What did you do for controls, Stefan? Unless you controlled for level
and experienced all of this double blind, it is likely that you are
experiencing placebo effect on some of these, particularly those
changes where the squeezebox or transporter is doing the final retiming
and conversion. 

Bits are bits. If you keep the Squeezebox fed and are feeding it the
same format, it should not care where the bits came from. After all,
the BBC makes it all the way to you from the UK with a second of
latency and maybe 0.1 second of latency jitter. Squeezecenter can
buffer it and spool it to Squeezebox which has yet another jitter
buffer. Squeezebox reconstructs and retimes the sample stream. All the
jitter in the world (well until the buffer runs dry :) never reaches
your ear, just that introduced by the Squeezebox itself.

With this understanding, moving your music library from an internal
disk to a NAS should have no effect on the sound coming out of the
Squeezebox or Transporter. Some of the other changes such as SPDIF to
TOSLink may. SPDIF is notorious for its jitter. But again, we are
filling a buffer and regenerating the timing. I can see a difference
resulting if the jitter is so bad that a bit is lost every now and
again. This would cause actual errors or dropouts in the sample
stream.

The way around this is to use "forward error correction" by using
Hamming coding or one of the newer forward error correcting techniques
like those used in disk drives. Even if a bit is lost every now and
again, the bit-stream can be recovered. I don't really know the details
of SPDIF, only that it dates back to the beginnings of digital audio
production equipment and is obsolescent.

Bit jitter should not have a big effect on playback as long as all of
the bits can be recovered correctly off the SPDIF waveform. The bit
stream can always be buffered and retimed. The last device to buffer
and retime the bit stream determines the pitch accuracy and sample
jitter. What happens along the way is not seen unless the pipeline goes
dry.

Sample jitter is interesting. From my introduction to digital signal
processing in college, jitter (non-uniform sampling in geek speak)
introduces noise in the reconstructed audio. A classmate did his
masters thesis on the subject deriving the effects of nonuniform
sampling. My recollection is that he concluded that the noise was
Gaussian for the sample jitter distribution that he examined. This
would show up as a raised noise floor in the playback. It would be hard
to notice a tiny amount of jitter on most program material.

FLAC and Apple Lossless do give you back the original bits. Folks have
verified this. I ripped an CD to one of each, loaded into Audacity, and
played the tracks. I couldn't hear a difference. There can be a
difference if you are not normalizing the levels or you are using
different play back level management schemes. These can cause the
levels to differ enough to fool you.

I demonstrated placebo effect to myself some years ago. I bought some
Nordost cables (the modest ones) to replace some way too long monster
cable that came with my trusty DQ-10's in the early days of Monster.
The difference was dramatic after the change. The next day, things were
back to normal. I expected a change and, by Jove, I heard one. I'm not
so quick to form opinions about a change after that experience.

Interestingly, the woofer surrounds failed and I never noticed. I had a
local pro audio shop renew the surrounds and the DQ-10s are still going
strong at 30 years young. It has been 13 years so I've forgotten what
prompted me to take the front panels off to look.

My rig: SB3, Conrad Johnson PV-1, Gas Ampzilla, Dhalquist DQ-10s. 
Computer audio: Presonus 1394 and Monsoon Audio planar desktop
speakers.


-- 
dlhamby
------------------------------------------------------------------------
dlhamby's Profile: http://forums.slimdevices.com/member.php?userid=12801
View this thread: http://forums.slimdevices.com/showthread.php?t=57449

_______________________________________________
audiophiles mailing list
[email protected]
http://lists.slimdevices.com/lists/listinfo/audiophiles

Reply via email to