Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Ashermusic]

I love to visit this forum even though much of the discussion is technically over my head. But man, it is confusing!

I recognize that this is all subjective. Still....

On one hand we have folks Dan Lavry and Bob Katz,  both highly respected, saying that even a mediocre internal clock (like I assume is in my Echo Layla 24) will out perform an external clock like Isocrone or Big Ben because of the laws of physics.

On the other hand we have highly respected people like Fletcher@ Mercenary audio, Marcel @ audiomidi.com and my own engineer who tell me that in their tests they hear a noticable improvement with these clocks.

So for a guy like me who would like to improve his D/A at a low cost I am torn between saviing for Lavry Blues or awaiting the new product; getting respected converters by either Mytek, Lynx or Benchmark which no doubt should be a substantial improvement over my Layla; getting an interface like RME Fireface  or Aurora 8 which have better convertors supposedly; or get a clock like the above mentioned.

On the other hand my friend,  a former president of AES who records some famous jazz musicians using an Echo Layla (and his recordings sound pristine) says to me, "Jay are you unhappy with how your recordings translate when you play them back on another system? Do your clients complain about your sound quality? If it isn't broken, why do you want to spend money to fix it?"

Tis a puzzlement.

[crm0922]

Dan, maybe you should post a sticky FAQ on this forum that explains the reasoning behind the argument that external converters cannot improve jitter performance?  It seems like a lot of people are coming here looking for a reason not to buy an external clock for "better sound".

Save your money for the high end stuff, I say.  I have a Layla24 kicking around somewhere, and, as I recall, it sounded quite nice.  I used it to mix to a few times, and had little to complain about, for the price, anyways.

A lot of improvement can be made in other parts of the recording chain, are you maxed out every where else?

Chris

[sui-city]

IMVHO, I think the turning point for me was the realizing the importance of the question: "What do you mean by better?"

Dan Lavry and Bob Katz, as you state, make it quite clear that from the point of view of the laws of physics, it is not possible for an external clock to improve the performance of even a mediocre internal clock. This is obviously absolutely correct.

On the other hand you have users who have great ears telling us it sounds better. So, "What do you mean by better?".

Just as Dan Lavry has said on a number of occasions, he does not question what these people are hearing. As I understand it, he is questioning the perception from a technical and scientific level. Since the physics clearly show how it is highly unlikely, we therefore need to spend more time understanding what it is that is in fact improving the sound.

I have reached a point where I am extremely suspect of someone who tells me that their converters make something sound so much more "open". This to me is not a clear statement. IMNSHO, converters are meant to convert analog information to digital data as accurately as possible, nothing else. And if the laws of physics state that an external clock cannot improve the performance of a mediocre internal clock and upwards, then before I spend money on an external clock, I think some scientific research needs to be done to truly understand why the people with the ears feel that it improves performance.

[Ronny]

sui-city wrote on Wed, 14 September 2005 17:22

IMVHO, I think the turning point for me was the realizing the importance of the question: "What do you mean by better?" Dan Lavry and Bob Katz, as you state, make it quite clear that from the point of view of the laws of physics, it is not possible for an external clock to improve the performance of even a mediocre internal clock. This is obviously absolutely correct. On the other hand you have users who have great ears telling us it sounds better. So, "What do you mean by better?". Just as Dan Lavry has said on a number of occasions, he does not question what these people are hearing. As I understand it, he is questioning the perception from a technical and scientific level. Since the physics clearly show how it is highly unlikely, we therefore need to spend more time understanding what it is that is in fact improving the sound. I have reached a point where I am extremely suspect of someone who tells me that their converters make something sound so much more "open". This to me is not a clear statement. IMNSHO, converters are meant to convert analog information to digital data as accurately as possible, nothing else. And if the laws of physics state that an external clock cannot improve the performance of a mediocre internal clock and upwards, then before I spend money on an external clock, I think some scientific research needs to be done to truly understand why the people with the ears feel that it improves performance.

If you want to find the answer to these questions, put more money into studying the auditory cortex. We know little about it other than it's not 100% accurate to determine sound differences when tests are not blind and double blind.

[sui-city]

Ronny,

Agreed.

I have a bunch of friends doing their doctorates in neuro-pyschology. The stuff they tell me is incredible. If our brains can completely alter things we see, and generate voices that are not there, then the capacity to believe something because we choose to is going to get in the way of a non- blind/double-blind test.

This is why for me at the moment, when it comes to converters, I'll take the maths over the subjective perception.

[Ashermusic]

Once again, my quandary comes into play. I don't give a flying you know what as to why something sounds "better". We all have had the experience of hearing a piece of gear that we believe sounds better than some that specs out better. Nonetheless, I have enough belief in the laws of science that when Dan says that the physics of external clock vs. internal clock determines jitter and I know that jitter cannot possibly addd anything aesthetically desirable to a sound to be confused.

