To Compress or Not?

[tranz]

My position is that where audio quality of digital audio is concerned, money is better spent on a DAC with isolation capabilities rather than tweaking the computer to reduce noise. These tweaks are not free, especially if one places any value on one's time. ...

 

Ah yes, the unicorn of DACs that is impervious to source noise. Let me know when you find one. All the gadgets like USB-SPDIF converters, Regen, etc. are not cheap either. I agree about time being costly when you start down the slippery slope of computer tweaking. But keeping your library in FLAC or uncompressed does not take more time and cannot really be considered computer tweaking.

 

Just my 0.2c. Every setup, ears, tolerances are different of course so it is just another one of these available choices in CA world.

[Paul R]

My position is that where audio quality of digital audio is concerned, money is better spent on a DAC with isolation capabilities rather than tweaking the computer to reduce noise. These tweaks are not free, especially if one places any value on one's time. They will ultimately be less effective than providing isolation. They will have to be redone everytime there is a significant technology change. These optimizations are based on using cheap mass produced components that were designed on the philosophy that "bits is just bits" when the audio application is "bits is not just bits". Thus, the problem, which amounts to "polishing a sow's ear" or "putting lipstick on a pig". Just my opinion. I don't see the point in trial and error optimization of systems where there is no visibility to the underlying implementation.

 

Last time I looked reliable tape drives were extremely costly. The market for corporate storage is not tuned to optimize library sizes in the 5 - 20 TB region.

 

Sez who? VTL's are available for well under $20K to service this segment of the market, and physical tape libraries, including LTO6 with VFS for around $12K. 18-20 Terabytes of SAN storage for under $14K.

 

Sure, those are not exactly home office prices for most people, but they are incredibly cheap for small and mid-sized businesses.

 

For home, there are things coming into place of course.

 

Given that preponderance of streaming services lately, quiet, fast running, power efficient processors with plenty of network connected storage over clean networks is a very big factor in audio sound these days. Especially if you happen to compare $10K of computer/network gear to $10K of DAC. Spend the money on the computer, and buy a less expensive DAC. IMNSHO of course.

 

-Paul

[gmgraves]

There is no credible evidence any of the lossless formats sound different. There is worrisome conjecture about what might be happening, and people who say they can hear a difference (said difference disappears if they don't know what they are listening to). The output you listen to is the same either way. So now all this hysteria about something dead simple and reliable. If you don't even have a system together the difference is at best small. It should be way,way down your list of concerns. Just go ahead and rip to FLAC.

 

 

I certainly do not know that there is any difference. I can, however, understand how a processor in a player would have to work harder to de-compress a FLAC file "on-the-fly" as it were, and that this extra work by the processor could affect the player's power supply by putting an extra load on it, and that this could, conceivably, impact SQ. However, I strongly suspect that whether or not SQ would actually (as opposed to theoretically) be affected would depend upon the design of the individual player, and not be a universal phenomenon. IOW, I would think that if this is really a known issue, any designer could get around it by simply designing the player's power supply to be more robust, and to be able to supply the processor with enough power so that decoding FLAC files while playing simply would not cause an SQ problem. Doesn't that make sense?

[esldude]

I certainly do not know that there is any difference. I can, however, understand how a processor in a player would have to work harder to de-compress a FLAC file "on-the-fly" as it were, and that this extra work by the processor could affect the player's power supply by putting an extra load on it, and that this could, conceivably, impact SQ. However, I strongly suspect that whether or not SQ would actually (as opposed to theoretically) be affected would depend upon the design of the individual player, and not be a universal phenomenon. IOW, I would think that if this is really a known issue, any designer could get around it by simply designing the player's power supply to be more robust, and to be able to supply the processor with enough power so that decoding FLAC files while playing simply would not cause an SQ problem. Doesn't that make sense?

 

It does and it does not. Nice hey!

