Interesting article, and fairly old: http://theaudiocritic.com/plog/index.php?op=ViewArticle&articleId=4&blogId=1
That is a silly conclusion to a test that shows something quite different. Maybe I’ll hit that in tomorrow’s post. Thanks for posting this.
That is a silly conclusion to a test that shows something quite different. Maybe I'll hit that in tomorrow's post. Thanks for posting this.
Indeed!
What cables did they use? =))
God makes Gaussian noise. If in the average people do not hear any difference, there must be some percentage of population that do. It’s like the average temperature of patients in hospitals 
But anyway, this paragraph sums it all up:
“Plausible reasons for the remarkable sound quality of these [hi-rez] recordings emerged in discussions with some of the engineers currently working on such projects. This portion of the business is a niche market in which the end users are preselected, both for their aural acuity and for their willingness to buy expensive equipment, set it up correctly, and listen carefully in a low-noise environment. Partly because these recordings have not captured a large portion of the consumer market for music, engineers and producers are being given the freedom to produce recordings that sound as good as they can make them, without having to compress or equalize the signal to suit lesser systems and casual listening conditions. These recordings seem to have been made with great care and manifest affection, by engineers trying to please themselves and their peers. They sound like it, label after label. High-resolution audio discs do not have the overwhelming majority of the program material crammed into the top 20 (or even 10) dB of the available dynamic range, as so many CDs today do.”
Paul, in your newsletter today regarding this topic, if I understood you correctly, you state that the conclusion is not really supported by the experiment.
This experiment takes an analogue signal from high end SACD player, digitizes it with 16/44 and then converts it back to analogue. I agree with you that the resulting sound will have some characteristics of the analogue sound produced by SACD, however there will be information loss during 16/44 conversion. The experiment is proving that this information loss is not audible.
I can clearly hear difference between CD and SACD/HiRes in my system and it makes it hard to believe the article. However, can you please explain your logic?
Cheers, Mikhail
mikhail said: I can clearly hear difference between CD and SACD/HiRes
The article explains it by different mastering process. Also, according to some audio magazines (AFAIR Stereophile is one of them) the CD-layers (on some hybrid SACD/CD disks) are intentionally made with a lesser quality to make the difference more obvious.
And again, some people can clearly hear difference where other people can't. Some do not care and subconsciously ignore the difference, some do care and subconsciously hear the difference where it does not exist. And reliable ABX is not always possible. Go figure... But at the end we are on our own with our money ;)
In a reply to the comments section to the Paul’s Posts, Paul states:
"None of this proves anything other than RBCD is capable of recording the analog output of a CD, LP, or SACD well enough most average listeners could not tell the difference between the original and the copy of the analog output."
The problem with this position is that the test subjects were 60 members of the Boston Audio Society, audio professionals, “serious students of the art,” and other interested parties. These are hardly “average listeners.” Moreover, they used high quality equipment. It took over a year as the ABX were as long as the listeners desired.
To whit: “[N]one of the more elaborate and expensive playback systems (for which the subjects were all dedicated amateur audiophiles, active students in a professional recording program, and/or experienced working professionals) revealed detectable differences on music.”
“We have analyzed all of the test data by type of music and specific program; type of high-resolution technology; age of recording; and listener age, gender, experience, and hearing bandwidth. None of these variables have shown any correlation with the results, or any difference between the answers and coin-flip results.”
The test, rigorous and blind, supports the conclusion that there is no discernible difference between SACD and Redbook; that is, any loss of resolution resulting from using 16/44 is inaudible.
Again, the analogy of high definition to standard definition television is useful. The average ten year-old can easily distinguish between standard and high definition television. If a high definition analog signal is recorded to VCR (or DVD recorder) and played back on a high def monitor, the same ten year-old can readily see and report the diminution in quality. If one could see no difference, we could fairly conclude Blu-Ray is no better than VCR.
Here, we have a controlled study finding that experienced listeners listening to hig hquality equioment cannot hear the equivalent difference between SACD, and SACD recorded and played back on Redbook CD. This is telling.
Elk, this is exactly my understanding of the experiment. I went through the blog discussion and I do not understand Paul’s logic.
However, every bit of our equipment has it’s own sound. I really enjoy my PWD and my Sony 777ES. It is fun to listen to the same tunes and enjoy different aspects. I do trust that the experiment was not biased and properly setup. And I still have hard time believing it listening to 16/44 and HiRes. BTW, I do not think my listening skills are only average.
