Le Sun, 22 Sep 2002 13:16:41 +0000, Ben Reser a écrit :

> On Sun, Sep 22, 2002 at 09:15:43PM +0200, Frederic Crozat wrote:
>> Le Sun, 22 Sep 2002 11:23:42 +0000, Ben Reser a écrit :
>> 
>> > On Sun, Sep 22, 2002 at 11:52:54AM +0200, Frederic Crozat wrote:
>> >> My bet would be a bug in VIA8233 sound driver.. There is absolutely
>> >> NO code for one specific soundcard in esd.. Try switching to OSS (or
>> >> to ALSA) to see if bug is still here..
>> > 
>> > Bull hokey Frederic.  If it was a bug in the sound driver then I'd get
>> > the same problem with artsd.  And I don't.  This *ONLY* happens with
>> > ESD.  And while ESD doesn't have code for specific sound cards in it. 
>> > I do believe it has code to determine the playback rate automatically.
>> > This is more than likely what is bugged...
>> 
>> Because it doesn't happen in arts doesn't mean the driver is bug-free..
>> ARTS and ESD don't use exactly OSS driver the same way..
>> 
>> The only oss specific code in ESD is about setting the playback rate..
>> and by default, it is 44.1 kHz..
> 
> I'm not using the OSS driver because there is no OSS driver for VIA 8233. 
> Only VIA 686.  Which is not the same thing.  So there is no way to test it
> by changing drivers.

There is one way : recompile esd with alsa support => it will bypass ALSA
OSS emulation..

> Switching between ALSA and OSS for ARTS has no apparent effect on the
> sound quality.  But ESD ALWAYS outputs digital garbage.

ESD always uses OSS interface, even for ALSA drivers.. 
I think ARTS is able to directly use ALSA backend when it is available..

> I guess I find it really hard to believe that ARTS OSS output is so
> terribly different from ESD's that it should cause a driver bug.

Trust me on that :))
We had to integrate some "hacks" on ESD because some OSS drivers weren't
working correctly when set to "pause" just after being opened.. That is
why I really thing this is a driver bug..



-- 
Fr�d�ric Crozat
MandrakeSoft


Reply via email to