 

At most, at MOST, it is like AC on a high performance car, something you have experience with. Okay, maybe one can tell when the AC is on or not. With it on, you have a touch less oomph, a touch less balls, a touch less power. In a very high performance car, the difference is negligible vs lower performance vehicles. Perhaps perceptible, but not huge.

 

Yet even plebian computing machines find doing audio processing less strain than AC on a car. So will doing FLAC at 2 percent more CPU really be something you can notice? Maybe, but how big will be the difference? Not much. And in a modern car, like the Corvette I own, rolling down the windows and turning off AC will cause more issues than leaving AC on and windows up.

 

Here in the SE USA, with high temps and very high humidity, the AC on and losing 2% absolute performance is an absolute bargain. BARGAIN I tell you. Especially over long journeys. 300 miles at 150+ I will take the AC and loss in performance gladly everytime.

[Paul R]

I certainly do not know that there is any difference. I can, however, understand how a processor in a player would have to work harder to de-compress a FLAC file "on-the-fly" as it were, and that this extra work by the processor could affect the player's power supply by putting an extra load on it, and that this could, conceivably, impact SQ. However, I strongly suspect that whether or not SQ would actually (as opposed to theoretically) be affected would depend upon the design of the individual player, and not be a universal phenomenon. IOW, I would think that if this is really a known issue, any designer could get around it by simply designing the player's power supply to be more robust, and to be able to supply the processor with enough power so that decoding FLAC files while playing simply would not cause an SQ problem. Doesn't that make sense?

 

 

I do know there is a processing difference George. And some folks are going to be sensitive to that difference in one way or another. Just as you belive that the only music worth lustening to is "live" - a point that is absolutely true for you, and to which you are more sensitive than other people. Because I for instance, disagree with you, does not make it less true for you.

 

Some folks here can reliably discriminate between AIFF and WAV files, a jaw dropping fact if there ever was one. A fact none the less, and trying to intimate it is anything less than a fact is just pure cussedness.

 

Whether a power supply redesign would eliminate the differences for those folks or not is a good question; only problem is in trying to falsify that idea. That gets expensive - rapidly - and really does not add to anyone's enjoyment. It is like grammar school challenges.

[goldsdad]

Some folks here can reliably discriminate between AIFF and WAV files.

 

How was the reliability determined?

[gmgraves]

It does and it does not. Nice hey!

 

At most, at MOST, it is like AC on a high performance car, something you have experience with. Okay, maybe one can tell when the AC is on or not. With it on, you have a touch less oomph, a touch less balls, a touch less power. In a very high performance car, the difference is negligible vs lower performance vehicles. Perhaps perceptible, but not huge.

 

Yet even plebian computing machines find doing audio processing less strain than AC on a car. So will doing FLAC at 2 percent more CPU really be something you can notice? Maybe, but how big will be the difference? Not much. And in a modern car, like the Corvette I own, rolling down the windows and turning off AC will cause more issues than leaving AC on and windows up.

 

Here in the SE USA, with high temps and very high humidity, the AC on and losing 2% absolute performance is an absolute bargain. BARGAIN I tell you. Especially over long journeys. 300 miles at 150+ I will take the AC and loss in performance gladly everytime.

 

 

I understand the concept, but I'm not sure that the analogy is an apt one. You seem to be saying that decoding FLAC (or any other compressed file) on-the-fly is analogous to a loss of power to an air conditioning compressor in a car. I doubt if that's the correct model. I believe that the processor doing the decompression would probably introduce noise onto the power supply, rather than cause a loss of voltage or current. This PS noise is what might manifest itself as a loss of SQ. But again, I don't know that any of this actually happening in the real world, I'm just saying it's possible that real-time decompressing could be compromising SQ in some manner.

[gmgraves]

I do know there is a processing difference George. And some folks are going to be sensitive to that difference in one way or another. Just as you belive that the only music worth lustening to is "live" - a point that is absolutely true for you, and to which you are more sensitive than other people. Because I for instance, disagree with you, does not make it less true for you.

 

OK.