I, too, find there is a difference in sound between my PWD and my Sony SCD-XA777ES (wonderful SACD/CD player). There is a lot of hardware differences however.
I have not recorded an SACD analog stream to 44/16 and then compared the two. I am now curious to try.
The difficulty is expectation bias is a huge, real problem when trying to informally compare. We often “hear” differences and improvements which do not really exist.
Well I think this just shows flawed testing and or listening (or simply that most people’s brains are not familiar with subtle sonic differences. I have the same Chesky recordings (Jen Chapin) in 96k then I bought again in 192K when it was released. I can definitely hear the differences, and that should be much more subtle than 44.1 and 16 bits VS. SACD. I can hear differences in different Hi res sources (like flac made from an SACD rip vs. an HD tracks release of the same material.) I hear subtle clipping (or rather hardness) in mic preamps as well that were recorded a tiny bit too hot.
I know that some engineers state that 24 bit is more important than a higher sample rate. I would also think the analog output of an SACD player could be an area of loss.
I like many of you have taken years to cultivate our aural memories and judgments (I often calibrate my ears with life acoustic music played in home settings) That is not a skill to be taken lightly. Look how many of us think NativeX is such an improvement over Native, and are willing to pay dearly for that improvement, but it’s quite possible that 99.5% of people could not hear the differences.
As part of my post production business, I color correct video, and often the subtle hue shift I A/B for a client is unnoticed, but I can easily tell. Were they not back to back, I probably wouldn’t be able to though. Another thing that completely annoys me is watching a film or video where the aspect ratio is off for some reason or another, or even cropped from 2.35 to 16X9, loosing the sides. I often can’t even watch a film like this on cable because I am so cognoscente of what is missing. I am the average person, or even average film professional? No, certainly my “E-meter” is set way too high in comparison to most others, but it doesn’t mean that differences don’t exist.
I didn’t read the system they tested on, but then again when I hear most high end rigs at shows I find them colored in some way. So it’s quite possible there are on a system that while considered high end, is no where near as resolving as mine or many of yours, or Paul’s system.
It could also be that it takes time to calibrate ones ears to a system. I can remember listing to a system with unfamiliar music and thinking it sounds good, then once hearing one of my reference tracks, I can hear the colorations within 20 seconds.
In fact the TAD CR-1 wasn’t even good enough for me “as is” so I recently added the Enigmacoustics Sopranino Electrostatic supertweeter and a pair of Velodyne DD+10’s. (plug note - I recently became a dealer for each)
So yes, we on this this board are rare birds, but saying because mere humans don’t see the color variation of our feathers, doesn’t mean the subtle shading we know are there, don’t exist.
Added note: Has anyone noticed that after intense listening sessions, you are hungry? The brain uses large amounts of energy when concentrating and I’m sure brain imaging could show that audiophile’s have increased activity and perhaps larger areas devoted to audition than other people, and perhaps even more reward center being tied to auditory experiences.
emailists said: I know that some engineers state that 24 bit is more important than a higher sample rate.
I prefer upsampling to the native. So does dCS, hence they use upsampling to 24/192 or DSD. And I do hear the difference, even if it's subtle. I do hear difference between CDR's written with different speeds, I do hear difference between CD's statically charged and treated with distilled water. I do not hear any difference between cable stands. Etc... We had a long discussion about testing in the PCM vs DSD thread, so not going to repeat myself again :D
Only we can decide if we hear any difference or improvements. However, it's not always possible to test "this" or "that" in our own systems, hence we have to rely upon "other ears" and then trust them or not. Vicious circle...
emailists said: The brain uses large amounts of energy when concentrating
I wonder if a proper qi-gong can help to treat digititis. Qi-gong vs Schumann resonator? :D Tea vs coffee. Whisky vs vodka. (I know, LSD will win anyway). Perception 1 vs perception 2. Something for this subforum http://www.psaudio.com/vanilla/categories/psychoacoustics-and-corn-flakes ?
Yeah, it would be nice to have something repeatable and measurable with probability a bit far from 50% than 47%. But even in this case there will be someone who will try to measure voltage with a ruler and state that "there is no difference" ;)
And finally - can we calibrate our brains?
emailists said: I didn't read the system they tested on . . .
