Steven Sullivan wrote:

Michel Hafner wrote:
s1.newsguy.com
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Path:
NNTP-Posting-Host: 213.200.229.23
X-Original-NNTP-Posting-Host: 213.200.229.23
X-Trace: news.cybercity.ch 1075757204 213.200.229.23 (2 Feb 2004 22:26:44 +0100)
Lines: 49
X-Original-NNTP-Posting-Host: 127.0.0.1

Harry Lavo wrote:
That does not really address my point that there is no audible difference
between 24/96 and 24/192 if you remove AD and DA issues.


Hey, you've got to get it from analgo to digital and back somehow. You
can't escape AD/DA issues.

Of course not. But my point/hypothesis is that 24/96 and 24/192 are better
formats to store the musical information than 1 bit DSD at 64 times 44.1 Khz.
And 24/96 sounds the same to human ears as 24/192 as far as the musical
information in the signal is concerned.
To investigate this you need to separate the AD and DA from the format.
So if people say SACD sounds closer to the analogue master we need to know
if that is (if it's true) because the AD is better and/or the DA is better
and/or there is more information in the signal.
Points to investigate a
a) is PCM 24/96 and/or PCM 24/192 transparent relative to DSD master.
b) is DSD transparent relative to PCM 24/96 and/or 24/192
for a) you take a DSD master, convert it digitally to PCM and convert it
digitally back to DSD. Listen to it via DSD hardware DA before and after.
Can you tell it apart in double blind testing? This establishes if PCM
is transparent relative to DSD using DSD DA. Can you tell apart the DSD
and the PCM? This establishes if PCM is transparent relative to DSD via
PCM DA.

etc.

Only if you know for sure that the conversion process used
was transparent. If you confirm a difference,
how do you definitievely rule out implementation
issues?

Yes, the conversion itself should be transparent. The mathematical
properties of the filters need to be well understood first.