I had an interesting discussion with someone yesterday who told me that he simply preferred the sound in an A/B test of the RME Fireface's convertors to some that he knew were technically better convertors. He found  their sound "sterile"  

Let's face it, if science were all that came in to play then theoretically everyone with decent ears would pick the most accurate convertors in a blind test versus any inferior ones or any with an external clock and we know that this is simply not so.

Hopefully next week with the help of a store here I am going to hook up a couple of  these externals to my Layla as well as a Fireface and compare for myself. I  will report back.

[bobkatz]

sui-city wrote on Wed, 14 September 2005 17:22

IMVHO, I think the turning point for me was the realizing the importance of the question: "What do you mean by better?" Dan Lavry and Bob Katz, as you state, make it quite clear that from the point of view of the laws of physics, it is not possible for an external clock to improve the performance of even a mediocre internal clock. This is obviously absolutely correct.

Well, I think it is the definition of "mediocre" that we'd have to define. Obviously, if you put in a 25 cent jittery crystal oscillator as an afterthought, then it is conceivable that even a crappy PLL will do better. But what a stupid way to run a company...



BK

[sui-city]

bobkatz wrote on Thu, 15 September 2005 17:35

sui-city wrote on Wed, 14 September 2005 17:22 IMVHO, I think the turning point for me was the realizing the importance of the question: "What do you mean by better?" Dan Lavry and Bob Katz, as you state, make it quite clear that from the point of view of the laws of physics, it is not possible for an external clock to improve the performance of even a mediocre internal clock. This is obviously absolutely correct. Well, I think it is the definition of "mediocre" that we'd have to define. Obviously, if you put in a 25 cent jittery crystal oscillator as an afterthought, then it is conceivable that even a crappy PLL will do better. But what a stupid way to run a company... BK

Bob,

Have you got any examples of businesses run like this? A yes or no will suffice if you don't wish to name names.

Asher,

That is why for now I will rather opt for the science. I won't take the risk on subjective perspective on something like converter at this point.

Pres and mics are a different thing.

[bobkatz]

sui-city wrote on Thu, 15 September 2005 16:54

Well, I think it is the definition of "mediocre" that we'd have to define. Obviously, if you put in a 25 cent jittery crystal oscillator as an afterthought, then it is conceivable that even a crappy PLL will do better. But what a stupid way to run a company...



BK[/quote]


Bob,

Have you got any examples of businesses run like this? A yes or no will suffice if you don't wish to name names.


[/quote]


It's not as bad as I describe. It's mostly ignorance and lack of test equipment or PLL or oscillator expertise on the part of the designers. Coupled with pinching pennies. In essence, consider that a single 2 channel converter of superior quality can cost 3 to 4 times as much as an 8 or 16 channel converter of "medium quality" and you'll see where the money goes.

I get people who write me that their brand x 16 channel converter for which they paid $800 to $1000 sounds better on external clock. OK, so then the only conclusion I can make is that the internal clock was designed too cheaply. The converter is what I call "defective". But that is not the same as saying, "you got good value for your money." Maybe you did  . The sound quality of even a "medium level" 16 channel converter made today is often better than that of certain more expensive 16 channel converters made 8 to 10 years ago. Simply because the parts are better.

But at $1000 for 16 channels, even $2000 you might expect a compromise somewhere, eh? Or be pleasantly surprised. I have not tested many, no one pays me to test these things  

[Bogic Petrovic]

bobkatz wrote on Sat, 17 September 2005 15:39

sui-city wrote on Thu, 15 September 2005 16:54 Well, I think it is the definition of "mediocre" that we'd have to define. Obviously, if you put in a 25 cent jittery crystal oscillator as an afterthought, then it is conceivable that even a crappy PLL will do better. But what a stupid way to run a company... BK

Quote:

Quote: Bob, Have you got any examples of businesses run like this? A yes or no will suffice if you don't wish to name names. It's not as bad as I describe. It's mostly ignorance and lack of test equipment or PLL or oscillator expertise on the part of the designers. Coupled with pinching pennies.

This isn't good, if it's true...

1. If I remember, some of jitter/wander measurement instruments costs are greatly lowered, but it is still very expensive even today... look at:
http://www.sunrisetelecom.com/pages/jitter.shtml
http://www.acterna.com/global/index.html (ex Wandel&Goltermann)
http://www.agilent.com/  (ex. Hewlett Packard)
This companies have some history in jitter/wander measurements, especially last two...
Some of their instruments _must_ be used if any digital audio equipment manufacturer tend to be authoritative.