 

Some folks here can reliably discriminate between AIFF and WAV files, a jaw dropping fact if there ever was one. A fact none the less, and trying to intimate it is anything less than a fact is just pure cussedness.

 

Has anyone tested people on this? If so, how was the test conducted and were the results statistically valid? People claim that they can hear lots of things that just don't seem reasonable. I used to know a guy who was a very well known audio critic (alas, he's no longer with us). He swore that digital sound sounded so bad to him, that he could tell the difference between an analog audio signal and a digital one every time. Some of us decided to test him on this. We took one of my 1/2 track, 15 ips Dolby A master tapes of the San Jose Symphony (CA) and digitized it to 16-bit, 44.1 KHz Beta tape using a Apogee modified Sony F1 digital encoder. Our "pigeon" was ensconced in a separate listening room from the equipment while we switched between the analog tape recorder and the digital tape. Our "golden-eared" digital expert got it right slightly less than 50% of the time - IOW, blind chance. And this was when digital was, if not actually in its infancy, then certainly it was, at best, just a toddler technology. People who make extraordinary claims must be held to extraordinary tests. Much of the time, their claims are shown to be mostly their overactive imaginations.

 

Whether a power supply redesign would eliminate the differences for those folks or not is a good question; only problem is in trying to falsify that idea. That gets expensive - rapidly - and really does not add to anyone's enjoyment. It is like grammar school challenges.

 

Well, I think that before anyone goes to any great engineering lengths to solve such a problem, we ought to distinguish whether or not that problem really exists by doing some actual research rather than listening to people's possible flights of paranoid imagination. Not saying that decoding FLAC files on-the-fly doesn't result in a loss of SQ, you understand; perhaps it does. But before I buy into it, I'd like to see a little hard data.

[Paul R]

OK.

 

 

 

Has anyone tested people on this? If so, how was the test conducted and were the results statistically valid? People claim that they can hear lots of things that just don't seem reasonable. I used to know a guy who was a very well known audio critic (alas, he's no longer with us). He swore that digital sound sounded so bad to him, that he could tell the difference between an analog audio signal and a digital one every time. Some of us decided to test him on this. We took one of my 1/2 track, 15 ips Dolby A master tapes of the San Jose Symphony (CA) and digitized it to 16-bit, 44.1 KHz Beta tape using a Apogee modified Sony F1 digital encoder. Our "pigeon" was ensconced in a separate listening room from the equipment while we switched between the analog tape recorder and the digital tape. Our "golden-eared" digital expert got it right slightly less than 50% of the time - IOW, blind chance. And this was when digital was, if not actually in its infancy, then certainly it was, at best, just a toddler technology. People who make extraordinary claims must be held to extraordinary tests. Much of the time, their claims are shown to be mostly their overactive imaginations.

 

 

Why assume that? To be honest, I do not hear a difference, on my equipment, with my tired old ears. But people whose opinions I respect do, and have reliably done so in ways that I have no reason to question.

 

Not saying you do this at all, but some folks will take it as a mission from on high to disprove anything they don't believe is possible. That is really just cussedness - and the people they are opposing are quite capable of cussin' back. Quite inventively in some cases. And yes, the same applies to the other side too - something appears to work for them so it must be true for everyone, in every case. That is just as silly.

 

Well, I think that before anyone goes to any great engineering lengths to solve such a problem, we ought to distinguish whether or not that problem really exists by doing some actual research rather than listening to people's possible flights of paranoid imagination. Not saying that decoding FLAC files on-the-fly doesn't result in a loss of SQ, you understand; perhaps it does. But before I buy into it, I'd like to see a little hard data.

 

I do not blame you. On the other hand, if you try something - say a different power cord, or comparing an AIFF and WAV file - and you *do* hear a difference, what on earth would compel you to spend the time and effort to prove it to a hostile audience? Would you not rather be listening to some music?

