[jcx]

 It was also a mistake to allow myself to be engaged by an ad hominem critical post by a forum member

I'm sorry that you still see it that way - how many times do I have to say I discuss Ideas - criticizing someone's statements, reasoning about a subject is not an attack on the person
 
 
I too have "senior and graduate level" DSP course books - needed to learn polyphase time interpolating filter design in multiplexed sampled data systems requiring time aligned data output
 
and I don't imagine myself an expert by those standards - many, probably most of the tens of thousands of EEs that have taken the actual courses, or done serious self study are ahead of me
 
 
not knowing what you are doing in DSP specifically I haven't said there are no possible improvements
 
 
I think I can still reasonably be skeptical of how important these unnamed improvements are for listening to digital audio music releases given so many other limitations in time, phase errors of transducers, inherent and added noise
 
 
I am not at all a "meter reader" reductionist - Psychoacoustics is broad, complex and incomplete
 
 
if you want a even more extreme example try inter aural time delay - clicks presented to opposite ears can be discerned to arrive at different times as small as 10, possibly 5 us, maybe 2 us? - the textbook number may still be 20 us - which is still intriguing
 
and I do criticize many surrounding claims, Signal Theory fails of even Kunchur - the differences can be shown with 16/44 digital audio
 
what I really try do debunk is the implication that 50, 100, even 200 kHz hearing is required to explain it - and after its all wrapped up Kunchur agrees, points to correlation - not 100 kHz hearing
 

 
 
for my part the only "fight " was to get considered seriously, get you to pay attention to what was said - not being stereotyped, my argument ignored - get past your nonresponsive and non technical criticisms

 

[JWolf]

   
shows you are using nasty rhetoric there - putting “fighting words” in my mouth that I never said, employing belittling tone, insulting “humor”
 
when you appear to not even understand the point - which was that the described signals can be audibly discriminated when band limited to below20 kHz - sorry but that anecdote don't hunt
 
makes the climb down a bit awkward – doesn't it?

 
Why is it you are splitting your lines and making them double spaced? I would find it much easier to read if you didn't split your line and double space. Thanks.

 

[Joe Bloggs]

It was also a mistake to allow myself to be engaged by an ad hominem critical post by a forum member. My worst mistake was to descend to the phallus-waving techs-narcissism established in the original post that publicly called me out. That does not serve any of the readers well. For the above two points, I apologize to all members of this forum, including the original poster. It is my experience that if one puts out slop, one gets it back. For that reason, I will not be further engaged on this matter.

While I have no idea what the original post was or whether you do indeed owe him an apology, calling the original-poster a "d**k-waving tech narcissist" and then apologizing to him for "descending to his level" doesn't really look like an apology to me.

 

[coli]

So... what's the conclusion? Someone write a tl;dr please?

 

[KeithEmo]

The problem with audibility tests is that they tend to be limited... and many of the ones that I see referenced over and over are VERY limited (this often happens with other types of tests as well). For example, the Boston Audio Society test where they "introducing a Red Book CD record-playback loop into the signal chain" to test the audibility of the differences between high-res and "ordinary" digital audio files..... 
 
The basic "logic" of that test is that:
 
1) We all know that there are measurable differences between high-res and ordinary-res files (wider bandwidth; better S/N ratio).
2) If we pass our high-res signals through a "CD quality loop" we will eliminate those differences.
3) If we hear no difference when we do so, then we can reasonably conclude that the differences were inaudible (because we didn't hear a difference when we eliminated them). 
 
Now, here are the "gaps" in their methodology:
 
1) It was never confirmed that the differences POSSIBLE between high-res files and regular ones were present in their samples. I don't recall seeing any test results showing that any of the sample content they chose actually included any signal content in the range of frequencies which high-res files can reproduce and ordinary files cannot.
 
