[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: Sluggish audio



On Tuesday 06 July 2004 04:50, Vincent Cojot wrote:
> Just to add my two cents. I too had a setup with RH7.3 which worked fine.
>
> After some googling and with the following in /etc/sysctl.conf, I still
> get the same behaviour than what you observed:
>
> # See also:
> http://bugzilla.redhat.com/bugzilla/long_list.cgi?buglist=114553
> vm.inactive_clean_percent = 100
>
> Just scrolling up and own the tabbed window inside mozilla hangs xmms
> until all scrolling has stopped. Not that I -need- to play sound under
> RHEL but it'd be nice to standardize on RHEL3 for desktops and
> workstations alike and desktop problems could be a show stopper in some
> cases..

Indeed, this could be a show stopper when RHEL 3 is used mainly as desktop. 

I tried 

I tried installing ALSA driver manually last night, and now I'm running xmms 
using ALSA. I don't know how, but the performance is much much better. I 
tried to reproduce all the stutter problem and it hasn't came up yet (Okay, I 
thought I heard a very slight stutter once when I load a large web page, but 
it's much more acceptable than the shipped kernel driver). Also tried 
reproduce the problem with the steps from bugzilla entry, and ALSA seems to 
still works fine.

So I guess this is the solution for me for now, although it would be kind of a 
pain that I have to rebuilt ALSA with every kernel update.

> Also, the integrated sound card on which the problem was reproduced is
> different from yours (it's in kernel*, not kernel-unsupported):
>
> [me where code]$ lsmod|egrep -i '(sound|i810|audio)'
> i810_audio             27048   1 (autoclean)
> ac97_codec             15624   0 (autoclean) [i810_audio]
> soundcore               6340   2 (autoclean) [i810_audio]

Do you have uniprocessor on this machine? My other server (RHEL 3 also), a 
dual CPU Athlon, uses the same sound module with its integrated sound card. A 
quick check on that machine with xmms didn't show the problem. I also read in 
one of the bugzilla entries that this problem under RH 9 kernel went away 
when the person added a second processor to his machine.

Thanks a lot for the info.

RDB



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]