Which MBs are you using for both the K6-2 450 and the PR200? Actually,
the PR200 does sound somewhat familiar... that was the kind of behavior
I had from the PR200+ and MB that it had prior to it biting the big one.
You might try (and this is only a thought based on a hint from what
happened to the PR200+ after it blew... it still technically works,
but only specs out as a PR90, regardless of what combinations I use
on FSB and Multiplier) you might want to try reducing the core
voltage to the PR200. Drop it from 2.5 to 2.2, and see if you
get an increase in stability (I know, it's counterintuitive to some
folks, but you shouldn't do anything to damage it by the drop).
You may be having some of Cyrix's famous thermal problems. Also,
try breaking down the system, and cleaning out the dust bunnies. They
may be causing some deterioration of the contacts between some of the
components, which could be causing problems... this would be a slow
slide from operational to unoperational. I've got a P100 here that
I brought back "from the dead" (it was about to meet Mr. Garbage Can,
Mr. Garbage Truck, Mr. Trash Compactor, and Mr. Landfill) simply by
disassembling the system and rebuilding it after cleaning.
I have doubts that that is the problem with the AMD, though, unless
you're in a particularly dirty environment.
Bill Ward
-----Original Message-----
From: Charles Galpin [mailto:[EMAIL PROTECTED]]
Sent: Wednesday, December 29, 1999 2:10 PM
To: '[EMAIL PROTECTED]'
Cc: recipient.list.not.shown; @nswcphdn.navy.mil
Subject: RE: hardware testing
well, so far my athlon has been fine. I have been running a kernel compile
in a while true loop for over 2 days straight now.
On Wed, 29 Dec 1999, Ward William E PHDN wrote:
>
> As for how to troubleshoot... if you have access to an additional
> Super Socket 7 Machine, I would first swap the memory between the
> two machines. Test that for a few days, then swap the CPUs. You
> should slowly be able to isolate which part is the problem by doing that.
Unfortuanately I don't. I'm going to talk to the place i bought the
components from later today.
I have tried running it with one memory chip at a time. With the first
chip it lasted about 11 hours, then barely an hour the second run. I'm
tryin gthe second dimm now. If this one behaves the same way, I don't
think it's the ram. based on what you are saying, the MB would be the next
most likely suspect.
> There are also some old DOS utilities (and a few that run under Win95
> et al) that can do integrity testing of memory, and can SOMETIMES
> do testing of the CPU. Try checking some of the download sites
> like Download.com, etc.
yup, looked at a couple already that work off a boot floppy. not impressed
by any yet. If you can recommend any I'd appreciaste it.
> You may need to install DOS or Doze on the
> system to run them, though, but if you're having this much problem,
> you're going to rebuild the system image when you are done, anyway.
certainly willing to do this. One reason would be to allow the place I
bought the parts from muck with it.
One more question. I have a cyrix 200 pc that I used to run NT on. It
worked reasonably well, with the occasional bsod. I made no hardware
changes, and installed win98 on it. It reboots itslef randomly every
couple of hours, even when not in use. I ran a memory test on it a while
back and it seemed to check out fine. Anything sound familiar with this
one? :)
>
> Bill Ward
--
To unsubscribe: mail [EMAIL PROTECTED] with "unsubscribe"
as the Subject.