2) It was never confirmed that, if any of that "extra information" was there in their test samples, that their equipment could reproduce it. If we're trying to test whether anyone can hear the "ultrasonic content" present in a high resolution disc, we need to confirm both that it is present in our test samples, AND that our test equipment (including the speakers) can reproduce it. (Unless we confirm all those unknowns, all we really know is that the high-res discs MIGHT contain content that would be eliminated by passing it through the standard-res loop).
 
In other words, what they did was logically equivalent to an audiologist handing you a headset, setting a dial on his equipment to 18 kHz, and declaring that "18 kHz was inaudible to you" when you didn't report hearing anything - without confirming whether his equipment was actually putting out a signal at 18 kHz, or whether the headphones he used could reproduce it. (In logical terms, if you heard the tone, that would confirm all three conditions; but your not hearing it only confirms that EITHER you can't hear it, or the headphones can't produce it, or the output simply isn't working.) In their test, their null result proved that EITHER there is nothing audible in a high-res signal that can't be reproduced perfectly by a standard-res signal, OR that there was nothing present in the particular sample content they chose which couldn't be reproduced perfectly by a standard-res signal, OR that, if there was something present in the particular sample content they chose which couldn't be reproduced perfectly by a standard-res signal, it couldn't be reproduced by their test setup anyway (perhaps because of some limitation of their disc player or speakers).
 
To put it simply, in order to validate that their test procedure would actually test what it was intended to they should have:
 
1) Tested their sample content to ensure that there was in fact some measurable difference between the high-res and the standard-res version.
2) Confirmed that their electronic reproduction chain could accurately reproduce both signals and the differences between them.
3) Used a measurement microphone to confirm that their speakers could accurately reproduce both signals, and the difference between them, IN THE AIR.
 
At this point, they could state with validity that they were actually delivering the correct test signals, and so could expect accurate results.
 
And, unfortunately, any "audibility testing" conducted by an individual is even more limited. You or I can only confirm whether a certain difference is audible with the content we currently have, and using the equipment we currently own (or have access to). The problem is that, by confirming that we hear no difference with our current music and equipment, we can't infer that there won't be some audible difference with the content and equipment we might own tomorrow or next year...  and, as I've noted before, we'd hate to spend money to put together a great collection of content, only to find out later that it has some audible shortcoming that's clearly audible with the new speakers we buy next year, or with some new recordings we may buy later. Therefore, the quest for "something that is audibly as good as possible" rather than simply "something that has no flaws that are audible at the moment" isn't really unreasonable.
 
Incidentally, I agree with you 100% - that the biggest problem today is simply poor quality mastering (or even music that is deliberately mastered with minimal dynamic range and maximal loudness to appeal to a certain audience). But that is, as you say, a different battle.....  (However, I do think that, overall, it's "a good thing" that it is at least becoming obvious that customers do in fact care about audio quality... )
 
 
Quote:

  for the interest of how things work, I always enjoy reading about jitter, digital and analog filters, LSB, oversampling, nyquist, and music's actual dynamic range. same for the ever growing information we have on psycho acoustic. but I fail to see why they should lead an argument about audibility? shouldn't audibility tests be the valid answer to audibility concerns?
 
and shouldn't we do our own audibility tests to come up with our own conclusions and choices? as far as I'm concerned, from time to time I notice a difference in the highres file, I convert it to 16/44 and abx it with the cd rip I have(if the differences aren't very obvious. with stuff like "random access memory", ABX really wasn't needed^_^). and sure enough the differences are still very much on the downsampled version. demonstrating that I'm listening to 2 different masters and not to highres vs CD. to me that's important so that we avoid fighting the wrong war.
based on my tests, clearly my war is on mastering job, and never once was about highres vs cd where trying to ABX the highres file to a downsampled then upsampled file, resulted in guessing statistics for as long as I remember doing it. I am yet to find one album that is the same master but sounds different when played in highres. I've looked long and hard, and listened on a variety of gears throughout the years. nothing, nada, rien.
so that gets me wondering how many of the people talking about how highres sounds better, are really talking about highres at all?
 
