System rebooting during heavy cpu tasks

Question

I have a K7N2 (MS-6570) board with 1GB ram using 2 mushkin PC2700 ram sticks running win2k with the latest bios and nvidia drivers.
When I try to perform intensive CPU tasks such as running the united devices distributed program or the distributed.net applications, render images or even build large source code modules the system reboots with no warning.
Occasionally I'll play a game and in the middle of it the system hangs and then claims I've executed invalid instructions and windows kernal gets corrupted and then reboots.
During rendering I get random STOP 0x errors, never the same one twice.
The timings are at 166fsb, memory at 1:1 (333mhz) clock at 12x5 CPU speed at 2088 mhz.
If I use auto-defaults memory timings goes to 7-3-3-2.5 but the mushkin website claims the ram should be at 2-2-2 so I set it that way and still I get wierd errors.
This is the 6th memory pair I've tried so far, could I possibly have a bum board or processor instead?
Is there a "safe" memory timing I can try to see if its the ram or my system itself thats bad.  I'm close to giving up on this board and going the intel route.

Answer 1

Whew.
I'm glad you had the spare supply; I usually recommend "trying before buying" anything, unless, of course, one has lots of cash to blow.
I also wanted to tell you that, if your 2.03 driver method doesn't work, try this:
Go to Start, Run, type in MSCONFIG, press Enter, press the Services tab, check the Hide All Microsoft Services box, press the Disable All button, select the Startup tab, press the Disable All button, Apply, OK, and Restart.
After the reboot, try your games again, and see what happens. If everything's ok, then go back and enable the startup services one at a time, and try playing again until you find the offending service. If all is ok, do the same with the Services tab, until you find the offending Service.
(I spent hours online with Microsoft on this a couple of months ago, and this method found that the intermittent crashing problem was caused by one of the nVidia services.)
So far, I have had four causes of this type of problem (BSOD, spontaneous reboot, freezing):
1) Corrupted OS/driver
2) Poor power
3) Bad RAM
4) Startup Service
Of course, #1 and #4 don't cost anything but time and effort to fix, and should be tried first, unless anything is grossly wrong, like an underrated supply, and/or ram which won't pass memtest, which one should replace anyway. (I've got a lifetime warranty on the RAM, and only had to return it for replacement.) So the only thing that cost me to repair this problem completely was the replacement of the PSU, and it was the very last thing I did.
Were I to do it over again, I would have replaced the PSU before I actually put the replacement RAM in, because I risked damaging the new RAM (and the old stick too) with my old 300W supply. (However, I was on a severe cost restriction which influenced my decision, making the "free fixes" the priority.)
Hope this helps.