[SOLVED] EVGA GTX 1080Graphics Card Starting to Die?

[CorMazz]

Hello everyone,

Windows 10 Home 64bit
EVGA GTX 1080
i5-8600k 4.5 GHz
AS Rock Z370 Killer SLI/AC Motherboard
Thermaltake Toughpower Grand 650W PSU
16 GEIL DDR4 @ 2,132 MHz


I have a GTX1080 that I purchased new in January of 2018. It is factory overclocked and ran fine up until about a month ago. Now, whenever I play Rainbow Six Siege, the game I have played fine for going on two years, the graphics glitch out and then the game crashes. Just the game. The rest of the pc runs fine. I "reset" windows to see if it was a background program causing this and it did not solve the problem. I lowered the graphics settings in the game and that lowered the incidence of the crashes. I have dusted my pc. I downloaded the 3D Mark Timespy benchmark in an attempt to stress my GPU. The Timespy benchmark crashes after about a minute. GPU temps are always steady under load at around 68 degrees Celsius.

I ran Prime95 to see if my CPU was failing somehow, and the stress test went fine, with temperatures about 86 degrees Celsius. Then yesterday I messed with the inside of my PC again and somehow the Prime95 temperatures went up to 100 celsius and it started thermal throttling, but it still did not crash. It just throttled itself. When in game the CPU temperatures stay around 75 degrees Celsius. I did kind of skimp on a PC cooler when I made this rig.

Any advice for what I could do to figure out the source of the problem? I am not even sure if it is 100% a GPU problem. I don't know how to diagnose the issue, let alone go about solving it.

Thank you for the help,
CorMazz

 

[Phaaze88]

It looks like you experience crashes when you push the gpu hard enough.

Thermaltake Toughpower Grand 650W

That would be the first thing I'd suspect. Thermaltake's rep in the psu market isn't great.
Do you have a spare gpu or psu to test with?

As for the ram, that info is printed on stickers on the back of every module. Open your case up and check that, or you can use Cpu-Z.

 

[CorMazz]

It looks like you experience crashes when you push the gpu hard enough.

That would be the first thing I'd suspect. Thermaltake's rep in the psu market isn't great.
Do you have a spare gpu or psu to test with?

As for the ram, that info is printed on stickers on the back of every module. Open your case up and check that, or you can use Cpu-Z.

I figured out what RAM it was, 3DMark told me. I also added my mobo in there because I had forgotten it.
I do not have a spare GPU or PSU to test with. Is there anyway I would be able to see what the power coming off the PSU is with software and analyze it that way?

Thank you for the help.

 

[Phaaze88]

You can find out how much power the gpu is pulling with HWINFO:

Free Download HWiNFO Sofware | Installer & Portable for Windows, DOS

Start to analyze your hardware right now! HWiNFO has available as an Installer and Portable version for Windows (32/64-bit) and Portable version for DOS.

www.hwinfo.com