 
now a guy with something like a pono will hear differences with all downsampled tracks because of how the pono rolls off the trebles when playing 16/44. so while the gear is clearly to blame, it could make sens(somehow) for the person using a pono to listen to highres. or at least for him to upsample his library as a way to avoid the stupidly strong roll off @16/44 if he doesn't want it.
 
and with different situations we get different reasons to go for, or avoid highres. but actual audibility should be our main audio reason. all other reasons like, fashion, being reassured, misunderstanding the actual limits of 16/44, or just wishing to have the best, those are ok reasons too at an individual level. but they have no relation to what we actually can or cannot hear.

 

[RRod]

And, unfortunately, any "audibility testing" conducted by an individual is even more limited. You or I can only confirm whether a certain difference is audible with the content we currently have, and using the equipment we currently own (or have access to). The problem is that, by confirming that we hear no difference with our current music and equipment, we can't infer that there won't be some audible difference with the content and equipment we might own tomorrow or next year...  and, as I've noted before, we'd hate to spend money to put together a great collection of content, only to find out later that it has some audible shortcoming that's clearly audible with the new speakers we buy next year, or with some new recordings we may buy later. Therefore, the quest for "something that is audibly as good as possible" rather than simply "something that has no flaws that are audible at the moment" isn't really unreasonable.

 
What year is this going to happen then, Keith? SACD came out in 1999, yet for some reason we aren't yet in a world where equipment is expected to be speced out beyond 20kHz or with 144dB of SNR; why might that be? And I agree with you that various published tests leave something to be desired, but I also think that you should probably also then go around the rest of the forum and ask all these people claiming "huge" differences with hi-res just how well their equipment is reproducing hi-res content. Personally, my E-MU delivers pretty flat performance up to 48kHz, and Sennheiser would have me believe my HD800s are only -3dB at 44.1kHz (which I can't really test myself, unfortunately). Intermodulation tests at 96ksps stay perfectly quiet even at high volumes (not so for 192ksps), so for up to that sample rate I'm at least somewhat sure my ABX results for hi-res testing aren't totally fallacious. I mean, what equipment would you have someone use that you would consider kosher?

 

[Joe Bloggs]

1) It was never confirmed that the differences POSSIBLE between high-res files and regular ones were present in their samples. I don't recall seeing any test results showing that any of the sample content they chose actually included any signal content in the range of frequencies which high-res files can reproduce and ordinary files cannot.

2) It was never confirmed that, if any of that "extra information" was there in their test samples, that their equipment could reproduce it. If we're trying to test whether anyone can hear the "ultrasonic content" present in a high resolution disc, we need to confirm both that it is present in our test samples, AND that our test equipment (including the speakers) can reproduce it. (Unless we confirm all those unknowns, all we really know is that the high-res discs MIGHT contain content that would be eliminated by passing it through the standard-res loop).

We're getting somewhere here though. All those arguments made that hi-res e.g. fills in the gaps in those standard-res "stairsteps" even within 20kHz, arguments about "timing accuracy" of signals within 20kHz, 19kHz signals "fading in and out", etc. etc. At least those can be put behind if we agree that you need reproduced >22kHz signals for the redbook standard to be tested against, no?

 

[KeithEmo]

In order to do a valid test, you need to be able to "control for all the variables" - and, until you do so, then your test is NOT 100% conclusive. I agree with you that a lot of the things people think they hear are probably just imagination, or placebo effect, and also that a lot of other things are so trivial or occur so rarely that they probably don't matter, but, as someone with a science background, I am forced to differentiate between scientific facts and generalizations that aren't based solidly on facts. 
 
