Re: USB and VIA chipset

kony wrote:

> On Sun, 22 Jun 2003 23:30:27 -0400, Stacey <fotocord@yahoo.com> wrote:
>
>>kony wrote:
>>
>>
>>>
>>>
>>> For the last time, you are just plain wrong.
>>>
>>
>>So the PCI latency patch was written for what reason? People were bored
>>and made up these 57,000 pages? OK I believe you because you have a system
>>that doesn't have a problem with the hardware mix you lucked into..
>>
>>http://www.google.com/search?hl=en&...G=Google+Search
>
> Again you jump to conclusions... I've had no less than 4 distinctly
> different video cards, 2 video capture cards (with different chipsets)
> on most Via chipset boards, 3 video capture cards tried on Via boards
> with KT266 and after, and at least 3 sound cards... These are only
> components used LONG-term, that I remember distinctly, but the odds
> are that even more components were swapped around, it's just been a
> few years since Via chipsets for Athlons debuted so cards that were
> swapped in just long enough for DOA-type testing are soon enough
> forgotten after they were confirmed working.
>
> Let's directly address what you refer to with the link, a latency
> patch. Do you understand EXACTLY what that patch does? It changes a
> chipset register. That is a register that was incorrectly set by the
> BIOS. I repeat, a register that was incorrectly set by the BIOS.


http://www.georgebreese.com/net/sof...v014_readme.htm

VIA has acknowledged a problem with their VT82C686B controller chip. The
VT82C686B is a chip that is responsible for, among other things,
controlling the IDE bus. When overloaded by an incorrectly-configured
motherboard, the chip can cause corruption and/or lockups. The SoundBlaster
LIVE! card seems to make the problem worse.


Since you claim this is only a bios issue, wouldn't you think the board
makers would just fix the bios (as companies like asus/gigabyte release
several different bios's through most board's lives to fix bugs) instead of
pointing their customers to 3rd party patches? Why does it seem only a few
of these boards are immune to this problem? And if it's only a bios
problem, why do so many boards using different bios's exhibit the same
symtoms? Why so many different 4 in 1 drivers? Other chipsets don't seem to
need constant revisions of the chipset drivers to work properly or special
video drivers to work with Via chipsets etc..

>
> What do you want to wager that I can take the most stable board you
> could name, to have EVER existed, and cause a major problem by
> changing a BIOS chipset register setting or two? The point is that
> the chipset isn't the problem, a buggy BIOS is.

Interesting Via acknowledged this problem though isn't it? And that so many
board have the exact same problems using totally different configs and
bios's? Glad you feel these chipsets are so great, I'm not interested in
saving $5 to screw with junk like this...

--

Stacey

On Mon, 23 Jun 2003 22:49:10 -0400, Stacey <fotocord@yahoo.com> wrote:


>http://www.georgebreese.com/net/sof...v014_readme.htm
>
>VIA has acknowledged a problem with their VT82C686B controller chip. The
>VT82C686B is a chip that is responsible for, among other things,
>controlling the IDE bus. When overloaded by an incorrectly-configured
>motherboard, the chip can cause corruption and/or lockups. The SoundBlaster
>LIVE! card seems to make the problem worse.

Note the part that reads "When overloaded by an incorrectly-configured
motherboard". Since when has it been guaranteed that an
incorrectly-configured motherboard will run ok?


>Since you claim this is only a bios issue, wouldn't you think the board
>makers would just fix the bios (as companies like asus/gigabyte release
>several different bios's through most board's lives to fix bugs) instead of
>pointing their customers to 3rd party patches?

Actually some did, and others didn't.. .i suspect it was for political
reasons, easier to pass the blame... Attention was already focusing
on Via because no one had enough information (at that time) to know
what the real problem was. At any rate, the boards had already sold
and the 686 Southbridge is no longer used, yet every tech-geek-wannabe
was recanting the little bit of speculation mixed with fact that was
floating around the 'net. It is clear that Via could've done a better
job to protect user's data from "themselves" or 3rd party hardware
that didn't function properly, but pointing fingers at Via doesn't
change the cause.



>Why does it seem only a few
>of these boards are immune to this problem?

Because you jumped to conclusions?
Most of the boards DON'T have the problem, and the few that do, have
it only in very specific circumstances. It's a situation of hearing
people complain when things go wrong, but remain the silent majority
when things are working right.

>And if it's only a bios
>problem, why do so many boards using different bios's exhibit the same
>symtoms?

The BIOS is modular, more similar than different in many cases. The
fact remains that the BIOS sets the chipset registers, and the latency
patch, in the cases where it is needed, wouldn't be needed if the BIOS
set that register correctly in the first place. No matter what else
we'd like to disagree on, that is a clear fact, the BIOS sets it, and
set it incorrectly, or rather, it was set to be less tolerant of
devices extensively using the PCI bus.


>Why so many different 4 in 1 drivers? Other chipsets don't seem to
>need constant revisions of the chipset drivers to work properly or special
>video drivers to work with Via chipsets etc..

Some companies release new drives with each little change, and others
release them as a cumulative set of changes... this really has nothing
to do with the Athlon/Via/etc problem, the 4in1 drivers were released
in this manner before the 686b SB era, and after.


>Interesting Via acknowledged this problem though isn't it? And that so many
>board have the exact same problems using totally different configs and
>bios's? Glad you feel these chipsets are so great, I'm not interested in
>saving $5 to screw with junk like this...

They acknowldeged that there is a potential problem, that can exist in
specific circumstances, which is different from acknowledging that "we
are to blame for this". If the problem were as extensive or common as
you imply, the boards would've all been recalled and there'd be a huge
lawsuit.

Actually the problem isn't so widespread and common as you'd like to
think... Many, MANY people still use these boards every day with no
problems. These are the 686 SB boards I'm talking about, while you
are still suggesting that Via boards are a problem even when they
haven't used the 686 SB since the KT133A, possibly a few KT266 boards.

If you feel that Via 686B Southbridge chipset boards for Athlons are a
problem, then it would seem reasonable to consider the other hardware
before using one, but to suggest that the more modern Via chipset
boards are still affected in any similar way is deceiving, isn't
supported by the evidence.


Dave