Re: sound issues in -CURRENT

On Tue, 9 Mar 2004 07:05:24 -0800
I wrote:

> hello, i've searched the web for hours and hours and only found one
> hint of evidence that i'm not the only one with the following problem.
> i found this post from several months ago that details the same
> problem i'm having:
> 
> --- snip ---
> 
> On Sat, 30 Aug 2003 18:45:49 +1000 
> matti k <overclocker_at_optusnet.com.au> wrote: 
> 
> 
> > On Fri, 29 Aug 2003 12:36:56 +0200 
> > Eirik Oeverby <ltning_at_anduin.net> wrote: 
> >
> > > I might also point out that there are two distinct kinds of 
> > > distortion happening: The click/pop/crackle kind of distortion, 
> > > and one where a sound buffer is repeated over and over 5-10 times 
> > > - as if the hardware has some kind of hickup. 
> > > 
> > > Again - this *only* happens in -CURRENT, not in -STABLE or in any 
> > > other OS (tested OS/2, BeOS, Linux, winXX). 
> > > 
> > > Anyone? :) 
> > > 
> > > /Eirik 
> > 
> > I get the same pops/crackling but using a SB Live! 
> > I think it started happening when I switched from 4.x to 5.x 
> > The strange thing is that sound works fine after a reboot but some 
> > hours after that I get the noises. It only happens with music 
> > (mp3,ogg). 
> 
> 
> I tried a sound card with a CS4630 chip and still got the same thing. 
> Strange that it takes a while for the noises to appear. Anyway, I was 
> fiddling with settings using XMMS to play mp3's and found changing 
> either "resolution" from 16bit to 8bit or changing "down sample" from 
> 1:1(44khz) to 1:2(22khz) fixes the problem. These settings are located
> under audio i/o plugins -> input plugins -> mpeg layer 1/2/3 player ->
> configure -> decoder. Also after using these settings (1:2(22khz)) for
> half an hour and then changing them back to defaults ... sound was 
> good. I should add that running gkrellm (/ports/sysutils/gkrellm) 
> _and_ having the cpu monitor enabled makes the noises much worse. 
> Again, on my system this only occurs after several hours and of 
> course if I reboot all is well. If nothing else at least I've found a 
> way to avoid rebooting, I guess.
> 
> --- snip ---
> 
> the only difference is that i don't experience the looping sound
> problem. the sound card i have is a sound blaster live 5.1 and i've
> also tried a hercules muse xl, both of which experience the problem
> after several hours of operation. i've tried the gkrellm quick fix and
> disabled several krells that seems to help. the downsample thing seems
> to work too. i'm just wondering if there has been any development on
> this issue since 5.2.1-RELEASE (what i'm currently using). here is a
> full list of my hardware:
> 
> - AMD Duron(tm) Processor _at_ 1.2GHz
> - 128MB PC133 RAM, 256MB PC133 RAM, totalling 384MB
> - 4.3GB Fujitsu hard drive (Ultra-ATA 66)
> - 20.2GB Maxtor hard drive (Ultra-ATA 100)
> - 50x max IDE CD-ROM drive
> - 40x12x48 LITE-ON CD-R/RW drive
> - SiS 7012 PCI sound card (integrated, disabled)
> - Creative SoundBlaster Live! 5.1
> - GeForce 4 Ti 4200 w/64MB DDR VRAM w/TV-out (latest nvidia drivers
> - SiS900 10/100 ethernet card (integrated)
> 
> thanks,
> 
> - ryan

well, i recompiled my kernel this morning with the following option
changes:

got rid of:
options DISABLE_PSE
options DISABLE_PG_G
(i added then because of a problem i thought was existing with slow disk
access but the problem was simply because of an old harddrive :/)

changed sched_4bsd to sched_ule

now then, the sound problems above seem to no-longer exist. at all.
after i finished compiling and installing the new kernel i had to go to
work. 9 hours later (well more like 11 actually after the walks to and
from) i am listening to some mp3s through my headphones to detect the
slightest tick or pop and i have yet to hear anything. if i had run the
system for that long before it would be popping and clicking like
popcorn right now. so i think i managed to get rid of this problem. i
really hope this helps the other people with the same problem. i've only
been listening to music for about 10 minutes but that should be more
than enough time for the problem to arise. if it happens later i'll
repost i guess but for now i hope its solved. thanks to everyone who
replied :)

also i am not sure which of the options actually fixed the problem...i'm
leaning towards the disable_pse and disable_pg_g options but i don't
know? thanks,

- ryan