As for equipment, I do know that there are test microphones claimed to have response well past 50 kHz, so it should be possible to find an SACD, or some other form of high-res file, that has content which includes harmonics out to 30 kHz or 40 kHz, play it through an amplifier, and see if we can find a pair of headphones or set of speakers which our microphone shows are actually reproducing those harmonics. If we do find them, then we'll have a valid test setup; and, if not, then we'll be forced to conclude that we are simply unable, for technical reasons, to perform a conclusive test. (It happens all the time in "real scientific circles" - where someone has an interesting theory, but the limitations in current technology prevent them from being able to test it. In the case of audio, we might even conclude that, since we couldn't find any content that contained significant information at the frequencies in question, it really doesn't matter... however, even then, we couldn't conclude that, IF such content existed, the difference wouldn't be audible.)
 
In other words, we need to be very specific in the way we phrase our questions.
 
Here are several different questions. Think very carefully about the answer to each - based on the results of all the tests you've read about,:
 
1) "Is it possible, under any conditions, using any recording equipment, and any content, for a human being to hear a difference between a high-res recording and a standard-res one?"
 
2) "Has anyone proven conclusively, using current sources and current equipment, that audible differences between high-res and standard-res files exist?"
 
3) "Has anyone proven conclusively, with every possible combination of current source material and current equipment, that audible differences between high-res and standard-res files do NOT exist?"
 
4) "Has anyone proven conclusively that there is no difference between high-res and standard-res files that will be audible to any human being, with any possible combination of content and equipment that currently exists or might ever be developed?"
 
I would summarize all of the information that I've read about so far as follows:
 
Technically, high-res audio files have wider frequency response and a better S/N ratio than standard-resolution files. However, several tests of reasonable but limited scope have been conducted, and none of them has produced evidence to suggest that these differences are audible to a significant number of test subjects, under a variety of typical listening conditions. Another data point is provided by a large collection of anecdotal evidence from people who claim to hear a significant difference. However, at least so far, whenever these claims have been tested using scientific test methods, they have not been shown to have merit, so it seems reasonable to conclude that some or all of them are simply the result of expectation bias or the placebo effect.
 
(To put it bluntly, many people out there believe quite strongly in things that, by available scientific information, seem less likely. And, every now and then, something widely believed by science to be true turns out NOT to be true. I remember when we were all taught that "matter was made up of protons, neutrons, and electrons - which were the smallest indivisible units of matter"; as it turns out, this is a useful model, but we now know that it isn't in fact true. And the only "evidence" we have so far that "high-res files aren't audibly superior" is simply a lack of evidence that they are... which is sufficient justification for an assumption, but not a statement of fact.)
 
To answer your final question directly, I would use a wider variety of equipment, in the hope of providing a good cross sample. Specifically, I would have added a variety of high-end DACs and at least a few pairs of electrostatic headphones to the test list - because I personally find them to do a good job of making tiny differences audible in general. I would also invite people, including both end users and equipment vendors, to submit equipment for consideration. And, if a difference was in fact audible with any of them, then I would be forced to conclude that "an audible difference exists"... but, if no difference showed up in the tests, I would be forced to state the list of equipment with which no difference was detected (and I would refrain from making generalizations that extended to equipment I hadn't tested). I would leave it to the person reading the test to draw their own conclusions as to whether the failure to demonstrate a difference with xxx pieces of equipment - as listed - was sufficient for them to decide "it probably didn't matter" or not.  
 
(The fact is that, in all of the tests I've read about, a truly tiny percentage of all the audio equipment currently in existence was tested, which sure makes it a stretch to extend those conclusions to all of the equipment out there.)
 
Quote:

   
What year is this going to happen then, Keith? SACD came out in 1999, yet for some reason we aren't yet in a world where equipment is expected to be speced out beyond 20kHz or with 144dB of SNR; why might that be? And I agree with you that various published tests leave something to be desired, but I also think that you should probably also then go around the rest of the forum and ask all these people claiming "huge" differences with hi-res just how well their equipment is reproducing hi-res content. Personally, my E-MU delivers pretty flat performance up to 48kHz, and Sennheiser would have me believe my HD800s are only -3dB at 44.1kHz (which I can't really test myself, unfortunately). Intermodulation tests at 96ksps stay perfectly quiet even at high volumes (not so for 192ksps), so for up to that sample rate I'm at least somewhat sure my ABX results for hi-res testing aren't totally fallacious. I mean, what equipment would you have someone use that you would consider kosher?

