I've received more than one suggestion that I run the CDROM's digital output
through my sound card and somehow the digital output from the sound card
will be valid when the CDROM is not playing.  I have a DIO2448, which does
have a header for the CDROM's digital signal.  Running it though the sound
card makes no difference.  To get the CDROMs signal to the TOSlink output on
the card, the card must be in "pass through" mode, where it is routing one
of its digital inputs to the digital outputs.  In this mode, the output is
not valid unless the input is valid.

The basic problem that seems to be confusing everyone is that the S/PDIF
signal is a *synchronous* signal where the clock and data are encoded onto
the same signal.  (The technique they use is called Manchester encoding.)
The clock must be extracted at the receiving device using a
phase-locked-loop.  This has two problems:

1) The PLL has a finite lock time, so you end up missing the first several
samples before lock occurs.  Remember, this is *synchrounous*, unlike normal
RS232, which is asynchronous acquires sync on every byte.
2) The extracted clock may be unstable, depending on the data.  Certain data
patterns are worse than others.  Stability is improved by increasing the
lock time.  Good receivers use a two speed lock algorithm: high speed when
acquiring lock and slow speed to improve stability, once lock is acquired.

There is no magic that the soundcard can do to acquire lock that the MD
cannot do.  The damage caused by interrupting the S/PDIF stream is
irreversable.  The bottom line is: it takes a finite amount of time to
re-acquire lock, regardless of whether its a sound card, an MD, a DAC or
whatever...

-----------------------------------------------------------------
To stop getting this list send a message containing just the word
"unsubscribe" to [EMAIL PROTECTED]

Reply via email to