 

I know I once offered to put up a significant amount of money to buy and ship a test set of gear around to CA members, in an attempt to maybe gather some significant data and put some of these theories to the test. I ran smack into a hostile little feller whose only goal was to make fun of people who didn't agree with his "super scientific knowledge." To put it mildly, that feller totally annoyed me. Unlike Bill and others who are skeptical to the extreme, but are at least willing to give things a try.

 

I would love some hard data on this stuff, but you know what? I do not think we are *ever* going to get that data. Too many people with too much emotional baggage on all sides of the argument to ever look at facts. On both sides of the argument. Or rather I should say, on *ALL* sides of that argument, since there are almost as many sides as there are people with opinions about it... (*sigh*)

 

-Paul

[goldsdad]

But people whose opinions I respect do, and have reliably done so in ways that I have no reason to question.

 

What are these ways, and how is the reliability ascertained?

[gmgraves]

Why assume that? To be honest, I do not hear a difference, on my equipment, with my tired old ears. But people whose opinions I respect do, and have reliably done so in ways that I have no reason to question.

 

To be honest with you, Paul, you question things because that's how we learn. Human history is strewn with ideas that people have blindly followed because someone they trust, or someone in authority has told them to follow. Often, too often, these blind alleys have resulted in war and pestilence and death. Not to say that this subject is in any danger of leading there, but the premise is real. It is our legacy and our scientific duty to investigate phenomenon that seem counter to accepted wisdom. At least, that's my opinion.

 

Not saying you do this at all, but some folks will take it as a mission from on high to disprove anything they don't believe is possible. That is really just cussedness - and the people they are opposing are quite capable of cussin' back. Quite inventively in some cases. And yes, the same applies to the other side too - something appears to work for them so it must be true for everyone, in every case. That is just as silly.

 

Paul, that's the scientific method. Someone puts forth a hypothesis about some observed phenomenon, and others try to shoot it down. If they can't shoot it down, then it gets accepted as a working hypothesis, and science tries to find the physics behind the phenomenon. Just about everything we know was learned, more or less, in that manner.

 

 

I do not blame you. On the other hand, if you try something - say a different power cord, or comparing an AIFF and WAV file - and you *do* hear a difference, what on earth would compel you to spend the time and effort to prove it to a hostile audience? Would you not rather be listening to some music?

 

Well, yes and no. Sure I'd rather listen to music, but something about my character compels me to know why. I suspect that I'm not alone in my insatiable curiosity about such things, but OTOH, there are obviously people who are satisfied to just accept the results and not worry about it.

 

I know I once offered to put up a significant amount of money to buy and ship a test set of gear around to CA members, in an attempt to maybe gather some significant data and put some of these theories to the test. I ran smack into a hostile little feller whose only goal was to make fun of people who didn't agree with his "super scientific knowledge." To put it mildly, that feller totally annoyed me. Unlike Bill and others who are skeptical to the extreme, but are at least willing to give things a try.

 

As humans, my feeling is, that we owe it to our collective knowledge base to always be willing to give things a try. But it must be done in a systematic and logical manner so that we don't mistake false findings, human bias, and inconclusive results for real data.

 

I would love some hard data on this stuff, but you know what? I do not think we are *ever* going to get that data. Too many people with too much emotional baggage on all sides of the argument to ever look at facts. On both sides of the argument. Or rather I should say, on *ALL* sides of that argument, since there are almost as many sides as there are people with opinions about it... (*sigh*)

 

-Paul

 

I'm afraid that you are right. The problem is that the type of people who should be researching these things; that is to say, people with no axes to grind, aren't in the least interested in pursuing these subjects. Audiophiles are the wrong people to do this research because they are like Creation Scientists who approach their research already knowing the desired result. To them, the universe is 6000 years old, and everything they investigate starts with the premise that their results must fit in the framework of what, to them, is an incontrovertible fact. Whether they are right or wrong about the age of the universe is something I'm not going comment on, but either way, this is not the right approach to science. Real science follows the evidence and goes where the evidence leads. It does not lead the evidence. So whether you believe in these audio differences or not, you are likely going to approach the subject already convinced of one outcome or the other. Nothing gets solved in that case.