 

[KeithEmo]

Absolutely....
 
Most of the arguments that include pictures of jagged stairsteps, with explanations of how such an unruly looking signal can't possibly sound anything other than awful, are either gross exaggerations or deliberate attempts to scare people with bad science, and most of the remaining few are simply poor explanations put forth by folks who don't understand the science themselves. The 16/44k Red Book CD standard provides for excellent sound quality (although you may have to dig around to find CDs that are mastered well enough to prove the point).
 
The audible differences between a high-res version of some particular content, and a Red Book version of it, which is produced from the same master, and with equal care, should be tiny at most. We're talking about (maybe) slight shifts in imaging which (might) be caused by microsecond shifts in phase relationships, and equally subtle alterations in how certain instruments (may) sound due to the loss or alteration of upper harmonics which many people classify as "ultrasonic" and "inaudible". The reason it's difficult to even do reliable comparisons is that the relatively simple process of converting from 192k to 44k, using the best commercial rate-conversion software available, produces differences of similar magnitude. (This makes it very difficult, if not impossible, to separate audible differences that might be due to the different sample rate from the tiny differences that are simply artifacts of the conversion process itself.)
 
Where things get a little tricky is that not all information can be expressed as a frequency. For example, a square wave is a collection of several harmonics, in particular proportions, and in a particular phase relationship to each other. If the frequency response of a device or recording isn't correct, then square waves played through it will both look unusual and sound wrong. However, if the frequency response is correct, but the phase response is wrong, you can end up with a signal that contains all the right amounts of energy at each frequency, but has a waveform that looks nothing like the original. Since our hearing seems to work mostly by analyzing energy like a spectrum analyzer, this difference is mostly inaudible. However, some research suggests that we are in fact somewhat sensitive to some aspects of the shape of the waveform itself - like the precise arrival time of the leading edge of it. One theory is that, even if we include all the audible frequencies up to 20 kHz so that we don't hear anything missing, because the missing harmonics above 20 kHz contribute to the overall shape of the wave, some other aspect of our hearing (perhaps the mechanism that figures out spatial location from phase relationships) may detect that the wave shapes are now incorrect, which may result in a perceived shift in the location of that instrument in the sound stage. In other words, the basic claim is that, even though we don't "hear" sound above 20 kHz, some of that information above 20 kHz does in fact contribute to other things we perceive about the sound - like its location - or even some other as yet not fully defined detail. And so we somehow sense when that information is altered or discarded. There have been some tests that at least suggest that this may happen - but they are far from conclusive. There is also lots of anecdotal evidence that a lot of people claim to hear a subtle difference.
 
However, to put this in "practical perspective", what we're talking about is a tiny bit of information that may be audible and, if it is audible, and if it manages to get picked up in the recording process, may contribute to a slight improvement in realism or fidelity. (Or, to put that perspective differently, I've heard one or two superbly recorded Red Book CDs that sound better than the vast majority of the high-res recordings I own... which proves just how close the capabilities of both formats are.)
 
We are NOT talking about some huge dramatic difference - we're talking about a difference far smaller than, for example, the difference between different speakers, or different phono cartridges.
 
Quote:

We're getting somewhere here though. All those arguments made that hi-res e.g. fills in the gaps in those standard-res "stairsteps" even within 20kHz, arguments about "timing accuracy" of signals within 20kHz, 19kHz signals "fading in and out", etc. etc. At least those can be put behind if we agree that you need reproduced >22kHz signals for the redbook standard to be tested against, no?