Four systems, including a mastering studio system, a dedicated university audio program system (University of Massachusetts - using students in their recording program as subjects) and a high end home setup (custom-built listening room, Quad ESL 989, Nordost cables, Conrad-Johnson pre, other goodies).
emailists said: It could also be that it takes time to calibrate ones ears to a system.
Potentially yes, and one of the reasons the test took over a year. This is one of the reasons they used calibrated system levels (85dB, wideband). Additionally, there were no time limits on ABX listening - short, long, your choice. Finally, the subjects were even allowed to choose source material and specific musical passages he or she thought were the most revealing.
The above two observations/arguments highlight the fact that, at best, the improvements with high resolution recordings are exceedingly subtle. Apparently the admittedly minute improvements can only be realized with specialized equipment, ideal listening conditions, time tempering, and calibrated ears. Are the improvements really this fragile and difficult to detect even by experienced, professional and/or enthusiast listeners?
Again contrast this with Blu-Ray v. DVD v. VCR. Each is a major improvement, not requiring calibrated eyes, long familiarity with the specific monitor and player, lighting of the room, etc.
emailists said: I have the same Chesky recordings (Jen Chapin) in 96k then I bought again in 192K when it was released.
The test included a couple of Chesky SACDs. :)
I suggest listening with an ABX comparator (there is a plug-in for Foobar, for example). It is sobering to learn you don't hear what you think you hear, or that the difference is exceedingly difficult to detect.
Alekz said: Only we can decide if we hear any difference or improvements.
Indeed. We each hear what we believe we hear. If a player, recording, tweak, format sounds better to you, it is better.
These discussions remind me of auto enthusiasts having invested $3,000 in a special exhaust. The car sounds different, of course. The owner swears his butt-dyno reveals massive power gains. The drag strip and actual physical dyno proves otherwise. The owner still argues there is a huge power improvement. Of course, if he thinks it feels more powerful - it is. :)
I forgot to note that claiming this study “proves” CD sounds the same as high resolution formats is ludicrous. It is some data, not proof of anything.
Note, also, the report is modestly titled: “Audibility of a CD-Standard A/D/A Loop Inserted into High-Resolution Audio Playback.” It points only to a conclusion that a 44/16 recording of SACD may sound no different than the original SACD playback.
Elk said: It is some data, not proof of anything.
Yes, my point as well.
Elk said: Audibility of a CD-Standard A/D/A Loop Inserted into High-Resolution Audio Playback.
Another question, I could not find an answer to (or missed it). Was that A/D/A loop constantly in the audio chain? Were they only switching the resolution (e.g. between 24/192 and 16/44)?
If there were two different chains (one through the A/D/A converter and one around it) I would doubt the whole method at once.
Alekz said: If there were two different chains (one through the A/D/A converter and one around it) I would doubt the whole method at once.
The high resolution signal went around the converter. This *favors* SACD and would increase the likelihood a difference would be detected between SACD and 44/16.
The entire point of the study was to determine whether the A/D/A 44/16 "bottleneck" was audible. If SACD can capture something 44/16 cannot, one should easily hear the bottleneck as a diminution of quality - but no one did.
This provides substantial evidence that SACD is not carrying any more meaningful, audible information than CD.
Hi All,
Correct me if I am wrong as I haven’t read the full article (only the article in the first post here) there appears to be two different situations/issues here.
The first situation/issue:
What they “proved” is that a source that has already had an analog to digital (AD) conversion (onto the SACD in the first place), then a Digital to Analog (DA) conversion (output from the SACD player) then goes on to sound the same as a file that goes through a further 44.1/16bit AD to DA conversion.
This entire test relies on the quality of both the original AD conversion and the quality of the DA in the SACD player (which I am sure was of high quality conversion, but still has been converted twice). This is one situation.
The second situation (which is quite different from the first):
If they had taken an actual analog source (such as real to real master tape) and converted it to 44.1kHz, 96kHz and also 192kHz I would think they may have found/heard differently. This is what I understand happens in a lot of the Hi-Res digital files being re-released today.
A lot of digital recording studios now days are recording at the higher sample rates (96khz or 192kHz) straight from the microphones with High Quality AD conversion.
In my days of past fiddling round with a small home studio (and it is alluded to in the article) I found that when recording the original signal from microphones to the different formats quite different, with the 96Khz being the best in terms of headroom and background noise (the 96kHz was the limit of my AD converters, didn’t have 192khz capability).
Hope that helps, Cheers,
Paul