Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › December 2013 Mid-Level DAC Comparison
New Posts  All Forums:Forum Nav:

December 2013 Mid-Level DAC Comparison - Page 67  

post #991 of 1331

i have asked him and await his response but he is a very well known and trust advisor in the forums over there.  Here is our conversation about the DAC's as he leaves easy to understand information and links.

http://www.avsforum.com/t/1299898/dac-comparison/810#post_24147695

post #992 of 1331

Copied his response lol

 

"Credentials questions are just d%^k measuring contests decorated with old sheepskin. ;-) Usually nobody looks at your degree after your first job. They look at your last job instead.

Sure, I have a BS in engineering and did all work for a master's degree but one class, but that was about 3/4 of the way through the previous century. In those days an audio grade DAC cost about a quarter of a million dollars and filled a rack cabinet.

Two of the most informative things I did in this area were my www.pcavtech,com and www.pcabx.com web sites, now departed."

post #993 of 1331
Quote:
Originally Posted by DoctaCosmos View Post
 

i have asked him and await his response but he is a very well known and trust advisor in the forums over there.  Here is our conversation about the DAC's as he leaves easy to understand information and links.

http://www.avsforum.com/t/1299898/dac-comparison/810#post_24147695

 

Fair enough.

I think I would argue that a headphone amplifier is easier to design than a DAC, or easier to make perfect.........or should I say easier to design a good headphone amplifier than a good DAC.


Edited by Chris J - 1/3/14 at 3:05pm
post #994 of 1331
Quote:
Originally Posted by Chris J View Post
 

 

Fair enough.

I think I would argue that a headphone amplifier is easier to design than a DAC, or easier to make perfect.........or should I say easier to design a good headphone amplifier than a good DAC.

 

Y, we need a definition of 'perfect' DAC.

 

My 2 cents:

In the computer world, one of the best measure / goal of artificial intelligence is:       Turing test:   http://en.wikipedia.org/wiki/Turing_test

"The Turing test is a test of a machine's ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of a human."

Same can be applied to ADC and DAC.  The best ADC, DAC, and headphone system should make human unable to distinguish it is a live concert or a playback.

But such test is hard to do.

 

So an easier test will be:

1) Use DAC to play an audio file.  

2) Then use ADC to convert it back to an audio file.

3) go back to 1).   repeat the process for 'n' times.   E.g. n=10.

 

Compare the final audio file with original audio file.    Which DAC can give the smallest difference will be the winner.     They are the most 'transparent' ADC and DAC.


Edited by yfei - 1/3/14 at 3:38pm
post #995 of 1331
Quote:
Originally Posted by yfei View Post
 

 

Y, we need a definition of 'perfect' DAC.

 

My 2 cents:

In the computer world, one of the best measure / goal of artificial intelligence is:       Turing test:   http://en.wikipedia.org/wiki/Turing_test

"The Turing test is a test of a machine's ability to exhibit intelligent behavior equivalent to, or indistinguishable from, that of a human."

Same can be applied to ADC and DAC.  The best ADC, DAC, and headphone system should make human unable to distinguish it is a live concert or a playback.

But such test is hard to do.

 

So an easier test will be:

1) Use DAC to play an audio file.  

2) Then use ADC to convert it back to an audio file.

3) go back to 1).   repeat the process for 'n' times.   E.g. n=10.

 

Compare the final audio file with original audio file.    Which DAC can give the smallest difference will be the winner.     They are the most 'transparent' ADC and DAC.

 

A perfect DAC is not necessarily 'transparent', though.

There wouldn't be a marked for Audio-GD stuff if it were the case :D.

post #996 of 1331
Quote:
Originally Posted by yfei View Post
 

But such test is hard to do.

 

So an easier test will be:

1) Use DAC to play an audio file.  

2) Then use ADC to convert it back to an audio file.

3) go back to 1).   repeat the process for 'n' times.   E.g. n=10.

 

Compare the final audio file with original audio file.    Which DAC can give the smallest difference will be the winner.     They are the most 'transparent' ADC and DAC.

 

A fine idea - only one problem. The DAC could be perfect and it would still fail this test due to an imperfect ADC. There's no way to distinguish which, of the ADC or DAC, is causing the problem.

post #997 of 1331
Quote:
Originally Posted by Clemmaster View Post
 

 

A perfect DAC is not necessarily 'transparent', though.

There wouldn't be a marked for Audio-GD stuff if it were the case :D.

Or Audio Note DACs for that matter.  

post #998 of 1331
Quote:
Originally Posted by CybDev View Post

I find the conclusions Gary in MD made quite interesting, particularly compared to the ones from purrin's thread over at http://www.head-fi.org/t/693798/ranking-of-17-dacs-and-dac-configurations

[...]

 

Clearly, purrin and co. are shills!

 

Gary, curious what you make of people latching on to your findings and taking them as gospel. I've noted that at various times you've taken care to state disclaimers (your ears, your gear, your preferences/priorities etc) but it's largely been overlooked from my pov.