 

[RRod]

 

 
Specifics, my good man. I'll agree with you that the final countdown, paradigm-shifting, über-proctored test of hi-res audibility done by whatever society should include a range of equipment randomized across the test subjects. But my question was more: "right now, if you had to make such a list, what would put on it, and would any of that stuff be within the reach of your mid-to-high-but-not-summit-fi member of this forum?"

 

[old tech]

Keith, in regard to one of you points on the methodology I do recall reading somewhere that the paper was criticized by a label on those grounds. The claim is that the SACDs and DVD-As used in the study were produced from 16/44 or 24/44 masters. If there is any truth behind the claim then more serious issues are raised. Have the labels misled consumers? Are they in effect saying that from 1999 to 2007, audiophiles were quite happy with 16/44 as long as they believe it to be hi Res? I think Moran responded to this and other criticisms over the years.

 

[castleofargh]

from my still limited understanding of a full sound system, I'm done with highres. it's as simple as that.  the only time I go get some, is when my friends keep telling me about some latest remaster and it's not available in 16/44.
I never bought a SACD player because I find the tech idiotic and that's my consumer's way to vote, but then again there are a few albums I wish I could get.
I would go back to concern myself with highres the day we get revolutionary improvements on speakers/headphones. when 0.1% THD stops being seen as amazing at the output of a headphone, then in my super basic logic, stuff below will matter again.  I know that's a narrow minded view of sound, not everything below will be masked by distortions, and that there is distortions and distortions. and maybe stuff happening in the time domain I know nothing about. but I just can't get myself to care about what's happening below -96db when I'm using gears that can't guaranty the first 60db to be clean.
 
 
so I agree with you Keith that my own tests only talk about me and my gears at the moment. and I kind of believe that things could be different with better drivers and everything(also it might be wishful thinking), but that's the thing, I don't really expect any revolution in the transducer's domain. they make small progresses, but it's more in how cheaper stuff get closer to the good ones, than the good ones really putting the old good ones to shame. so I fail to have much expectations.
my hopes are toward letting the sound engineers go back to doing their best instead of the loudest and fastest. we all agree on this, but in the world economy few studios will do that. 

and also HRTF and DSPs becoming the usual thing.  but all I see are people who don't understand a thing, crying "stranger danger" anytime someone wants to touch their sound, and begging to be left alone in the 20th century.
 
so I have hopes that sound can get better, but it doesn't involve highres audio and even then I'm not optimistic that it will come soon.

 

[KeithEmo]

Here goes (and you'll notice that most of it isn't expensive). Also, as an industry rep, I'm sort of precluded from "recommending" specific equipment. Again, and to be very clear, this is equipment that I am either both familiar with and consider to be pretty good at revealing differences in other equipment, or that has a reputation for being "neutral" and "revealing". This means that it lacks overwhelming colorations, and seems to me to be pretty close to neutral. (I'm not mentioning equipment that, even though I think it sounds good, doesn't seem to me to be very revealing.) Note that, since we're testing for audible differences, and not for personal opinions of "what sounds good", a lot of the equipment I would recommend would be characterized as "analytical" or "dry" to many audiophiles - which is just fine; after all, we are performing an analytical test.)
 
1) I would use a separate DAC, so I would say that almost ANY disc player would be fine. It should be tested to verify that it actually produces a bit perfect output, and doesn't upsample or otherwise alter the bitstream. I would probably pick the latest bottom Oppo model (at the moment that would be the 103), but anything that delivers bit-perfect response would be fine. Likewise, any computer should work fine as a source. Again, the only requirement would be that the playback software provides a bit-perfect output rather than re-sampling or doing anything else "interesting".
 