[Paul R]

To be honest with you, Paul, you question things because that's how we learn. Human history is strewn with ideas that people have blindly followed because someone they trust, or someone in authority has told them to follow. Often, too often, these blind alleys have resulted in war and pestilence and death. Not to say that this subject is in any danger of leading there, but the premise is real. It is our legacy and our scientific duty to investigate phenomenon that seem counter to accepted wisdom. At least, that's my opinion.

 

And it is an opinion I generally share - but like anything, questioning everything can and sometimes is used as a method to bully people around. That's especially true online, where they do not risk getting punched soundly in their snouts for such boorish behavior. It may be that a few people view online forums as "safe" places to express a little of their inner psychopath. I honestly don't know...

 

 

Paul, that's the scientific method. Someone puts forth a hypothesis about some observed phenomenon, and others try to shoot it down. If they can't shoot it down, then it gets accepted as a working hypothesis, and science tries to find the physics behind the phenomenon. Just about everything we know was learned, more or less, in that manner. [/Quote]

 

Well, *just* asking questions is not really an expression of the scientific method, though I can see where some folks believe it is. That's just the first part, all the*hard* work comes later. Defining a testable hypothesis, designing and carrying out the tests and observations, evaluating them and publishing. A lot of folks here have tried to short cut that process, by claiming they understand and are "scientific." They might as well say they are Catholic, or Baptist, or Zoroastrian, whatever. Anything but truly scientific.

 

Well, yes and no. Sure I'd rather listen to music, but something about my character compels me to know why. I suspect that I'm not alone in my insatiable curiosity about such things, but OTOH, there are obviously people who are satisfied to just accept the results and not worry about it.

[/QUote]

 

Well of course, but it isn't so much just "accepting" as "not being willing to pay the price to find out." In the example I gave earlier, super scientific guy was not interested in the research part, he merely wanted to design some activity that would "prove" his point. No real research, just ego there. One thing *I* have discovered, is that if you are really researching something, you must - absolutely must - subsume your ego.

 

 

As humans, my feeling is, that we owe it to our collective knowledge base to always be willing to give things a try. But it must be done in a systematic and logical manner so that we don't mistake false findings, human bias, and inconclusive results for real data.

 

 

 

I'm afraid that you are right. The problem is that the type of people who should be researching these things; that is to say, people with no axes to grind, aren't in the least interested in pursuing these subjects. Audiophiles are the wrong people to do this research because they are like Creation Scientists who approach their research already knowing the desired result. To them, the universe is 6000 years old, and everything they investigate starts with the premise that their results must fit in the framework of what, to them, is an incontrovertible fact. Whether they are right or wrong about the age of the universe is something I'm not going comment on, but either way, this is not the right approach to science. Real science follows the evidence and goes where the evidence leads. It does not lead the evidence. So whether you believe in these audio differences or not, you are likely going to approach the subject already convinced of one outcome or the other. Nothing gets solved in that case.[/Quote]

 

Pretty much, so it is an "enjoy that music and good fellowship" kind of thing I think. Maybe sneak in a bit of research on the side...

 

-Paul

[esldude]

 

Well, I think that before anyone goes to any great engineering lengths to solve such a problem, we ought to distinguish whether or not that problem really exists by doing some actual research rather than listening to people's possible flights of paranoid imagination. Not saying that decoding FLAC files on-the-fly doesn't result in a loss of SQ, you understand; perhaps it does. But before I buy into it, I'd like to see a little hard data.

 

I think in a similar manner on this. But there is lots of money to be made just fixing something perceived as a problem whether it is a problem or not.

[esldude]

I understand the concept, but I'm not sure that the analogy is an apt one. You seem to be saying that decoding FLAC (or any other compressed file) on-the-fly is analogous to a loss of power to an air conditioning compressor in a car. I doubt if that's the correct model. I believe that the processor doing the decompression would probably introduce noise onto the power supply, rather than cause a loss of voltage or current. This PS noise is what might manifest itself as a loss of SQ. But again, I don't know that any of this actually happening in the real world, I'm just saying it's possible that real-time decompressing could be compromising SQ in some manner.

