Re: [GnomeMeeting-list] Sound problem under FreeBSD 4.7

["Kevin Oberman" <oberman es net>]

> From: Damien Sandras <dsandras seconix com>
> Sender: gnomemeeting-list-admin gnome org
> Date: 05 Dec 2002 16:35:08 +0100
> 
> Le jeu 05/12/2002 à 16:23, LEZIN Lo a écrit :
> 
> > 
> > Thanks a lot for your help.
> > 
> > But in fact it's very strange :
> > 
> > cat /dev/sndstat gives me :
> > 
> > pcm0: <VIA VT82C686A> at io 0xdc00 irq 11 (1p/1r/0v channels duplex)
> > 
> > So can i think that this means that sound card is full duplex ?
> 
> Yes, but that doesn't mean that the driver is full-duplex.

I have exactly the same problem with FreeBSD 4.7. I see it on three
different systems all running with embedded Intel ICH sound
devices. All are duplex and the drivers claim to be full-duplex. 

> > I don't know how to test record and listen a the same time :-( How can i 
> > test it ?
> 
> First try to record with rec -d /dev/dsp myfile.wav
> of course, change /dev/dsp to your device.

This works fine (as far as I can tell). But when I try to paly back,
the sound is badly garbled! Does what is sometimes called
"motor-boating" and plays back VERY slowly in tenth second (time is an
estimate/guess) intervals, although it is just barely recognizable.

> Then, something like cat /dev/dsp > /dev/dsp should echo everything of
> what you say.

It does, but, like the output of rec(1), it plays back very slowly
like it is playing a tenth second of data, pausing a tenth second and
then playing another tenth second of data. It is just recognizable.

On the other hand, if I use the Gnome Sound Recorder, I can record and
playback just fine. If I play back the .wav file I generated, it
sounds just like when I cat'ed it to /dev/dsp. And, If I play back a
file recored by the Gnome Sound recorder, I hear horrible squawking,
not at all like what I get going the other way.

Clearly there is some sound problem here, but I am very confused by
the conflicting results of my tests. And, when I try to send audio
with GnomeMeeting, I get absolutely no sound, garbled or otherwise.

R. Kevin Oberman, Network Engineer
Energy Sciences Network (ESnet)
Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
E-mail: oberman es net			Phone: +1 510 486-8634