the wall”. It was about telling people the best way to use their gear.

It is not about taking a single phrase to a product. It is about technical leadership vs. the claim for technical leadership. It is about how marketing leadership is willing to step all over what is technically correct, disregarding what is really good for the industry and the customers. And Goliath can and does get away with it. With huge exposure, advertising resources, claims to be the best in the world in print all over, with year after year tech awards that are about name recognition and do relate to advertising money, not technology.

I am not a King Kong with overseas manufacturing. We make everything in the U.S. I take time to help uplift the industry by running a forum and Apogee had to come here and try to muck with my honest efforts, stopping at nothing. Did you see what Max wrote about me at Fletcher site? He was not at Apogee when it was in start up mode and his facts are all screwed up.  

But they are losing that argument because they are technically so very wrong! They may continue to sell a lot of gear but they lose the technical arguments here. I have to assume the reason they sent you here is because no one with technical chops would back them up. So why are you here? I call it Chutzpa.

Your tone is inappropriate and disrespectful. Such technical people and better people do exist, you are not one of them. I already saw you are not up to par technically to come up against me. You argued and made some terrible mistakes on this forum, don’t you recall. struggling with the issue of relating dynamic range to bits, which was remarkable for someone writing a book about the subject. There was that idea of common clock for AD and DA to cancel jitter, with you suggesting a flash audio AD and DA which was extremely indicative of your lack of hands on and any design experience. There were some comments about filters that left a lot to be desired…
.
As far as I am concerned you are a recent graduate with no design experience and you share something in common with Apogee. Commercial ties? Maybe so, maybe not so. But you both are into creating the impression of technical leadership. They are not talking or answering anyone’s questions.

Once again you are taking too much of my time. I cannot accept you coming in here as if you were some technical guru, splitting hair about words, and misstating facts. I certainly did not agree that cables induce jitter.


Dan Lavry
Lavry Engineering
www.lavryengineering.com
Title: Re: Proper word clock implementation
Post by: Fletcher on March 25, 2005, 02:41:49 pm
danlavry wrote on Fri, 25 March 2005 13:04



You just pointed out another error in their marketing. They are selling 110 Ohms cable under their WORD CLOCK.  

     http://www.digitalriver.com/dr/v2/ec_MAIN.Entry10?xid=28122& amp; amp; amp; amp; amp;      amp;PN=1&SP=10023&V1=331437&DSP=&CUR=840& ;amp ;amp ;amp ;amp ;amp ;PGRP=0&CACHE_ID=0

So is that another BIG ERROR in the marketing department? Or are they selling 110 Ohm cables instead of 75Ohms. to people that use 75Ohm termination?


Apogee may very well be full of shit... and none of us should ever stop making sure that each and every manufacturer [be that manufacturer Apogee, Lavry, Digi-Design, Studio Projects, or Behringer] can support each and every claim they make 100% with scientific evidence.  

Subjectivity aside [we are in ways a community of artists that employ technology for artistic purpose] anything less than 100% supportable/factual claims must be challenged at all expense, within reason... with that in mind I quote the immortal words of Albert Einstein [who I'm told was a hell of guitar player] "As far as the laws of mathematics refer to reality, they are not certain; and as far as they are certain, they do not refer to reality."

In this [one] particular case Apogee isn't full of shit [doesn't mean they're not somewhere else, just not in this case], the link to the store on Apogee's site is general description of all their cable products, not any one specific cable product.

"Digital River" is what's known as a "fulfillment house".  A fulfillment house, in this case "Digital River" hasn't the slightest remote clue of anything technical.  The purpose of this fulfillment house is to warehouse hardware,offer it for sale via the internet, take orders, process those orders and ship those orders.  For all anyone at "Digital River" knows or cares these things could be TV's, Ginsu Knives, or talking bears.

Why Apogee has chosen to employ this fulfillment house to hawk their wares is a mystery... and personally I think it's rather unnecesary [I would think that now wouldn't I Twisted Evil ] and frankly, counterproductive... but it's not my place to tell them what to do.

The "official" Apogee cable description page http://www.apogeedigital.com/products/wydeeyecables.php does indeed have the correct product descriptions and designations.

At this point, as one of the administrators here at R/E/P, I have to say that this 14 page thread has run it's course.  All threads over time and multiple pages will tend to mutate from the original call and response, this one has done just that.

At this time, this thread will be locked.

I wholeheartedly encourage the various participants of this thread to begin new "sub threads" that focus specifically on the myriad of concepts and concerns that began and were continued through a couple hundred posts scattered over 14 pages.

As is the charter for Dan's forum, subsequent threads that will grow from this thread will be based firmly in science, and that all statements made in these subsequent threads can be backed up with empirical data.

On behalf of the staff and management of R/E/P I would like to thank each and every participant for their time and knowledge in these areas, and sincerely hope to see several productive threads spawned from the seed planted here.

Peace.



Title: Re: Proper word clock implementation
Post by: Sin x/x on August 21, 2006, 11:50:54 am
It's very simple to me:

A properly engineered converter will sound the same, no matter what cable or master clock you use.
It should accept data from the serial input signal and regenerate its own clock.
Audible differences simply mean poor design.
Things like that simply don't belong in a professional environment.

The questions about cable jitter are therefore total nonsense.

Keep up the fight Mr. Lavry.
Title: Re: Proper word clock implementation
Post by: Fletcher on August 21, 2006, 07:09:50 pm
Sin x/x wrote on Mon, 21 August 2006 11:50

A properly engineered converter will sound the same, no matter what cable or master clock you use.


