Simon Wrote: 
> Is streaming the data across to the SB2 much the same as reading a CD? 
> If so the differences in sound quality between different transports is
> well documented; disk copies never play as well as their orginal
> counterparts, and dirty CDs also make for poor playback, although you
> could argue that data is data.  Passing all those ones and zeros around
> is never completely without loss and error.
I would bear in mind that a lot more is happening in a normal CD
transport.  Yes it is full of digital logic, but it also has high
current motors and servos to rotate the CD, move the pickup to the
right place, focus the pickup etc, plus sensitive analog circuits to
read the disk.  

The difference between transports in my view occurs for at two places:

1) Reading the analog pattern on the CD and turning it into 0s and 1s. 
The CD contains different length pits/lands which equate to 0s and 1s,
but reading them is an analog process - scratches on the disk, poor
power supplies [especially noise from the motors & servos] etc mean
that there is distinct probablity of noise on the signal which lead to
reading errors.  Now the CD standard includes error correcting codes so
that some degree of errors can be tollerated and the correct bit stream
reproduced even though some is missing.  Other errors can't be
corrected and it transport specific how it 'fills in the gaps' to make
up the missing data.  Hence differences occur in the both the analog
domain reading the disk and in the signal processing of errors. 

2) Getting the data out of the transport to the dac.  In the case of a
CD transport and with a separate DAC connected via spdif, clocking out
the data such that the DAC receives it with minimal minimal jitter is
critical.  Depending on the DAC it is likely to use the transport
stream directly as the source for the critical clock that it uses to
decide the instance in time to turn each digital sample back into
analog.  Hence the quality of the transport clock and the spdif
interface circuit which is used to get this to the dac can produce
significant differences between tranports. 

Now for PC audio and SBs - all of 1) is done at ripping time and hence
is not relavent in discussing SB2.  [but does mean that an accurate
rip, e.g. using EAC is important] If we assume that all signal
processing can be done without creating bit errors, then the only thing
that remains is 2).  [If bit errors occur then the entire computer
industry would have a problem...  Well they do with 'cosmic ray events'
and other very rare bit flips, but thats why ECC memory was invented -
these are so low rate they have no impact on this case.]

Hence if you are using SB2 as a transport, and your rips are accurate I
would expect the differences in sound quality to be down to how your DAC
interfaces to SB2.  It is reasonable to expect data to be data and for
the right bits to turn up at the DAC all the time.  The critical bit
will be whether it turns them back into analog at the right instance in
time...

Adrian


-- 
Triode
_______________________________________________
ripping mailing list
[email protected]
http://lists.slimdevices.com/lists/listinfo/ripping

Reply via email to