Discussion PC Black Screen When GPU Under Load (FIXED Somehow)

Status
Not open for further replies.
Oct 20, 2021
2
0
10
I thought I'd post this since it seems so random and maybe it could help someone and possibly give me some advise on if I should do anything else. The recommendations online were either it's a Driver Issue, PSU, the GPU being faulty, and a rare occasion could be the motherboard being faulty.

Problem and Testing:
So out of now where my GPU would lose signal to my monitor when under heavy load. I would still be able to hear the game in the background even though my screen was black and I would hear the coil whine on my GPU stop. It only happened on demanding games or if the scene was really demanding.

So I tried a number of things like test it with a different PCIE GPU PSU Cable, a different PSU(EVGA G2 Gold), reinstalling Windows 10, reseated the RAM. Gave my PC the good old OXY clean. Nothing would work. So I went back to the drawing board. That's when I noticed my GPU would crash when MSI afterburner was reading around 250 to 350 Watts of Power. And the way I found this out was by under volting the GPU, lowering the power limit, lowering the core clock, and the memory to reach a lower Wattage. I was able to be underload and play my games without getting the black screen for about 2 weeks. I wasn't entirely happy about losing a lot of performance but I thought it was fine for now until I get another GPU. However, it came back on a certain scene on one of my games and my last resort was to take apart my hardline water loop to see if cleaning it would fix the issue.

Solution:
I had 0% hope that me taking the GPU block apart, cleaning it, and reseating it on the MB would solve the issue but it somehow did fix the issue. I just took apart the water block then I noticed the thermal pads were a little to big so I decided to cut them a little bit and removed the paste and applied it again. Made sure the PCB was clean but I didn't really do much really in terms of cleaning it was mostly just repasting it and putting it back together but I made sure to not tighten the screws a lot. I made note of doing this. But idk what fixed it ultimately. Fingers crossed it doesn't come back but it has been about a week and I am able to overclock the GPU again and shoot my GPU's load at my games again!
I kind of feel like I should go back to the EKWB stock thermal pads though. Instead of getting the ones I got from amazon which were more expensive and probably not worth the price jump.

Didn't want to RMA:
I didn't bother RMA-ing the GPU since I've had the GPU more than 3 years and I believe NVIDA doesn't do warranty on GPUs past 3 years. Also I took apart the NVIDIA cooler and put my own water block, and not to mention I also Flashed the GPU's BIOS from a different GPU to increase the power limit on my 2080 ti...

My PC Specs: CPU and GPU water cooled.
CPU: 10900k
GPU: 2080 ti
Moth: z490g
RAM: Corsair Memory 16GB
PSU: Corsair SF750
Resolution: 3440 x 1440p

***UPDATE: Seems like the potential issue could be a mounting issue. Such as one user over in reddit mentioned that when they banged their desk the GPU would not make good contact on the MOBO slot and would need to be pressed back in. This could be the same Issue I had. Instead of me cleaning it maybe it just needed to be reseated but since it was water cooled I decided I might as well open the water block up. Hope it helps someone!
 
Status
Not open for further replies.