PDA

View Full Version : Update on s/pdif issues



Dan Foxley
2004-02-16, 21:50
Thanks for the update, I'm anxiously waiting to go digital...the m-audio
superdac 2496 doesn't work either.

dan

-----Original Message-----
From: discuss-bounces (AT) lists (DOT) slimdevices.com
[mailto:discuss-bounces (AT) lists (DOT) slimdevices.com] On Behalf Of Sean Adams
Sent: Monday, February 16, 2004 12:13 PM
To: Slim Devices Discussion
Subject: [slim] Update on s/pdif issues



B&K:

Just doesn't work with any format. Got a reference 50 receiver here and
confirmed this. One thing that's interesting though is that if you connect a
s/pdif cable from a CD player, and then quickly switch it over to
squeezebox, the squeezebox does play for about 1 to 5 seconds before the
input switches off again. I believe the problem is that it doesn't like
something in the control bits that we're outputting in the s/pdif stream.
I've ordered some equipment which will analyze our s/pdif output and that of
other products, so we can find out exactly what it is that this receiver
does not like. I called B&K and they confirmed that the receiver does not
use an off-the-shelf s/pdif receiver chip (eg Crystal, Burr-brown), but
custom logic that they've done in a xilinx chip. This means I can't look up
the specs for the s/pdif receiver and will have to treat it like a "black
box" and just figure out what works. Will have more info on this later in
the week when I get the testing equipment.

Track startup delay:

This issue is only for MP3 format, only with some receivers. The symptom is
that the start of tracks will be cut off by about 0.5 sec.
We will have a fix for this soon. For PCM formats, squeezebox should already
be outputting 44.1KHz silence correctly between tracks - if you are hearing
otherwise, please let me know. (Please note there was an unrelated streaming
issue which could cause similar symptoms when the server is running on
Windows - it has been fixed already in 5.1.)

Wrong sample rate for WAV/AIFF:

Using a frequency counter on the internal frame clock signal, confirmed
wrong sample rate of 44.19 KHz. For MP3 the sample clock is correct at 44.10
KHz. This is easy to fix in software - it's an error in our calculation of
the PLL offset for the clock. Will have a firmware update to address this
ASAP.

Meridian speakers:

Need more information. Could somebody please confirm whether these work for
mp3s but not WAV/AIFF? If that's the case, then it's probably just the error
in the sample clock. Otherwise it's probably something more like the B&K
issue.

Inverted signal (*-1; not swapped channels) for WAV format

Confirmed - easy fix server-side.

Sample clock off-by-one (right channel shifted by one sample WRT the left
channel)

Fixed in 5.1

Channels swapped in some formats:

Fixed in 5.1