On Wednesday 20 June 2001 06:00, you wrote:
> On Tue, Jun 19, 2001 at 09:57:50PM -0500, you [Robert Redelmeier] claimed:
> So K7's do actually fry themselves so that they won't ever recover? I was
> under suspicion that most CPU's will halt before they actually totally fry
> themselves - this one booted and ran fine until suddenly died completely.
When I had lockup and oops problems with a KT133 chipset and PC133 RAM, the
CPU seemed to run rather hot, and the rack mount case airflow was not
optimal, so I was suspicious of thermal causes.
The dealer told me after checking, that AMD said, that a K7 automatically
throttles back, if it's overheating, and that it could not be a cause of the
problems I saw. I don't believe that, certainly the thermal protection is
less than the total cutout with Intel CPUs, and I haven't 'confirmation' of
throttling on any technical sites, though I later elminated thermal effects
in that troubleshoot.
Certainly if you run a K7 without heat sink, it fries very quickly.
> Yes, but he only booted into Win98 shortly to install the sound card
> driver. He had ran DivX decode, opengl stuff and stuff like that in Linux.
Which is a minimum of 5 minutes busy idling. If the thermal protection is
going, it's the last straw that breaks the camel's back.
If your friend tried reseating everything, CPU, RAM as well as PCI cards,
Roberts explanation is a plausible one. He should be glad the CPU was still
under warranty.
Rob
--
=- To unsubscribe, email [EMAIL PROTECTED] with the -=
=- body of "unsubscribe linux-abit". -=