 

You are correct of course. Or maybe not. I don't really know what people think this extra processing is doing. It doesn't rob power hurting the conversion. It might cause a slightly different noise profile, but is it one large enough to really matter? This is one of those cases where we don't even bother to see if the problem is real, don't even define what the problem is in concept or even propose a model of what is happening to result in the problem. The noise profile emanating from two very similar collections of PC hardware is so different one would have difficulty teasing out what 2% doing the decompression would be.

[goldsdad]

What are these ways, and how is the reliability ascertained?

 

As expected, you cannot substantiate your claim, Paul. The truth, as far as I'm aware, is that nobody has demonstrated an ability to reliably distinguish AIFF from WAVE without knowing which format they're listening to.

[Paul R]

You are correct of course. Or maybe not. I don't really know what people think this extra processing is doing. It doesn't rob power hurting the conversion. It might cause a slightly different noise profile, but is it one large enough to really matter? This is one of those cases where we don't even bother to see if the problem is real, don't even define what the problem is in concept or even propose a model of what is happening to result in the problem. The noise profile emanating from two very similar collections of PC hardware is so different one would have difficulty teasing out what 2% doing the decompression would be.

 

You test at the extremes of course. PC people rarely think that loading a machine down is a good thing, but it is only when you have proccessors near 100% and memory pressure near stavation, or in the case of modern systems even oversubscribed, that the effects of stuff like this are exaggerated enough to be easily observable.

 

As to what effect you would be looking for - noise and timing at the DAC of course. This is why something like the UA USB Regenerator can, at least potentionally, have a beneficial effect. Better isolation = better sound is the hypothesis to test.

 

And yep- people have done that. With results that are respectable. See John Swenson.

[Tony Lauck]

As expected, you cannot substantiate your claim, Paul. The truth, as far as I'm aware, is that nobody has demonstrated an ability to reliably distinguish AIFF from WAVE without knowing which format they're listening to.

 

I find no need for Paul to substantiate his claim, since I've heard similar differences, myself.

 

If you would like a demonstration I suppose someone could write some buggy player code that produces gross sonic differences when playing different formats. Such a program could be further modified to create subtle differences as well... (My point here is that if differences exist they reflect defects in the playback chain, not the file formats.)

[goldsdad]

I find no need for Paul to substantiate his claim, since I've heard similar differences, myself.

 

If you would like a demonstration I suppose someone could write some buggy player code that produces gross sonic differences when playing different formats. Such a program could be further modified to create subtle differences as well... (My point here is that if differences exist they reflect defects in the playback chain, not the file formats.)

 

Buggy player code? OK, I concede that a sufficiently broken system, or, as you suggest, a deliberately defective and corrupting system, may produce differences that are audible to almost anyone.

 

Why not take it to the extreme: a system that plays one format and simply doesn't play another. I guess that certainly would prove Paul's claim correct.

[gmgraves]

I think in a similar manner on this. But there is lots of money to be made just fixing something perceived as a problem whether it is a problem or not.

 

 

Yes. It's what I call marketing-driven pseudo-research. Make up a problem, pretend to research a solution and then market that solution. Laundry detergent manufacturers have been doing that with New, and Improved TIDE since the 1940's. Unfortunately, high-end Audio companies do it too!

[Paul R]

I find no need for Paul to substantiate his claim, since I've heard similar differences, myself.

 

If you would like a demonstration I suppose someone could write some buggy player code that produces gross sonic differences when playing different formats. Such a program could be further modified to create subtle differences as well... (My point here is that if differences exist they reflect defects in the playback chain, not the file formats.)

 

Ah- I didn't realize Owen was still posting his usual abuse around the system. I don't reply to him or even read his messages, bless that "ignore list" feature...