2) I would use several middle-of-the-line DACs, and I would specifically avoid DACs with tube output stages, and models otherwise known for various euphonic colorations. Therefore, I would EXCLUDE: DACs with tube output stages, non-oversampling DACs, and DACs with "unusual" procesing built-in. An Emotiva DC-1, or a Schiit Gungnir, or a Wyred4Sound DAC2, or any Benchmark DAC should work fine. (Note that I am NOT excluding Sabre DACs. Even though I believe that most of them introduce some euphonic coloration, I find it to be of the sort that emphasizes differences in other areas rather than covers them up.) If we're using a computer source, which would almost certainly suggest a USB connection, I would specify that the DAC have an ASYNCHRONOUS USB input - to eliminate the major effects of jitter on the input signal. If the DAC includes the option of multiple user-selectable oversampling filters, I would let the test subject pick the one they like - but I would exclude any that specifically produce unusual frequency response (like "slow roll-off filters" which produce a significant drop at 20 kHz) - because they may mask differences that would otherwise be obvious. (It's probably easier just to avoid DACs with selectable filters and frequency responses other than very flat.)
 
3) For speakers, I would strongly suggest models known for being revealing (note that this doesn't specify that they sound "good"). I've found our Emotiva Airmotiv line to be good at this, and I assume that most speakers with folded ribbon (or "true" ribbon) tweeters would be as well. I also assume that most electrostatic speakers would excel in this regard (although I'm not especially familiar with current models, and some are not). Again, I would specifically avoid speakers known for euphonic colorations (like high-efficiency horns, and speakers known to have rolled-off high ends). I would definitely avoid speakers with horn tweeters, vintage models with cone tweeters, and unusual designs with large heavy drivers equalized to produce high frequencies (like Bose 901's) - because I don't think most of them reproduce transients well.
 
4) For amplifiers, I would suggest any "normal" solid state power amp (meaning a current model from.... Crown, Emotiva, Marantz, Parasound, Rotel, etc.). Again, I think most current middle of the road models would work well, but I would avoid tube amps and hybrids with tube stages, which are known to produce euphonic coloration. Since some Class D amps are known for being somewhat colored, and I'm not familiar enough with the various models to list which is which, I would stick with "standard Class A/B amps".  
 
5) Alternately, to avoid the complexity and cost of selecting several good speakers, I would probably prefer headphones. I find electrostatics to be the most revealing: I think Koss ESP-950's would be an excellent choice (since they come with their own amplifier, they also eliminate that variable, and they cost less than $1000). Any of the Stax models I've heard would also work well. (Again, though, if using a separate headphone amplifier, I would avoid using a tube model, because many of them produce various euphonic colorations, which may include "smoothing over" differences in the program source.) Any of the higher-end dynamic models from Sennheiser, or AKG, or Beyerdynamic would probably also work well, although I find electrostatics to be more revealing. I would AVOID planars because, while I think many of them sound quite good, I also find many to be somewhat colored - and in the direction of "smoothing the high end" - which seems to me to be likely to conceal differences in high-frequency and transient response.
 
(You will also notice a preference on my part to choose components that are specifically good at reproducing transients and high-frequencies, and not to say much about low frequency response. This is mostly because, from personal experience, and from my technical theoretical knowledge, I don't expect high-res files to reproduce low frequencies any differently than standard-res files, but I DO expect significant variations in that regard between various speakers and amplifiers. Therefore, I expect any differences that may be audible to be in the areas of high-frequency response or transient response, or related to them.)
 
6) For test content, I personally happen to prefer rock and pop music, so I'm not all that familiar with many classical tracks - or how they should sound. Also, sadly, most rock and pop music isn't mastered especially well (IMHO). However, here are a few of the albums and tracks that I normally use to "show off high-res sound reproduction" and which I think would be good to include:
 
- The Tempest (Reference Recordings - 176k)
- The Eagles - Hotel California (24/192k HDTracks)
- Grateful Dead - Studio Album Remasters - American Beauty (24/192 HDTracks)
(slightly problematic for testing high-res because the new remasters have been "heavily restored", and so sound very different from the originals)
- Kodo - Tataku (SACD)
- Pink Floyd - Dark Side of the Moon (high-res version from the Immersion set)
- Alison Krauss - New Favorite (SACD)
 
 
Quote:

   
Specifics, my good man. I'll agree with you that the final countdown, paradigm-shifting, über-proctored test of hi-res audibility done by whatever society should include a range of equipment randomized across the test subjects. But my question was more: "right now, if you had to make such a list, what would put on it, and would any of that stuff be within the reach of your mid-to-high-but-not-summit-fi member of this forum?"

