Re: [linux-audio-dev] Soundcards: USB, Firewire, PCI, PCMCIA

New Message Reply About this list Date view Thread view Subject view Author view Other groups

Subject: Re: [linux-audio-dev] Soundcards: USB, Firewire, PCI, PCMCIA
From: Ariel Salomon (audio_AT_oscillatory.org)
Date: Tue Jan 22 2002 - 20:02:54 EET


Richard C. Burnett wrote:

> I also wanted to point out for those who are not aware, be very weary of
> the VIA chipset on many AMD motherboards. Apparently there are some
> communication issues between the north and south bridge chips and it
> causes quite a bit of latency. I have the IWILL KK266 motherboard and
> experienced these problems. I found an unofficial patch that seems
to fix
> the problem, but just wanted others to be aware, because it is REALLY
> fustrating!
>
> Rick

There have definitely been a few issues with those Via KT133/266
chipsets. Seems to be due to some control bits Via did not originally
document, or at least not well enough. But it looks like Iwill updated
their BIOS to deal with what seems to be this problem a few months ago
(I've got the same board, more info below).

A probably related issue was that under Linux, Athlon optimizations in
the 2.4 kernels exposed problems booting on motherboards with these Via
chipsets (fer instance the one that RedHat 7.2 tries to boot after the
initial install, if it detects an Athlon). But .. there was a patch a
few kernels back (2.4.10?) that fixed some cases, and a proper patch
based on info from Via that went into 2.4.18-pre1 (also included in
Jussi Laako's 2.4.17-jl11 low latency patches, which is was I run).

It seems that with a current BIOS and current drivers (or Linux kernel)
that these problems are finally behind us .. though certainly a blemish
on Via's record.

     - Ariel

in particular this BIOS update looks impt: (see support.iwill.net)
"KK266 for Product No is 35103A, BIOS Date Code: Apr.26.2001
1.Fixed system performance lost.
2.Fixed SoundBlaster LIVE! card are distorted
3.Fixed files copied from one IDE hard disk to another are corrupted.
4.Fixed files are corrupted when copying from to/from DVD, CDROM or CDRW
device.
5.Correct CPUID=0643H ratio table.
6.Fixed data corruption with 686B."

-- 
Ariel Salomon (asalomon_AT_mit.edu, ariel_AT_oscillatory.org)
RLE Speech Communication Group, MIT, 36-545
O: (617) 452-3598   C: (617) 803-8099


New Message Reply About this list Date view Thread view Subject view Author view Other groups

This archive was generated by hypermail 2b28 : Tue Jan 22 2002 - 19:55:17 EET