Print

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

[Fletcher]

We've been doing quite a bit of chatting about work clocks and converters over beers on the loading dock lately... which caused me to go back an read a good bit of Dan Lavry's 14 page thread on word clocks during his residency here from November 2005-November 2006.

In re-reading a majority of that thread, and after hearing a good bit of new [and some pretty exciting convterts... like the JCF and the Burl]... along with the new Antelop "Atomic Clock"... it seems to me that Dan's premise is still a crock of shit.

I have to wonder if he actually listens to music or is just reading measurments?

We hooked up the "Atomic Clock" to our RADAR V [Nyquist]... and I'll be damned if it didn't make things sound "better"... cymbals really sounded like cymbals [my long standing biggest bitch with digital audio]... the "soundfield" seemed even larger [which is pretty amazing because the RADAR V is still the "largest" sounding multi-track converter/storage system I've ever experienced]... and then there is the JCF 8 channel tube D/A converter [which is based on Ampex 351 tube electronics]... it's simply remarkable sounding... and the Burl B-2 A/D converter, which while different than the sound of RADAR's converters is still one of the best sounding unit's I've met.

I guess my point is that science is science and no one can deny the "scientific fact"... but what we work with is music... and music is art, art is subjective and scientific reason does not necessarily prevail in what we hear and feel [on an emotional level... which is where music is supposed to live].  Yes, there are scientific reasons for the things we find appealing and unappealing in any given unit... but the fact of the matter is that science is second to the emotion of music which appears to have escaped Mr. Lavry.

Anyone else observed anything like this?

[beau]

well, i have been doing some research on converters and clock lately too... after switching from the digi 192 to the lynx aurora 16, months later, i have been doing some research on clocks... i finally settled on the big ben... then yesterday, i brought my converters and clock to a friends place, where he had the antelope clock.  and we did a not very scientific listening test... and here is what i found..  i think that converters and clocks are almost just like another studio tool...  and you need to learn how to use all of your tools to suit your needs...  i felt like the antelope clock was clearer, and the cymbals were nicer, like you said... smooth, prestine, but not harsh.... the bottom end was tighter as well... but.... personally, i liked the vibe of the big ben better... it just has a bit of a looser feel to my ears.  so i guess i am saying that i agree with you...  we are working with an art, and so much of it is based on taste and preference...  and i heard a quote by Quincy Jones, i believe, where he says. "a great song, great performance by great players will transend everything else in the chain."  i am pretty sure i just slaughtered that quote, but it really hits home with me... and keeps me focused on the music, rather than all of the toys and technical data about them, that we get caught up with..

peace

beau

[PookyNMR]

You're right science is science.  Science tells us that a well designed internal clock can produce lower jitter than any externally linked clock.

You're right subjective listening experience is up to the users ears.  We may prefer a number of different sounds, different types of and amounts of jitter, distortion, etc.  And our preferences will be driven by our limited perception of transparent replication, even if they don't measure as well with a number of the common scientific tests.  And in the end our satisfaction with what our ears tell us is the most transparent replication is what matters.  Measurements are only one tool in the tool box.

You and I may indeed prefer the sound of converters externally clocked to other devices.  Does that mean that science is wrong about an internal clock being inherently more accurate than an external one?  No.  It just means that we subjectively prefer something else.

I know you don't like Dan, but that doesn't mean that his statement about internal vs. external clock accuracy was incorrect.  It just means that the implications of it may not be as black and white as he suggests.

Peace.

[Tomas Danko]

PookyNMR wrote on Mon, 02 June 2008 17:54

You're right science is science.  Science tells us that a well designed internal clock can produce lower jitter than any externally linked clock. You're right subjective listening experience is up to the users ears.  We may prefer a number of different sounds, different types of and amounts of jitter, distortion, etc.  And our preferences will be driven by our limited perception of transparent replication, even if they don't measure as well with a number of the common scientific tests.  And in the end our satisfaction with what our ears tell us is the most transparent replication is what matters.  Measurements are only one tool in the tool box. You and I may indeed prefer the sound of converters externally clocked to other devices.  Does that mean that science is wrong about an internal clock being inherently more accurate than an external one?  No.  It just means that we subjectively prefer something else. I know you don't like Dan, but that doesn't mean that his statement about internal vs. external clock accuracy was incorrect.  It just means that the implications of it may not be as black and white as he suggests. Peace.

Very well said, Nathan.

There is no denying the facts Mr. Lavry is pointing out from a technical point of view, as well as there is equally no denying that if it sounds better it sounds better.

If external esoteric clocking sounds better, then use it. Until someone figures out why this is so. Meanwhile you've been able to record shitloads of kickass albums.

FWIW I once implied to Apogee if the idea behind the Big Ben is some form of noise shaping within the time domain but that question was met with silence.

