GPU Crashes when Playing Games

[KyleAparthos]

Hello,

So I've had this GPU (a Gigabyte model NVIDIA GTX 760) since about May of this year - previously I had to RMA it because the first iteration they sent me in March of this year was dead on arrival.

It worked perfectly for about 2.5 months, until the end of July, at which point in time I began experiencing the occasional system crash when the GPU was under high load. The screen would cut out, along with all input from the keyboard/mouse/power button/etc. and the only way to reboot the computer would be to cut all power from the PSU.

At first the GPU would be at a slightly higher idling temperature (the result of a higher idling clock speed) when it was about to suffer this issue (about 7-9 C higher) and a simple driver reinstall would fix the problem for a while, though it would inevitably come back.

Unfortunately at this point the driver reinstalling strategy is no longer effective, though the higher idling temperatures have also gone away. I have also noticed that a simple stress test does not appear to crash the computer - it only occurs when playing a video game with high rendering requirements, such as EVE Online, Dota 2, Skyrim, etc.

As a gamer though, this deeply concerns me due to one of my favorite pastimes being gaming, and with the crashes happening every few days now, I am strongly concerned that the repeat crashes will affect my HDD and system eventually. I am uncertain as to if this is a software or hardware issue, and am hoping that a software fix or simple problem that I have overlooked will be the solution.

My system specs are as follows:

Motherboard: Gigabyte GA-990FXA-UD3
GPU: Gigabyte GTX 760 2 GB model
CPU: AMD FX-6300 6-core 3.5GHz
RAM: 8 GB Corsair 1600MHz
PSU: Corsair 750 W (A slightly old model, approx. 5 years old, but all voltages appear to be functioning normally in the BIOS.)

 

[hang-the-9]

First try a clean Windows setup with new drivers.

If that does not fix things, possible power supply or video card issue, need to swap those out to test with new ones in your system and/or yours in another system to see where the issue is.

 

[KyleAparthos]

First try a clean Windows setup with new drivers.

If that does not fix things, possible power supply or video card issue, need to swap those out to test with new ones in your system and/or yours in another system to see where the issue is.

I actually just got done trying a clean Windows setup with new drivers about a week ago. No real improvement. Unfortunately I do not have access to a spare PSU or GPU.

However, one thing I have noted since I posted this: In the NVIDIA Control Panel, I disabled Adaptive Power Management and forced the system to prefer maximum performance. It's been about 18 hours since I did so and I have not suffered a crash since then. I'm cautiously optimistic, but not sure if this is a matter of luck or not.

Update: For whatever reason, after 30 hours of standard performance, the clock speeds of the GPU suddenly began to lock in to maximum level, and have crashed once since then. Driver reinstalls have been minimally effective at fixing this.

 

[Druzial]

I have a GTX 765m and have the exact same problems. all since this past summer. screen tearing while playing games only, BSOD, and complete freezes resulting in a hard reset. i have updated drivers. loaded old drivers. i run Furmark, with GPU-z and coretemp logging in background, and my GPU does bat an eye. but as soon as i load CS:GO, or WOW, or LoL artifacting screen and crashing. always with the same message nvidia drivers have failed and been recovered. the BSOD gives a BCCode of 116, which is GPU related. i have cleaned it and reapplied thermal paste and nothing seems to fix the problem. i am about ready to pull out the my OS stash and format all partitions. any advice on matter?