At Mon, 19 Jan 2004 22:54:23 -0800,
Noah Roberts wrote:
>
> Renai wrote:
>
> >Hi all,
> >
> >I had no expectations of ever getting the on-board sound to work under linux
> >when I first bought my VIA motherboard with 8233A sound card. The only scenario
> >where it works for me is with a 2.4.22 kernel and ALSA 0.9.0rc6. I was quite
> >surprised when it did.
> >
> >
> I guess I have become spoiled. See, I have been using Linux now for
> several years. In the beginning I had a piece of crap sound card that
> barely managed to work. It was labeled as unsupported and I could only
> get 8 bit sound out of it after using isapnptools and claiming it as an
> SB. I was happy to get it to work at all because it was one of those
> "If it works for you count yourself very lucky" kind of things.
>
> Later, I bought a real SB16 and used it for a few years until ISA was
> unavailable (sad too, it was one of the ones with the built in amp).
>
> So I had to buy a new soundcard. I bought a piece of crap for $15 and
> ran lspci to find that it was a maestro, which is listed as supported
> with OSS and worked flawlessly. Once I saw that it was supported I had
> no doubt that it would work.
>
> Now I have this via8233/A, which is listed as working...but doesn't. I
> am dissapointed because it is one of those "Works, use driver XXX and
> run XXX" kind of things. Once I saw that it was supported I had no
> doubt that it would work.... oops...
>
> See, when something has a version number of 1, I have a reasonable
> expectation that it works. When it is incorperated into the kernel and
> is threatening to become the new thing, I expect it to work as
> advertized. When my soundcard is listed as supported, I really expect
> that it works. I have never, in all these years, run into a situation
> like this - especially with regard to the kernel itself. It really
> surprises me that ALSA was accepted into the kernel.
such an argument is nonsense.
do you expect that 2.6.0 kernel works flawlessly with EVERY SINGLE
driver since it's released as the "stable" tree? and, you believe
Linus shouldn't release 2.6.0 until all modules get fixed?
so it is for ALSA. we released 1.0.x, but not every single driver can
work well yet, unfortunately. the total system is becoming stable, i
believe, though.
i know that via driver is often problematic. it's just because it's
the onboard chip, and it's a combination among quite different
AC97 chips. you cannot test/debut it unless you have the
*EXACTLY SAME* motherboad, CPU, and the EXACTLY SAME kernel, compiler,
linker, etc. don't expect me to buy hundreds VIA mobos on the market
only for testing and debugging this driver module :)
ok, rant gives no help. let's debug it.
seriously, i'm not sure whether the VIA823x revision 0x60 on your mobo
works. it must be pretty new. so far, it's not recognized as a
workable revision (i have no datasheet for that) although i guess
it's compatible with the older revisions.
(and, your chip is *not* VIA8233A. VIA8233A is the older revision,
which behaves fairly differently from other revisions. don't be
fooled by the output of lspci :)
for debugging, please provide the following info:
1. what happens if you play WAV file?
i know no sounds come out, but the important thing is whether
it looks as if it's working, or it hangs up, or got any errors,
etc.
run
% aplay -vv foo.wav
and it will show the level meter. check whether it works fluently.
check the kernel messages. also, check /proc/interrupts whether the
number of interrupts increases during the playback.
the most of problems on via driver is in this part.
in 99% case, they get sounds but distorted. in such a case,
tuning the dxs_support module option can help. see the
ALSA-Configuration.txt for details.
2. if the playback looks ok, then the problem is likely in the AC97
codec part.
in many cases, some mixer elements are not initialized correctly.
the names of some mixer elements are not intuitive.
if you're sure that it's not the configuration problem, please give
us the contents of /proc/asound/card0/codec97#0/* files, and
/etc/asound.state (after running "sndctl store"). this will help
the debugging of ac97 codec part.
ciao,
Takashi
-------------------------------------------------------
The SF.Net email is sponsored by EclipseCon 2004
Premiere Conference on Open Tools Development and Integration
See the breadth of Eclipse activity. February 3-5 in Anaheim, CA.
http://www.eclipsecon.org/osdn
_______________________________________________
Alsa-user mailing list
[EMAIL PROTECTED]
https://lists.sourceforge.net/lists/listinfo/alsa-user