Yes, a properly engineered converter will sound like a properly engineered converter, the sound of a converter is very much dependent upon that converter's clock, and jitter can indeed be induced by cabling.

An external master clock can make an inferior converter design [see RME for details] sound better.  An external master clock can make an entire system sound different [see your sense of aesthetic for "better or worse"].

I put a Big Ben on my RADAR V system.  I thought it fucked up the sound.  I put the RADAR V clock on my Lavry "blue series" converters [which I use for most mixes] and thought it fucked up the sound.  We use Apogee AD and DA-16x devices on our DAW... which have a similar clock to the Big Ben.  We clocked those units from both the RADAR V and the Lavry "Blue Series" [at different times of course] and found those clocks fucked up the sound [in our opinion].  However, all of these different converters and different clocks were different sounding... there were differences within our own camp as to which sounded better and which sounded worse.  Each of our staff has their own opinion.

My personal opinion is to use each converter working on it's own internal clock, but YMMV.

Peace.
Title: Re: Proper word clock implementation
Post by: Sin x/x on August 22, 2006, 01:37:54 am
A properly engineered DA-converter should have at least 2 separate pll's.

One for the FM data separator. This pll is much to jittery to drive the DAC directly, but unfortunately most manufacturers use just this one. And now you will hear cable and masterclock differences.

The other PLL (or a couple in series) you need to drive the DAC itself. This PLL gets fed by the PLL of the FM data separator. This PLL should have high damping and regenerates the clock signal without any jitter.

If this is the case jitter has no effect on the analog audio signal.

The jitter of the AES/EBU signal should be of cause within certain limits. Otherwise the first PLL can't lock. But if the jitter is within these limits and you have a properly engineered converter, everything is fine: No jitter at all.

Its analog to error correction, If the errors are within certain limits you can retrieve the original signal.
Title: Re: Proper word clock implementation
Post by: danlavry on November 16, 2006, 01:04:57 pm
The "Proper word clock implementation" thread is back after a year and a half.




Regrads
Dan Lavry
http://www.lavryengineering.com
Title: Re: Proper word clock implementation
Post by: crm0922 on November 20, 2006, 02:45:26 am
Maybe Apogee will come back and answer the (still unanswered) questions posed to them:

Quote:

I. QUESTION: WHICH SOLUTION PROVIDES LESS JITTER AT THE AD LOCATION?

1. A REASONABLE FIXED CRYSTAL
2. BIG BEN DRIVING A CLOCK INTO A CHASSIS EQUIPPED WITH A PLL VIA A 10 FOOT CABLE?

II. QUESTION: CAN ONE EVER CANCEL OR REDUCE A RANDOM NOISE AT ONE END OF A CABLE WITHOUT KNOWING WHAT KIND OF JITTER ACTIVITY IS OCCURRING AT THE OTHER END?
(can one cancel, remove or reduce the receiver and PLL noise, some of it unknown, some totally random, by driving it with a clean low jitter clock? Driving it with any clock?)


Can't wait.

Chris
Title: lavry
Post by: Lucas van der Mee on November 21, 2006, 01:31:57 pm
The answers to those questions we have given several times before in this thread. But to make it easy for you and not have to search for it I’ll say it again:

Jitter at the converter chip will be lower, in most cases, when clocked to internal. There is no doubt about that, no mystery, we never claimed anything else. However, we do say a number of things that are beyond this question.
1.   Big Ben is an excellent solution if you need a master clock. It is low jitter and offers a lot of features the competition does not have.
2.   If you need to clock to a jittery source, for instance a SPDIF output of a computer or a commercial cd player, having the Big Ben re-clock the data first, will very often improve the sonic and MEASURABLE quality of the converter used. The amount of improvement is all dependent of the quality of the PLL circuit of said converter. In other words, the better the PLL, the least improvement will be noted.
3.   For this one I will be quoting Max, because he put it so well:

“Yes, Jitter is a deficiency that causes a certain type of distortion to the signal. However, the ear, being a non-linear device with respect to frequency, will not hear all jitter the same. It is very easy to create a scenario whereby two signals can be induced with jitter, one of which has significantly more jitter than the other, but sounds audibly more ACCURATE to the ear. The distortion due to jitter on the one with more jitter can be relegated to frequencies that are entirely out of the human audible spectrum in various ways, whereas the other can have far less jitter, but at frequencies that are much more audibly apparent.

It is certainly possible to design an external clock in such a way that the increased amount of jitter that results will have less of an audible impact than the internal clock by itself. For this reason one cannot claim that an internal clock will always be more accurate - merely that it will theoretically always have less jitter. There is a significant difference between the two, and the testimony we have been seeing on the market and in our own testing with respect to Big Ben over the last few years certainly attests to this.”

I would like to add to this: There are numerous examples in technology, where we find that an actual reduction in performance of one parameter can lead to a perceived and sometimes actual improvement, in others. A good example in audio is dither, (noise shaped especially). By adding dither we actually reduce the dynamic range, yet we improve the perceived noise floor by masking the annoying noise of rounding errors.

My last word on this topic is, people do perceive more than science can explain. Our senses are intrinsically subjective, however the more experienced we become, the more we train our perception, the more OBJECTIVE we can become and obviously, your mileage may vary.

Lucas van der Mee
Sr. Design Engineer
Apogee Electronics