Question Is my GPU slowly dying? - Applications crash with high power draw.

[bigfootheyy]

I'll quickly get to the point, thanks for reading.

The games I want to play (BFV, Witcher 3) keep crashing, when I don't limit the power target in ASUS GPU-Tweak II to around 93-95%. When I limit it this way, I can play, but get artifacts from time to time. After crashes in games I have seen GPU-power of above 100% (102-108%) in HWMonitor every time.
Something similar happens in stress tests. If I set the power target to 120% and do a FurMark stress test, the power slowly creeps up to around 115% before the program shuts down.

BIOS is on the newest version, set to default. The newest NVidia drivers are cleanly (ddu) installed, Windows install is new and no other metrics are altered in GPU-Tweak II. I have tested my processor with intel’s diagnostics tool, my ram with memtest and got 0 errors.

What is wrong here? Any help would be appreciated enormously.


System-Specs:

Asus Z270F Gaming
Asus Strix 1080 8GB OC @ 11GBps (under load @ 68 celsius)
16GB DDR4-RAM @2400MHz (stock, XMP off)
i7 7700k with Noctua fan (under load @ 81 celsius)
Windows install on SSD drive - ADATA XPG Gammix S11 Pro 1 TB
2 seagate HHD's (1TB)
BeQuiet SystemPower 9 500W PSU
Monitor: Samsung CF971 3440x1440 @100Hz
Case: Fractal Define R6

 

[Flayed]

It does sound like the card is failing

 

[bigfootheyy]

It does sound like the card is failing

Thanks for answering. Could you elaborate on that, or is it a gut feeling? If it is the latter, could you give me a ballpark of your experience in order to assess what that hunch should mean to me?
Again, thanks for answering.

 

[Flayed]

I had a similar issue with an Asus Strix Vega 56. It would crash very predictable in games. When I lowered the power in MSI Afterburner it didn't crash. It would also cause the system to restart.

I posted on a different forum about and people convinced me it was my PSU so I bought a new one. It didn't fix it. Then I found out the GPU VRM was running hot so I removed its heatsink and replaced all the thermal pads and repasted with grizzly paste. This fixed the issue for a while a couple of weeks. Then the card suddenly died during gaming.

 

[darkzim]

might be loaded with dust try and see if it has a lot of dust in the cooler

 

[bigfootheyy]

might be loaded with dust try and see if it has a lot of dust in the cooler

Thanks for answering. I had it in my hands while changing my system SSD a while ago, and cleaned it: No excess dust on there.

 

[bigfootheyy]

I had a similar issue with an Asus Strix Vega 56. It would crash very predictable in games. When I lowered the power in MSI Afterburner it didn't crash. It would also cause the system to restart.

I posted on a different forum about and people convinced me it was my PSU so I bought a new one. It didn't fix it. Then I found out the GPU VRM was running hot so I removed its heatsink and replaced all the thermal pads and repasted with grizzly paste. This fixed the issue for a while a couple of weeks. Then the card suddenly died during gaming.

Thanks for the tip, I'm going to have an eye on the VRM temps.

[EDIT] Actually, my Motherboard doesn't seem to have a sensor for VRM-Temps and I don't have thermometers lying around, that are capable of measuring that kind of stuff. How'd you find out about your VRM running too hot?

[EDIT 2] Found the Readout in HWiNFO. While artifacting and crashing in FurMark the VRM-Temps on my GPU never got above 80°. This does not seem to be my problem. Thank you anyway.