Greetings. My hardware consists of the following: AMD P2 x4 955 BE ASUS M5A99X EVO MSI GeForce GTX-560Ti (2) 1Tb Seagate Baracuda 7200.12 1Tb WD Caviar Black Corsair TX-850V2 Now lets start a log. From the start I got the cpu, its stayed at 3.7Ghz with 1.4v overclock. In between I had the north bridge at 2600mhz last year, but it was causing blue-screens, because of which I turned it off, and ultimately the culprit was the faulty hard drive. Now nearly a month ago, my pc froze randomly after usage. This was when I had my northbridge at 2600mhz exactly 1 day prior to the hanging. And then finally when I was installing nvidia 314.22 there was a bsod just after it told me to restart. Unfortunately it just restarted n I couldn't see the code. After that I dropped the ball and dialed it back to 2000mhz north bridge. Running stable, no problems. Now around 1 week back, I dialed back to 1.35v, and 3.2Ghz because of the weather. SLI was still there. Just yesterday, I sent a card in RMA(TwinFrozr II Fans so OP), so now I'm stuck with 1 card. I just get these random freezes/hang ups again. No BSOD. It just hangs. Only 1ce I got an error of 'nvidia driver 326.19 stopped working' (updated recently), and the other 1 time I recovered from the freeze, my usb keyboard just got messed up. I had to disconnect it for some time and then plug it back in. Any idea whats causing these issues? My bets are slightly on my 850W PSU, but then Corsair psu's dont break often, and it has fairly served me right till now. Plus both gpus running at 99% load for hours it sustained. Hell it even rode through crysis 3. Just cant figure out whats exactly wrong. Also, I hard disk smart stats check out. 0 errors there. GPU also sorta checks out. Only culprits possible are motherboard,cpu,ram,psu. Any suggestions or help will be appreciated.
The last time I ran the test, it took 8hours for atleast 6 passes. How many passes do you advice I do?
If your memory can get through 2 passes, in most cases the memory is fine. However, for the sake of ensuring the most accurate results, it's best to allow it to run for longer. I'd say 8hrs should be sufficient though. If you don't see any errors at that point, the memory and controller are fine. Keep in mind though, errors in memtest don't necessarily mean that the memory itself is faulty. A faulty memory controller will also cause errors in memtest.
One time I remember seeing the same issue is on my uncle's pc, where the pc just hanged in the middle of nothing. Apparently dust was prohibiting the ram from functioning fine. But the difference is, the pc wasn't booting again. In my case the pc is booting just fine. Also you do understand, checking each module individually for 8hrs takes a hefty amount of time. It will probably take 4 days to check the 4 modules while I'm at work. Will report my findings.