I just built a new machine and it worked great for about two days and now it freezes up completely about every two minutes. I've seen it lock up even before XP loads and in the bios. The CPU temperature seems ok even running prime95 it doesn't get over 51C. I ran memtest86+ and everything checked out. I also measured the voltage of the power supply from the plug supplying the cpu and it had the correct voltage. I bought the parts from tigerdirect and I'd like to be sure before I start sending things back. Thanks for any help
ZEROtherm Butterfly CPU Cooler BTF90 778/754/939/A
Ultra LSP750 750w Power Supply
OCZ SLI 4096MB PC6400 DDR2 800MHz (2x2048MB)
EVGA GeForce 8800 GTS 512MB PCIe w/Dual Link DVI
MSI K9N2 SLI Platinum nForce 750a Socket AM2+ MB
AMD Phenom Quad X4 9850 2.5Ghz Black Edition CPU
I remember a long time ago when I built a new AMD ATHLON 700MHz Slot A system I had the same problem.
What I did to fix it was, boot the computer up to a DOS prompt and let it sit like that for 24 hours. This is known as a Burn in test. After the burn in test, I installed Windows 98 and everything worked fine with it for 8 years.
If a computer locks up during the burn in test, send it back starting with the motherboard, CPU, and then the memory.
Well I ran both sticks of ram by themselves and still had freezing problems. I also checked my CPU temp with coretemp and it only reached 35C before it locked up. I also tried reseating my fan, which also got me nothing. I also ran linux mint off a live cd just to be absolutely sure it was a hardware issue and the computer still froze up. Is there anything else anyone can think to try. I'm tempted to pack up the PSU, MB and the CPU and send them back. I don't really know what else to do. Thanks again for the help
At this point, when you don't have parts to swap out, it's a pain.
I don't want to rag on MSI, there are far worse brands, but they are not the best either, in terms of percentage of dead boards shipped. I favor the board as the bad part because it worked great for two days... It's more likely to be a part on the MB that failed.