2. Manufacturer can use their own, already properly tested, old designs copied and pasted in new design, but this does not means that new design is already jitter/wander free... Manufacturer must have possibility to check design again and again... there is too much jitter/wander sources than we can expected (PCB design may have great influence for example)

3. PLLs in digital audio has very sharp requirements if we look at AES recommendation named as AES11 (look at http://www.aes.org/publications/standards/)
I don't have idea how it's possible to make some 1000$ AtoD audio equipment if this is a comparable to price of one single good (but not the best!) TCVCXO unit (Temperature Compensated Voltage Controlled Crystall Oscillator)

Quote:

In essence, consider that a single 2 channel converter of superior quality can cost 3 to 4 times as much as an 8 or 16 channel converter of "medium quality" and you'll see where the money goes. I get people who write me that their brand x 16 channel converter for which they paid $800 to $1000 sounds better on external clock. OK, so then the only conclusion I can make is that the internal clock was designed too cheaply. The converter is what I call "defective". But that is not the same as saying, "you got good value for your money." Maybe you did  . The sound quality of even a "medium level" 16 channel converter made today is often better than that of certain more expensive 16 channel converters made 8 to 10 years ago. Simply because the parts are better.

Probably only analog audio can have "good value for money" scale...
If digital audio devices aren't so good, this is not that inaudible or tolerable.

And yes, new component are almost always better and cheaper than  oldest one... this is wonderfull Engineers always has new tasks (jobs!)

Quote:

-snip- I have not tested many, no one pays me to test these things

Why?


Best regards,

-boggy

[danickstr]

two thoughts -

layla is also your adc? then it is doing two things that affect sound...clock and conversion.  different things to listen for.

go to a reputable dealer where you buy yours stuff and ask them to let you listen to a high end adc/clock and decide for yourself.

good luck

[Ashermusic]

danickstr wrote on Sat, 24 September 2005 17:46

two thoughts - layla is also your adc? then it is doing two things that affect sound...clock and conversion.  different things to listen for. go to a reputable dealer where you buy yours stuff and ask them to let you listen to a high end adc/clock and decide for yourself. good luck

As I said in an earlier post, I  am setting  it up to  do just that comparing my Layla, PT HD2, and the RME Fireface all alone and then all with a Big Ben and an Isocrone. It will obviously be subjective but it will be me and three others whose ears I respect so if we all reach the same general conclusions that will be enough to satisfy my curioisty.

[Gunnar Hellquist]

To me the real test is that the makers of Big Ben and such never publishes any jitter measurements on equipment they supposedly improve, before and after.

If from some reason a typical unit x goes from, say, 10 to 5 in jitter by beeing connected to a Big Ben, they would surely publish such stuff. But there is nothing such published. They probably own the equipment to measure it already, and doing the tests should not take much time. Why do you think they do not publish their results?

[rankus]

Ashermusic wrote on Mon, 26 September 2005 08:18

As I said in an earlier post, I  am setting  it up to  do just that comparing my Layla, PT HD2, and the RME Fireface all alone and then all with a Big Ben and an Isocrone. It will obviously be subjective but it will be me and three others whose ears I respect so if we all reach the same general conclusions that will be enough to satisfy my curioisty.

My money is on the RME.... I run a Multiface.  It won out in a shootout against a company that begins with "A" at my stuio. (This was a very subjective test done by myself)

If you have the money I would suggest you look at the Multiface with their Octa Mic (or Quad Mic) front end....  This will give you more inputs (8-16 mic + 8 line analog + 2 ADAT + SPDIF) at slightly more than the Fireface that is "closed end"  with regards to upgrade paths....  The Multiface can handle 28 inputs total if I recall.... (plus 8 analog outs)

[Ashermusic]

rankus wrote on Tue, 04 October 2005 19:27

Ashermusic wrote on Mon, 26 September 2005 08:18 As I said in an earlier post, I  am setting  it up to  do just that comparing my Layla, PT HD2, and the RME Fireface all alone and then all with a Big Ben and an Isocrone. It will obviously be subjective but it will be me and three others whose ears I respect so if we all reach the same general conclusions that will be enough to satisfy my curioisty. My money is on the RME.... I run a Multiface.  It won out in a shootout against a company that begins with "A" at my stuio. (This was a very subjective test done by myself) If you have the money I would suggest you look at the Multiface with their Octa Mic (or Quad Mic) front end....  This will give you more inputs (8-16 mic + 8 line analog + 2 ADAT + SPDIF) at slightly more than the Fireface that is "closed end"  with regards to upgrade paths....  The Multiface can handle 28 inputs total if I recall.... (plus 8 analog outs)

Hi Rick and thanks for the response. When I  bought the Layla I also tried the Multiface and could not hear a difference so I went with the less expensive Layla.

RME claims however the the Fireface has superior convertors to the Multiface. I simply do not need that many ins and outs, 8x8 is  more than enough fpor me here as most of wehat I am doing are softsynths in the  box. When I add real players I go to a full on studio.