 

In any case, he is smart and when not being viciously abusive, does have some good thinking worth listening to. Not, I fear, good enough for me to listen to his rants, but upon occasion. He is still pissed from when I refused to "name names" a few years ago. (shrug)

[esldude]

You test at the extremes of course. PC people rarely think that loading a machine down is a good thing, but it is only when you have proccessors near 100% and memory pressure near stavation, or in the case of modern systems even oversubscribed, that the effects of stuff like this are exaggerated enough to be easily observable.

 

As to what effect you would be looking for - noise and timing at the DAC of course. This is why something like the UA USB Regenerator can, at least potentionally, have a beneficial effect. Better isolation = better sound is the hypothesis to test.

 

And yep- people have done that. With results that are respectable. See John Swenson.

 

When I asked John Swenson if he had measured differences in the analog output at the DAC (which is the signal we are listening to) he said such a task was complex and he did not have measurable results there. So nope, so far, I haven't seen that. Miska I believe posted some results where the ultrasonic noise was different. But we don't hear that. Perhaps in some systems it has other effects, but again, in our hearing range those effects appear hard to come by.

[esldude]

I find no need for Paul to substantiate his claim, since I've heard similar differences, myself.

 

If you would like a demonstration I suppose someone could write some buggy player code that produces gross sonic differences when playing different formats. Such a program could be further modified to create subtle differences as well... (My point here is that if differences exist they reflect defects in the playback chain, not the file formats.)

 

I think I would prefer recording the two formats with DSD256, and then transferring the two files to vinyl. Then see if you could hear the difference. Nothing like a direct test of the phenomena under discussion with SOTA formats. (Mutters under breath, and shakes head)

[audiventory]

I think I would prefer recording the two formats with DSD256, and then transferring the two files to vinyl. Then see if you could hear the difference. Nothing like a direct test of the phenomena under discussion with SOTA formats. (Mutters under breath, and shakes head)

 

Goog idea: move all to single media for testing.

 

It inspired me to thought about other way: capture flac player's output from virtual audio device (PCM format) to wav.

And compare binary content in hex-viewer. Most sophisticated here is synchronizing.

 

First need capture played back original wav.

 

Virtual device must not do any DSP. It possibly check via playback original wav and compare captured content with original wav. Via binary content comparison.

[Paul R]

Goog idea: move all to single media for testing.

 

It inspired me to thought about other way: capture flac player's output from virtual audio device (PCM format) to wav.

And compare binary content in hex-viewer. Most sophisticated here is synchronizing.

 

First need capture played back original wav.

 

Virtual device must not do any DSP. It possibly check via playback original wav and compare captured content with original wav. Via binary content comparison.

 

 

Dennis was being sarcastic there, as he has stated many times before that although he can hear differences in equipment based upon cables, formats, or other "audiophile" beliefs, he firmly believes all the differences he hears are imaginary and there is no benefit anyone anywhere can realize with hi-res formats.

 

There is a lot of real research to do on the subject, but there is not a lot of time or money being invested into it. Dennis often has a lot of good things to say, but he has sensitivities in this subject area that often result in rather sarcastic postings like that one.

 

-Paul

[Shadorne]

Ah yes, the unicorn of DACs that is impervious to source noise. Let me know when you find one. All the gadgets like USB-SPDIF converters, Regen, etc. are not cheap either. I agree about time being costly when you start down the slippery slope of computer tweaking. But keeping your library in FLAC or uncompressed does not take more time and cannot really be considered computer tweaking.

 

Just my 0.2c. Every setup, ears, tolerances are different of course so it is just another one of these available choices in CA world.

 

There are Unicorns around since 2002. They are no longer rare. Benchmark DAC1 started the mainstream movement to asynchronous DACs and we haven't look back since then.

 

The problem in the past was that the clock signal came from the master and couple that with the Inherent jitter issues with spdif, toslink and USB and HDMI and you have universal problems.

 

You need to read up on asynchronous designs - you can couple them with pretty much everything and you can ignore the jitter from the source....