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

Re: ALI chipset

>In message <Pine_(_dot_)_BSO_(_dot_)_4_(_dot_)_21_(_dot_)_9911230927460_(_dot_)_9509-100000_(_at_)_dqc_(_dot_)_org>
>	so spake Chris Cappuccio (chris):
>> I have rev 4 of this chip on a moderately utilized server that does DMA with
>> IDE, on both controllers, as well as Intel fxp, etc... I've never noticed a
>> fault with it so far (after many months of operation.)
>This is an all-scsi system (tried with several different scsi controllers).
>> This may be a settings problem?  The K6 was announced by AMD as having a
>> voltage of 2.1v for the 266MHz and higher models When it was actually
>> released, it was a 2.2v chip.  Some motherboards that do auto-voltage
>> still default to 2.1v.  Or, the motherboard documentation may prompt you
>> to set 2.1v.  This will cause problems of course....
>No, the setting are correct.  I'm not the only one who has seen this
> - todd

K6-2-300 series running at 100Mhz on VIA chipset boards are known to have
cache timing problems. We have heaps of the afflicted processors here at
work. It really shows up under windows and netware regulary. The only work
around i've found for that is to radically reduce the bus speed and change
the multiplier. So far I haven't seen the same problems with the K6-2-450's
or the K6-III-450's running full tilt.

On the other hand I currently run at home a K6-2-450 on an ali chipset
motherboard with OpenBSD 2.5 and haven't had any problems at all.

Catch ya,

Ian McWilliam.
Technical Officer,
School of Computing & IT
Uni of Western Sydney, NEPEAN.
P.O.Box 10 Kingswood,
N.S.W, Australia 2747.


Telephone: 61 247360 757
Fax:       61 247360 770