More noise can sound better to the ear than less noise, depending on where it is within the frequency domain.

Less mid range distortion can apparently sound worse to some ears, compared to more distortion distributed equally across the audio band.

We often prefer some types of distortion to clean and accurate reproduction.

But more beer is always better, so no sleepless nights needed!

Oh, and beer is also a form of pleasant (to some of us) distortion. So there you go. YMMV.

[Kendrix]

In my experiments with Clocks Ive found fairly subtle differences between internal and external clock sources on my on-board Yammy converters.

While difference were there I couldnt really say one clock was significantly better than the other.    

However, i ve always wondered if the architecture of the clocking on the Yammy (with optional I/O cards and a total 24 converters involved) really matches Dans definition of an "internal" clock configuration.

[Etch-A-Sketch]

Fletcher wrote on Mon, 02 June 2008 05:27

We've been doing quite a bit of chatting about work clocks and converters over beers on the loading dock lately... which caused me to go back an read a good bit of Dan Lavry's 14 page thread on word clocks during his residency here from November 2005-November 2006. In re-reading a majority of that thread, and after hearing a good bit of new [and some pretty exciting convterts... like the JCF and the Burl]... along with the new Antelop "Atomic Clock"... it seems to me that Dan's premise is still a crock of shit. I have to wonder if he actually listens to music or is just reading measurments? We hooked up the "Atomic Clock" to our RADAR V [Nyquist]... and I'll be damned if it didn't make things sound "better"... cymbals really sounded like cymbals [my long standing biggest bitch with digital audio]... the "soundfield" seemed even larger [which is pretty amazing because the RADAR V is still the "largest" sounding multi-track converter/storage system I've ever experienced]... and then there is the JCF 8 channel tube D/A converter [which is based on Ampex 351 tube electronics]... it's simply remarkable sounding... and the Burl B-2 A/D converter, which while different than the sound of RADAR's converters is still one of the best sounding unit's I've met. I guess my point is that science is science and no one can deny the "scientific fact"... but what we work with is music... and music is art, art is subjective and scientific reason does not necessarily prevail in what we hear and feel [on an emotional level... which is where music is supposed to live].  Yes, there are scientific reasons for the things we find appealing and unappealing in any given unit... but the fact of the matter is that science is second to the emotion of music which appears to have escaped Mr. Lavry. Anyone else observed anything like this?

All of the subjective opinions are pretty useless unless you do a truly blind A/B/X comparison, AND consistently identify which sound is which more than around 55% of the time. Otherwise, it's just your biases and opinions that are making the decision for you.

[Halfway Competent]

Etch-A-Sketch wrote on Mon, 02 June 2008 12:43

All of the subjective opinions are pretty useless unless you do a truly blind A/B/X comparison, AND consistently identify which sound is which more than around 55% of the time. Otherwise, it's just your biases and opinions that are making the decision for you.

I was once sitting behind the board and thought the bass track needed a bit of a low-freqency bump.  So I slowly tweaked up the low shelf knob until it sounded sufficiently bassy.  A bit later, I found the bass track again sounding lean, so I tweaked a bit more...  And a bit more...  And was wondering why it wasn't making as big of a difference as it should have been.  The reason, of course, is that the EQ on that channel was disabled.  

[Fletcher]

Isn't that what all music is based upon?  Our "biases and opinions"?  I'm all for the "advancement of science" but not at the cost of what I subjectively feel best suits the music on which I'm working.

A/B/X tests are all well and good if you believe there is a premise called "best", I don't believe in a "best" anymore than I believe in Santa Claus or the Easter Bunny... but I am a firm and ardent believer in what's "best" for a given project at a given time.

I think converters can be best equated to tape formulations... for years we listened to different tape formulations to determine which we felt was "best" for a project... then we stressed about minutia like "bias" and level at which we hit that tape... there are an equal number of variables in the conveter debate... the question is still, which best suits your sense of aesthetic vs. which is technically "best".

[tom eaton]

I'm pretty sure Dan never told anyone what sounded better.  He just explained the math.  Neither he, nor the math and physics involved, are "full of shit" as far as I can see.

He's not here to defend himself, or course, so allow me.

Dan works very hard to make serious gear that many pros use day in and day out to do serious work.  

Just like Fletcher he pulls no punches and calls them as he sees them.

Unlike Fletcher and most of us, he doesn't RECORD for a living, so he lives and dies by the math, whereas most of us only care what it sounds like in the end.

-tom

[bruno putzeys]

I still see a lot of half baked arguments floating around.

"Internal clocking will always have lower jitter than external clocking."
Factually wrong. Below the corner frequency, a good PLL duplicates the phase noise spectrum of the source. If that source happens to be cleaner than the internal clock, you get lower jitter below the PLL corner frequency. PLL's used in run-of-the mill converters have corner frequencies in the kHz region, and clocking them externally can improve results.

