[LAU] Sampling rate and perceived audio quality [WAS]: Re: edirol fa101 on differant rate that 48k ?

[Lorenzo Sutton]

Fons,

fons at kokkinizita.net wrote:
> On Thu, Nov 18, 2010 at 06:29:02AM -0500, Paul Davis wrote:
>> there has never been, to my knowledge, any double blind test that has
>> revealed that any more than a tiny number of individuals (if even
>> that) can hear the difference between the 4X kHz SR range (44.1, 48
>> etc) and the 8X+ kHz SR range (88.2 and above).
> Two years or so ago there was an in interesting 'engineering report'
> published in the AES journal. It reported on the results of a long
> series of listening tests involving hundreds of listeners, all of
> then selected for their interest in high quality audio.
>
> For these tests the authors used 'audiophile' DVD-A recordings
> (mostly classical music and jazz IIRC), all of them 24-bit, 96
> or or 192 kHz, and had the listeners compare them to a version
> transcoded to CD standards (44.1 kHz, 16 bit). Two results
> emerged from this:
>
> 1. Nobody could hear any difference between the original recordings,
> reproduced using the best equipment available, and the transcoded
> versions.
>
> 2. Almost all listeners preferred the 'audiophile' recordings to
> other versions of the same music released on CD.
>
Thanks for this one. I had been looking for this type of more scientific 
information with tests about something I've personally been going a long 
for ages against the statement that "CD is technically crap", that is: 
do some bloody blind tests!

Do you have the reference for this paper? I can't seem to find that 
exact report through google.

> The latter result is quite surprising, but given the first one it
> says nothing at all about the merits of higher sample rates. It is
> just the result of the 'audiophile' recordings (targeted at a very
> critical niche audience) being produced with more attention to audio
> and musical quality than the average CD.

Point 2 makes me think of the fact that someone discovered [citation 
needed] that many so called CD "remasterings" of '70s and '80s records 
have actually been compressed (as in using an audio compressior), in 
turn creating a belief in 'nostalgia audiophiles' that "the vinyl 
sounded better than CD", were probably what is sounding better (despite 
noise and annoying clicks and crackle) is a better S/N ratio which was 
'killed' (or at least strongly reduced) by the compressed ("remastered") 
version (and of course it wouldn't surprise me that "remastering" may 
also have involved some eq etc, but one should do a more scientific 
analysis).

Something else (not related directly to the topic and, again, I would be 
happy to learn if tests exists). One could argue that "some" noise could 
actually make a recording subjectively more pleasing/interesting due to 
the fact that we are immersed in a background noise in everyday life and 
that hearing some of that noise makes us feel more "reassured"... But I 
can see how I'm probably drifting a little  too far here :)

Lorenzo.