Blackscreen crashing during gaming - no BSOD, GPU/PSU fans going wild

BPCadet

Honorable
Sep 24, 2012
20
0
10,510
Hi,

Recently I've encountered an issue where my PC has been randomly crashing while I've been gaming. To date, it's only happened with Overwatch. When it crashes, the sound keeps going, the screens go black and say there's no input, and the fans on the GPU and PSU go nuts.

I've had my PC for several years, and have never had any problems with it to date. I'm not sure why this would be starting now, other than the warm weather. I think the problem may be my GPU overheating or my PSU not drawing enough power, but I'm not sure how to test either, and when I try to use furmark, the temperature shoots up to 97°C within a minute and I start worrying about shutting the whole thing down, because that's scary hot and I don't want to burn out my card (GTX 670). I'd tell you the wattage of my PSU, but I honestly don't know, and don't quite know how to get at it.

It's not the CPU or the RAM - that would have come up when I ran Prime95 all night.

Can anyone give me some pointers?
 
Hi,

I think you did a great diagnostics. (bloack screen while full load is problem with CPU / GPU PSU most of the time)

The best thing to do , download hwmonitor and monitor your temps while playing overwatch.
The computer should have problems if CPU reaches 80 degree or more while GPU should run ok unless it goes to 90 degree or more.

I try to use furmark, the temperature shoots up to 97°C within a minute and I start worrying
What hit's 97 degree ? CPU or GPU ?

For the power supply, you need to test it out with a known working PSU.

If the cpu is overheating, you need to reapply thermal paste or get an aftermarket cooler.
For the gpu, reapplying themal paste might help but sometime they also have thermal pads.

Good luck
 

BPCadet

Honorable
Sep 24, 2012
20
0
10,510


The CPU is not overheating. I hit 95°C when running at 100% capacity for hours, but that's still entirely acceptable, and the PC keeps on trucking along just fine despite that. It's nowhere near that while gaming - I have an i7 Ivy Bridge, and Overwatch is not that taxing on the CPU, from what I understand.

The GPU hits 97°C within a minute after starting FurMark - at least, according to furmark. And that makes me worry that it's gonna blow out my GPU, and I shut the whole thing down, because if I have to buy a new GPU, I am completely screwed.

Which, BTW, is my problem with your suggestion to test the PSU against a known working one. I don't have a known working one. I can see if I can bum one off my workplace (unlikely) but I don't have a lot of money, and if the only test is "get a new PSU and test that", then I'm in a bit of trouble. :(

Also, I never had any thermal paste on my GPU. The fans on it are built-in.

I'll try running the diagnostic kit while playing Overwatch, and see if I can figure out what's going on. Thanks in any case for the advice! :)

A big part of the problem is that I don't really get how to evaluate these results. :(