 

I did give some thought as to how I could reconcile my own experience upgrading dacs (times when I had various components on hand for direct comparison) and purrin's findings with your own polar opposite conclusions. I can understand purrrin's impressions and get a sense of components I've not heard and therefore derive some value from his posts. This is because I can calibrate against his take on the Gungnir, even though we differ on how we rate that component.

 

Your grouping of the Gungnir and M51 as indistinguishable is a mystery to me though. Again, I had both components on hand for weeks and also took great care to volume match. The differences weren't subtle to my ears, on my system and were substantial enough to prompt the sale of the Gungnir. Perhaps the differences were evident for you, but those aspects just don't have a bearing in your reckoning? But then again I find it confounding that differences in dynamics, resolution and transients could be so easily dismissed.

 

But in the end it's your gear, your ears, your hard-earned, your preferences and priorities. Kudos for the effort. I hope people take it for what it is - an interesting document of an individual search, and a prompt to audition components for yourself, rather than to take things as gospel.

post #999 of 1331
Quote:
Originally Posted by Chris J View Post
 

 

Fair enough.

I think I would argue that a headphone amplifier is easier to design than a DAC, or easier to make perfect.........or should I say easier to design a good headphone amplifier than a good DAC.

 

Or just follow the spec sheets for the dacs and amps, then spend all the money and brain magic on the power supply.

post #1000 of 1331
Quote:
Originally Posted by Chris J View Post

Fair enough.
I think I would argue that a headphone amplifier is easier to design than a DAC, or easier to make perfect.........or should I say easier to design a good headphone amplifier than a good DAC.

I think designing a good anything requires the extra something. Whether analog or digital. In fact I wonder how many Dac designers have an equal competence on the analog output side....
post #1001 of 1331
Quote:
Originally Posted by olor1n View Post
 

 

Clearly, purrin and co. are shills!

 

Gary, curious what you make of people latching on to your findings and taking them as gospel. I've noted that at various times you've taken care to state disclaimers (your ears, your gear, your preferences/priorities etc) but it's largely been overlooked from my pov.

 

I did give some thought as to how I could reconcile my own experience upgrading dacs (times when I had various components on hand for direct comparison) and purrin's findings with your own polar opposite conclusions. I can understand purrrin's impressions and get a sense of components I've not heard and therefore derive some value from his posts. This is because I can calibrate against his take on the Gungnir, even though we differ on how we rate that component.

 

Your grouping of the Gungnir and M51 as indistinguishable is a mystery to me though. Again, I had both components on hand for weeks and also took great care to volume match. The differences weren't subtle to my ears, on my system and were substantial enough to prompt the sale of the Gungnir. Perhaps the differences were evident for you, but those aspects just don't have a bearing in your reckoning? But then again I find it confounding that differences in dynamics, resolution and transients could be so easily dismissed.

 

But in the end it's your gear, your ears, your hard-earned, your preferences and priorities. Kudos for the effort. I hope people take it for what it is - an interesting document of an individual search, and a prompt to audition components for yourself, rather than to take things as gospel.

my immediate reaction to your experience is that it wasn't a level playing field in your comparison. You had a brand new M51, a lot sexier then the Shiit; twice the price and I would think it would hard to be objective.  I would accept your comparison if you had done a blind test. Just my .02 and how I would probably have reacted.

post #1002 of 1331
Quote:
Originally Posted by mordicai View Post
 

my immediate reaction to your experience is that it wasn't a level playing field in your comparison. You had a brand new M51, a lot sexier then the Shiit; twice the price and I would think it would hard to be objective.  I would accept your comparison if you had done a blind test. Just my .02 and how I would probably have reacted.

 

That's not always true.   I still believe olor1n.

 

There were numerous times that I purchase more expensive, better looking gears such as RCA cables, speaker cables, DACs,   but after listening I found them to be crap, worse than my other cheaper, uglier gears,    and returned them, or never used them again.    And there are also times higher priced stuff worth the price difference.   

 

So  I don't think we are so easily fooled by the look, $, or anticipation.


Edited by yfei - 1/3/14 at 9:37pm
post #1003 of 1331

I recently had the opportunity to demo a bunch of Nordost cables from the Lief series all the way up to the Odin and everything in between. I think one 2.5m Odin power cable runs around $21K USD. (A Lief cable of the same length may start around $500.) And we're not even starting to talk about their speaker cables, interconnects, line harmonizers, and power purifiers. You could easily drop $100K on a cable/power setup for a simple 2.0 system. I in fact heard these cables on a 2.0 setup of Avalon Acoustics Indra speakers ($25K/pair) powered by a McIntosh MA700 integrated ($7K), sourced by a McIntosh MCD550 CD-player ($6500).  When I demoed them, I was not interested in buying anything nor had I purchased anything. But I have ears and I did indeed hear differences as we demoed up the line. The noise floor perceptibly decreased and as a result the vocals, strings, and symbols, etc. all had clearer definition. Now was the difference worth spending $100K on the whole setup? I couldn't dream of it. Heck, I can't afford their Lief line! But to many people out there who share our hobby, spending upwards of $100K on cables is a drop in the bucket. And to them it is indeed worth it. These are also the same people who hire sound engineers to custom build rooms and houses in which this equipment will be used. It's all part the package.

