R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 ... 3 4 [5] 6 7 ... 15   Go Down

Author Topic: Proper word clock implementation  (Read 132997 times)

chap

  • Newbie
  • *
  • Offline Offline
  • Posts: 16
Re: Proper word clock implementation
« Reply #60 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
Logged

bobkatz

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2926
Re: Proper word clock implementation
« Reply #61 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.
Logged
There are two kinds of fools,
One says-this is old and therefore good.
The other says-this is new and therefore better."

No trees were killed in the sending of this message. However a large number of
electrons were terribly inconvenienced.

Albert

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 56
Re: Proper word clock implementation
« Reply #62 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.
Logged

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: Proper word clock implementation
« Reply #63 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
Logged

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: Proper word clock implementation
« Reply #64 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
Logged

Roland Storch

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 406
Re: Proper word clock implementation
« Reply #65 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.
Logged

bobkatz

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 2926
Re: Proper word clock implementation
« Reply #66 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.
Logged
There are two kinds of fools,
One says-this is old and therefore good.
The other says-this is new and therefore better."

No trees were killed in the sending of this message. However a large number of
electrons were terribly inconvenienced.

Albert

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 56
Re: Proper word clock implementation
« Reply #67 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.
Logged

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: Proper word clock implementation
« Reply #68 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
Logged

Roland Storch

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 406
Re: Proper word clock implementation
« Reply #69 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.
Logged

Max

  • Newbie
  • *
  • Offline Offline
  • Posts: 45
Re: Proper word clock implementation
« Reply #70 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.
Logged
Max Gutnik
Apogee Electronics

danlavry

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 997
Re: Proper word clock implementation
« Reply #71 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
 

Logged

chap

  • Newbie
  • *
  • Offline Offline
  • Posts: 16
Re: Proper word clock implementation
« Reply #72 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
Logged

Bill B

  • Full Member
  • ***
  • Offline Offline
  • Posts: 139
Re: Proper word clock implementation
« Reply #73 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.


Logged
BB

Lucas van der Mee

  • Newbie
  • *
  • Offline Offline
  • Posts: 12
Re: Proper word clock implementation
« Reply #74 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
Logged
Lucas van der Mee
Sr. Design Engineer
Apogee Electronics
Pages: 1 ... 3 4 [5] 6 7 ... 15   Go Up