 

[interpolate]

To be honest guys, you only get what you pay for and the rest is just a commercially induced placebo effect.
 
24-bit is just a binary range of possible values right. In digital it's purely mathematical and has to be converted to analogue again. Where your limitation is the human hearing range.
 
I suppose the best way to look at it is, if you have the studio masters in FLAC or WAV format for example then it's how the artist intended.

 

[KeithEmo]

At one level you're entirely wrong.... or may be. The whole point of this discussion is the question of whether the current digital technology has in fact reached the limit of what we can hear (that's a technical question). For example, if you or I can in fact hear some audible difference between two recordings, or two DACs, then either one of them fails to be better than the limits of our hearing, or both of them fail. (If they sound different, then, at most, ONE of them can be right, and perhaps neither. If we can hear a difference, then both CANNOT be audibly perfect.) You've only reached the level of "just placebo effect" once you've passed that point.
 
However, at another level, you have a point. If the "master itself" isn't much like the "original" (whatever we mean by that) to begin with, then delivering actual fidelity to that master may be sort of moot. Now, when you're talking about physical instruments, then there is some sort of absolute standard... regardless of who's playing it, or how good or bad the composition he's playing is, a recording of a flute should in fact sound like a flute. However, with modern music, often there is no "actual original" - and, in fact, the "band" may have recorded separate tracks, in different studios, in which the "original" only exists in the mind of the mixing engineer, or the instruments themselves may only be "sounds" created in a computer.
 
And, yes, if I were to "Photoshop a picture for a web page", I might not worry about artifacts and flaws that I know won't be visible when people look at that picture scaled to the size of a postage stamp, it's also possible that an artist who knows that his or her recording will be distributed in MP3 format may allow flaws and imperfections such that it really won't sound better when reproduced at 16/44 (or 24/192) - because they know that their audience will never hear it that way. (In fact, for years, some artists have deliberately added "record ticks" to certain tracks, and phony film scratches to videos, for "artistic reasons".) And, in that case, you might fairly argue that there's nothing to be gained by reproducing the poor quality original at better quality.
 
You also get into a sort of philosophical argument in that some artists and engineers may actually create a master that has been "tweaked" specifically "to sound good on certain equipment". For example, a song that has been mixed specifically to sound good on a car radio, may have anomalies designed to "work well with the way a car radio sounds", and may actually only "sound as the engineer intended", when played on a car radio - in which case you might argue that "playing it on a better system will NOT sound as he intended it" - and may actually sound worse. This argument usually comes up in the context of whether it's more accurate to listen to an album using "studio monitors" (so you can hear what the mixing engineer heard), or on home speakers (which is what you can assume the engineer expected you to listen on).
 
(If you were planning to display a painting by Rembrandt, would it be more accurate to display it in "pure white light", or in the light of lamps like those that may have been used in Rembrandt's studio, or in light that matches the light in the room where it was intended to hang when it was originally commissioned? Reasonable arguments can be made for all of those possibilities.)
 
However, to most audiophiles, the overriding idea is that we want to hear the recording exactly as the artist or engineer intended it, and, to be sure we do, we would want to have a system that can reproduce whatever is in that recording with an accuracy that exceeds our ability to notice the differences. (But, yes, the requirements of a system that can achieve that may vary depending on the source itself.)
 
 
Quote:

  To be honest guys, you only get what you pay for and the rest is just a commercially induced placebo effect.
 
24-bit is just a binary range of possible values right. In digital it's purely mathematical and has to be converted to analogue again. Where your limitation is the human hearing range.
 
I suppose the best way to look at it is, if you have the studio masters in FLAC or WAV format for example then it's how the artist intended.