 

What I am saying is that I do not believe that HIFI is about smoke and mirrors and snake-oil salesmen. My Nordost demo is proof of this. But I believe that for 99% of the readers of this forum there does come substantial diminishing returns above say $1000 or $2000 spent on a DAC. Especially if they don't have the appropriate source, cables, and headphones to match the DAC's performance.

 

I have no doubt that Gary is being as honest and objective as his body will allow him to be. I also know his opinions pose a great threat to salesmen and users alike who sell or have already purchased high-end equipment, respectively. At the end of the day, both he and those who doubt his results are both right since sound is so subjective. But it never hurts to remind the readers of this forum that not everyone here is honest and objective. There are forum members here who are paid-to-post and to create mindshare for product A over B. I am in no way insinuating that Gary is working on behalf of Yulong or Metrum, etc. but those who are quick to attack him may have more to lose than you realize.


Edited by schneller - 1/3/14 at 8:39pm
post #1004 of 1331

   I have to say the lack of difference makes me wonder.  Not about Gary's methodology but the DACs themselves.

 

    If you think about the companies who make the DAC chips (I mean Burr Brown, Cirrus Logic, Sabre, etc), they all have some sort of reference design on their web site. You can buy their chip, glue on capacitors and resistors and a power supply and such and have a working box that turns digital into analog based on the reference design. (Anyone who knows more about electronics than I do feel free to correct that assumption if I am wrong, but they have "application notes" that show a presumably working implementation.).

 

    So if you can get a DAC from Schiit for $99, they have to be "cutting some corners" to meet that price point.  Not that makes it bad in any way, but that it means they had to choose the parts carefully to hit that price point.  You would think that as the prices go up, that the reference design starts to get maybe better grade components, more hmmm what if we changed this capacitor to a different value, better noise filtering on the power supply, that sort of thing. Less reference design and more innovation. For example, the NAD 51 uses "Pulse Width Modulation" and Chord uses a FPGA and doesn't even use traditional DAC on a chip.

 

    It's *astounding* that so many DAC can sound the same including the NAD and Chord offerings which you might expect to sound different vs 5 dacs all using slightly different implementations of the sabre chip. (Sorry Gary I know its not literally 5 and you were selecting products on purpose to reduce this situation)

 

    Having said all that, I ordered my DA8 and I am eagerly awaiting its arrival.  You have no idea how restricted the choice of DSD files is on acousticsounds.com if you are in Canada; I need to hunt around other places. And maybe find an old PS3 for "backing up" my SACD discs to something the DA8 can swallow.


Edited by ellemir - 1/3/14 at 10:07pm
post #1005 of 1331
Quote:
Originally Posted by ellemir View Post
 

   I have to say the lack of difference makes me wonder.  Not about Gary's methodology but the DACs themselves.

 

    If you think about the companies who make the DAC chips (I mean Burr Brown, Cirrus Logic, Sabre, etc), they all have some sort of reference design on their web site. You can buy their chip, glue on capacitors and resistors and a power supply and such and have a working box that turns digital into analog based on the reference design. (Anyone who knows more about electronics than I do feel free to correct that assumption if I am wrong, but they have "application notes" that show a presumably working implementation.).

 

    So if you can get a DAC from Schiit for $99, they have to be "cutting some corners" to meet that price point.  Not that makes it bad in any way, but that it means they had to choose the parts carefully to hit that price point.  You would think that as the prices go up, that the reference design starts to get maybe better grade components, more hmmm what if we changed this capacitor to a different value, better noise filtering on the power supply, that sort of thing. Less reference design and more innovation. For example, the NAD 51 uses "Pulse Width Modulation" and Chord uses a FPGA and doesn't even use traditional DAC on a chip.

 

    It's *astounding* that so many DAC can sound the same including the NAD and Chord offerings which you might expect to sound different vs 5 dacs all using slightly different implementations of the sabre chip. (Sorry Gary I know its not literally 5 and you were selecting products on purpose to reduce this situation)

 

    Having said all that, I ordered my DA8 and I am eagerly awaiting its arrival.  You have no idea how restricted the choice of DSD files is on acousticsounds.com if you are in Canada; I need to hunt around other places. And maybe find an old PS3 for "backing up" my SACD discs to something the DA8 can swallow.

to me it shows more how limited our hearing ability actually is. all these dacs cannot possibly output the exact same analog signal (given the same digital signal) as they are implemented differently, yet to the ears (gary's at least), they sound the same. to me this just shows that the difference between competently implemented dacs are beyond what we can hear. the ones that do sound different are either through dsd files (ie not given the same digital signal) or is using some special dsp processing to make it sound different (deviating from what you call "reference design" i suppose)

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Dedicated Source Components
This thread is locked  
Head-Fi.org › Forums › Equipment Forums › Dedicated Source Components › December 2013 Mid-Level DAC Comparison