I've looked around every corner of the internet and seen that many people have had this problem. Unfortunately though I can never find threads with solutions that work. I apologise if this has been asked/answered before but I couldn't find anything so far that's worked for me. The Problem:
My PC will crash to a black screen (like someone has pulled the video cable out of the monitor) and whatever sound was playing at the time will loop for about 30 seconds, then the speakers go dead. It seems to occur solely during gaming (although not at particular points within games, nor does it seem to depend on how demanding the game is in terms of processing power, gfx or RAM). Due to the apparent randomness of it, it could
be that it is not limited to gaming, although I have never seen it happen at any other time so this seems unlikely.
The PC does not reset, it idles in "power on" requiring a manual hard reset.
No dump files are written (either kernel or minidump) - the PC doesn't seem to know it's crashed. Event log logs my manual reset but nothing else "Critical".
This problem does seem pretty common, as I say Google pulls up a fair few threads where people complain of this. Some are resolved but as yet nothing that worked for them appears to be working for me. PC Specs:
AMD Athlon II X4 645 - 3.10 GHz
4 GB RAM
NVIDIA GeForce GT 440
ASUS M4A87TD / USB3 MoBo What I've tried:
- Monitored CPU, Core, GPU, MoBo temps. CPU peaks at around 60 C (which I think is safe for my CPU model). GPU peaks ~70 C. MoBo doesnt get much higher than 50 C. The crashes don't seem to be dependant on temps, i.e. the crashes occur at lower temperatures than the max recorded. Eg a crash before occurred with CPU and GPU temperatures ~5 degrees lower than their peaks within the log (which were 75 and 59).
- Monitored voltage on 3.3, 5 and 12v lines. The values do creep over their "maximum", the 3.3 is frequently at 3.36, the 5 will reach 5.05ish, and the 12 will reach 12.05. Again these variations and maximums don't coincide with crashes (at least apparently not)
- Ran Memtest86 several times - no errors
- Ran FurMark on various different settings - no crashes or problems
- Uninstalled, sweeped and reinstalled the nVidia driver (I did think that this might be the issue, maybe not though?)
EDIT: I've also upgraded chipset and bios to their most recent iterations
None of that seems to have changed anything, crashes still occur at random times. They have been going on for quite a while now (at least 6-7 months) but can go months without one, then have 3 in a day with no change in PC usage over that time.
As I say I have not got any dumps to upload, unfortunately. I've also dl'd the SF diagnostic tool but unfortunately it's misbehaving. All tests other than system info give an "Unhandled exception error" citing "Could not find file C:\*.txt". The filename changes depending on which test I try to run. I have a feeling this is probably a stupid error on my part
If anyone has any
idea of anything more I can try, or what the problem might be I would be eternally grateful.