GTX 780 Ti hits core clock then dies

Habbasi

Honorable
Oct 15, 2013
58
0
10,530
I recently got my GPU replaced/repaired by PCspecialist who I bought my custom PC from. I got it over a year and a half ago and its been wonderful. However, a few months ago the GPU just died and the screen would show up with vertical green, yellow and blue lines and would just be so slow/unusable.

I sent it in a while later under manufacturers warranty, and it came back yesterday. I plugged it in and the lines error was gone but now when I play some games on higher graphics the GPU dies I guess and the monitor goes black, losing connection as it is connected via HDMI right into the GPU. I then have to restart the PC via the reset button near my power button so that it begins to work again.

I took a look at all the fans/gpu etc. and it all runs fine after the black screen but when I downloaded GPU-Z and checked the temp etc. here are some screens from the GPU-Z window:

n8u.png


Thats the info on the GPU itself and here are the screens for the GPU sensors under normal usage:

cd7.png


However, when I play games that before it broke down for the first time, they could run at max settings and would have absolutely no problems. E.g. CSGO max settings, GTA 5 best settings I can reach while staying in 3GB memory limit and more. But now when I play them on those settings or even csgo on mid-low settings the core clock is what goes insane and far above its default clock or even the boosted clock. This is surprising as running csgo was no problem before but now as soon as the core clock begins maxing out the gpu dies and woop black screen :/

Here is a pic of GPU-Z whilst csgo is running on mid settings:

4rx.png


This is really bugging me as mid settings should be nothing for a GTX 780 Ti on csgo right?

Thanks for reading and if you can help me :/








 
Can you play a more demanding game than CSGO and post the sensors section again? CSGO is barely touching your gpu, (look at gpu load while your playing csgo) its barely doing anything. I want to see if its still under such a low load even in a more demanding game.
 
Other basic spec:
6bj6wH7.png


Wow while I was posting this and testing gta V on mid-high specs just as I turned shadows on the pc died, I barely got a look at the load but it looked to stay between 0-5%

restarted and now using geforce experience optimal settings as a base and here is the gpu-z screen before i crash again rofl:

rf2.png


Highest readings:

esb.png


Beginning to wonder if it could be a PSU problem? GPU was just recently replaced/repaired took 3 weeks to get it back :/

VDDC values:

4hg.png


Thanks for the quick replies

 
I really dont know whats going on there. What power supply do you have make and model? The max GPU load is only at 67% it should be hovering around 90-100% especially with a 4770k, so it cant be a cpu bottleneck unless its overheating badly, have you checked your CPU temps before the crash occurs? I would guess its some kind of driver issue or like you already mentioned it could be related to power especially with all the crashing, it just goes black and restarts?