GPU Going Under Load Too Easily, Crashing When It Does

Larry_92

Commendable
Jun 13, 2016
4
0
1,510
For the last week or so I've been having problems with my GPU crashing and it's gotten progressively worse. Here are my system specs first:

GPU: XFX Radeon HD7850 2GB
CPU: AMD FX4100 3.6GHz
RAM: G.Skill Ripjaw 8GB (2x 4GB) DDR3
PSU: Seasonic M12II 620W
Motherboard: ASUS M5A97
Case: Cooler Master HAF 932 (4x case fans which come with it. Front and side are intake, top and back are exhaust)

All of these parts besides the GPU are about 4 years old, the GPU is about 2 and a half years old.

So now my problem: For the last week my GPU is going under heavy load way too easily compared to before. It started off with it going under load after a few hours of gaming but now today it will go under load within seconds of starting a game (I tried both Dota 2 and Overwatch and it happens while I'm still on the main menu). Not only is it going under load easily but when it does it completely shuts down. The rest of my PC stays on but I lose video completely. At first I thought it might just be overheating so I took my entire case apart (motherboard, GPU, etc. all completely out of the case) and thoroughly cleaned everything. I took apart the heatsink and fan for both the CPU and GPU and removed the old thermal paste and applied new and also cleaned all the dust from both.

After putting everything back together my idle temperatures are great (35C for GPU, 30C for CPU, 30C for motherboard) but I'm still having the same problem with my GPU crashing when I try to do anything intensive at all. When I start up Overwatch the temperature and RPM on my GPU will go from 35C and 1000RPM to 80C and 3500RPM almost instantly and then the crash happens. From what I understand 80C is fine for a GPU under heavy load so I'm not worried about that but it shouldn't be going under heavy load so easily and crashing once it is. Other than cleaning I've also tried reverting to older drivers and trying new beta drivers for my card and neither made any difference. I used the display driver uninstaller from Guru3D before installing the old/new drivers just to make sure and no luck. My CPU and motherboard temperatures stay relatively normal when I'm opening up these games so I'm pretty confident the problem is GPU-related but I'm not sure exactly what it could be other than the card just going bad.

So yeah, lots of text but I wanted to be as detailed as possible. Any help would be super appreciated.
 
Solution
If a different Windows setup did not help, then it's likely a card issue. Safe mode does tend to work even with bad cards. Maybe a power supply thing also, if you can try the card in another system see if it does the same thing there.
When you say "going under load" what do you mean exactly? The fans run fast? Or the GPU usage goes up to a higher % than it has before? That is often the heat issue, if not heat, could be a bad video card, maybe power supply, maybe motherboard.

Best thing to do is test the card in another system that can run it, if issue follows card, bad card.
 

Larry_92

Commendable
Jun 13, 2016
4
0
1,510
When I say under load I mean HWMonitor says the utilization is going up to 92% just from being on the main menu screen of Overwatch or Dota 2 and that has never happened before. It shouldn't be a heat issue because, as I said in the original post, I cleaned the entire system and eve reapplied new thermal paste. Even when it's under load the temperatures are in normal range (80C at 92% utilization). The problem is its going under load when it shouldn't (idling on the main menu of a game) and turning off when it does.

I don't have anyway to test the card in another PC. =/

Edit: My PC keeps blackscreniing even when I have nothing open now. I didn't hear the fans speed up before it happened this time.
 

Larry_92

Commendable
Jun 13, 2016
4
0
1,510
I restored to a point from two weeks ago (before this all started) last night and it didn't help. I haven't changed any drivers since way before the problem started.
 

Larry_92

Commendable
Jun 13, 2016
4
0
1,510
Okay so I took out my current HDD that has Windows 10 installed and put in one of my older ones that has Windows 7 and got the same exact results. As soon as I logged into Windows I got a black screen and the video card shut off.

After putting my Windows 10 HDD back in I decided to try booting into safe mode with networking and that is actually working (obviously can't test games but Windows itself is working which is better than it was doing in a normal boot). I'm thinking the video card is the problem and safe mode works because it's much less intensive than regular Windows on the GPU. Does that sound right or do you think it could be something else?