[bigshot]

If you want to do a comparison test for yourself, get two copies of a DSD hybrid SACD by a company that doesn't release CDs. Then run them into a line level balanced switcher and go to town.

I did this. I used a Pentatone SACD of Stravinsky by Jaarvi. Lined them up and there was absolutely no audible difference.

Comparing legacy titles was all over the map. Some had completely different masterings on the cd layer as on the SACD layer. The SACD wasn't a guarantee of better mastering either. Some of the classic Rolling Stones tracks were completely remixed (poorly). They might have sounded cleaner to people who had never heard them before, but they didn't sound like the Stones.

I gave up on SACD. My player is in a closet now. The crap shoot as to whether a SACD would be well mastered was just the same as the crap shoot with CD. No reason to buy the same albums again on a doomed format.

 

[Prog Rock Man]

Thanks Albedo. So to counter the opening study here is a blind test which finds there is a difference between CD and SACD
 
"It is currently common practice for sound engineers to record digital music using high-resolution formats, and then down sample the files to 44.1kHz for commercial release. This study aims at investigating whether listeners can perceive differences between musical files recorded at 44.1kHz and 88.2kHz with the same analog chain and type of AD-converter. Sixteen expert listeners were asked to compare 3 versions (44.1kHz, 88.2kHz and the 88.2kHz version down-sampled to 44.1kHz) of 5 musical excerpts in a blind ABX task. Overall, participants were able to discriminate between files recorded at 88.2kHz and their 44.1kHz down-sampled version. Furthermore, for the orchestral excerpt, they were able to discriminate between files recorded at 88.2kHz and files recorded at 44.1kHz."
 
 

 

[nick_charles]

 
Quote:

Thanks Albedo. So to counter the opening study here is a blind test which finds there is a difference between CD and SACD
 
"It is currently common practice for sound engineers to record digital music using high-resolution formats, and then down sample the files to 44.1kHz for commercial release. This study aims at investigating whether listeners can perceive differences between musical files recorded at 44.1kHz and 88.2kHz with the same analog chain and type of AD-converter. Sixteen expert listeners were asked to compare 3 versions (44.1kHz, 88.2kHz and the 88.2kHz version down-sampled to 44.1kHz) of 5 musical excerpts in a blind ABX task. Overall, participants were able to discriminate between files recorded at 88.2kHz and their 44.1kHz down-sampled version. Furthermore, for the orchestral excerpt, they were able to discriminate between files recorded at 88.2kHz and files recorded at 44.1kHz."
 
 

PRM and others - before placing too much faith in this paper (Pras) which I do own along with the Meyer and Moran paper - I suggest you get a full copy and read it all - not just the abstract. The stats are extremely iffy and shows signs of cherry-picking data. They used a two-tailed T-test which is inappropriate for this test of the ability to correctly discriminate and lowers the bar by 50%. They unilaterally drop 3 outliers from their already very small sample which skews the results and none of their individuals managed a statistically significant ability to correctly detect the difference between modes.
 
The M and M is methodologically far better and uses a bigger sample and a more sensible (simpler) comparison protocol with more iterations of the comparisons.
 
The critique of M and M is that not all samples used were native high res. This is a fair criticism. However some **were** high res native and for no sample and no individual was there a reliable detection of the degradation when a 16 bit stage was included. M and M have (sadly) not released their raw data so it is not possible to know how often native low res and native high res samples were used.
 
 
EDIT: 5/19 samples used were not native high res the other 14 were and the bulk of the testing was (according to Moran) done with the Chesky samplers which were definitively High-res
 
Quote:

 
Nevertheless, to address this question we added up all the trials where the original sources were very recent recordings touted as being of demonstration quality from their labels (Chesky, Telarc, ECM, Turtle and Kimber Kable). The average correct score for this group was 45.4% (109/240), slightly lower than our overall average...

 
 
 

 

[Prog Rock Man]

Thanks Nick_James. So the M & M paper is less flawed.

 

[nick_charles]

Quote:

Thanks Nick_James. So the M & M paper is less flawed.

In my opinion as a professional reviewer of academic papers, yes. But it is always best to read these things yourself and make your own mind up. This means joining the AES I am afraid as both papers are copyrighted and the Pras paper has not yet been linked to elsewhere

 
but in general just reading abstracts can be very misleading !