"External clocks shape the jitter to sound nice".
Wrong. This hypothesis has been proffered by people who believe that internal clocking must necessarily be technically better. As per the above, that's not quite the case so we don't need to postulate such a hypothesis to explain the state of affairs.

"Dan L. says internal clocking is always better".
That's certainly a misrepresentation. Dan's take, as I understand it, is "why on earth would anyone want to stick with a converter that performs worse on its internal clock?". He'd say if your converter improves with external clocking, get a better converter instead of buying an external clock (i.e. don't throw good money after bad).
I'd say: it simply depends on which option is cheaper. If you already have a 48 channel PT system, it stands to reason that a 1950 euro clock from a company I shan't name is a good investment. It's technical idealism vs economic reality.

Having said that, Dan has this propensity to take perfectly valid technical insights and then heavily argue oversimplified versions of them. On one occasion I fell out with him in a discussion about dither where he stated that 4 bits worth of dither is perfect. His actual point was that 4 bits is probably the point of diminishing returns, but he told it as if it were an immutable truth. I think that a statement like "internal clocking is always best" fits well with this experience.

Anyhow, coming back to clocks. No manufacturer of converters should be selling external clocks for the purpose of improving the sound quality of their own converters. However, that should not stop anyone from making clocks to improve other people's gear, using technical back doors in same (such as fast PLL's in the external clock input).

[Etch-A-Sketch]

Fletcher wrote on Mon, 02 June 2008 20:10

Isn't that what all music is based upon?  Our "biases and opinions"?  I'm all for the "advancement of science" but not at the cost of what I subjectively feel best suits the music on which I'm working. A/B/X tests are all well and good if you believe there is a premise called "best", I don't believe in a "best" anymore than I believe in Santa Claus or the Easter Bunny... but I am a firm and ardent believer in what's "best" for a given project at a given time. I think converters can be best equated to tape formulations... for years we listened to different tape formulations to determine which we felt was "best" for a project... then we stressed about minutia like "bias" and level at which we hit that tape... there are an equal number of variables in the conveter debate... the question is still, which best suits your sense of aesthetic vs. which is technically "best".

Who said anything about which is technically "best".  All I'm saying is, for YOU to say that YOU like something better... then you should CONSISTENTLY like that one thing better every time you hear it.  AND you should be able to identify it as sounding "better" to you EVERY time you hear it.  This has absolutely nothing to do with the advancement of science or what is empirically "best" at all.

All I'm pointing out is that you saying "...and I'll be damned if it didn't make things sound 'better'..." is the "crock of shit" that you are referring to if you can't hear the "better" sound consistently even if you dont' know which is which.  

Otherwise you could very well be giving in to your biases from sales and marketing pitches that have been thrown your way over the years.

If you said, "I had my assistant flipping and I honestly wasn't looking... and 9 times out of 10 I could hear when it was the atomic clock and identified it correctly..."  Then I would say you might have an argument against Dave's previous 14 page thread...

See my point?

[Fletcher]

I suppose we could run a "test" with someone flipping from clock to clock and hear what we get... most of the time what I've heard hasn't been subtle so it should be an easy test to take/recreate.

My general problem with A/B/X tests is that the brain becomes confused... I've found quick "A/B" [or A/B times 2 or 3] is a far more relevant test that doing something 10 times.  I sat in a "Digi-192 vs. RADAR" A/B/X test once... I got it right the first 4 times and after that it became a total crap shoot as my brain was simply confused [fatigued?].

From my perspective, I work with music... I make a LOT of decisions about that music in very short intervals [see signature file for details]... sometimes it's just a gut aesthetic decision... and more often than not [but not always] it's the right decision as it applies to my and the artist's sense of aesthetic.

Peace.

[Eliott James]

So what brand or model number converter has a clock so good that an external clock would not make a difference?

[tom eaton]

An external clock will ALWAYS make a difference... that's kind of the problem.  Whether you like the sound of any device clocked internally versus externally is up to you.

I would guess that Dan would say his converters perform worse on external clock, but may sound better to you.

I have the HEDD192, an Apogee AD16x (same clock as BigBen), a RADAR and an Isochrone OCX, and any one of them as clock master makes my MOTU1296s sound better.    I'm not sure that I've heard the HEDD do better on an external clock... if that answers the question to some degree!

-tom

[Chad Sims]

Wow that JCF stuff looks amazing.  I have been wondering why somebody hasn't put a bunch of brawny transformers and tubes into ADCs and DACs.  So how does this stuff sound?  It seems like it should sound more like a tape machine.  What difference did the clock seem to make on these units?