[SOLVED] Computer crashing after a game runs some time

[Nakroma]

Hey, heres my issue:
Since a few days ago, after running a game for some time, my entire computer crashes, both monitors flash a solid color which is random each time (I assume picked from the current color pallete on screen). It's generally random when it crashes, but seems to appear faster on graphically complex games (like Apex Legends) and slower on less complex ones (like Northgard).

First of all: I've ruled a software defect out, I just completely reinstalled my OS and the problem persists. It only seems to appear in games, thats why im posting in the Graphics Card board. I've also updated drivers in the meantime, the problem remains.

Specs:
Windows 10 Pro, 10.0.18362 Build 18362
AMD Radeon R9 280 with driver 19.7.3
Intel Core i5 4460
Motherboard: ASRock B85M Pro4

Naturally, I've thought it might be a temperature issue, with the current temperatures and all. Running GPU-Z, the temperature goes quite high (peak 97° gpu temp with 100% fan speed), however im not sure if the crashes are related. For example, if you look at this log while playing a game, my temperature stays at ~93° quite long, then goes down to ~80° (when switching to a loading screen) and only then it crashes: https://pastebin.com/Sm24uiu6

Any ideas what could cause it? The case is also freshly cleaned.

 

[cin19]

Running GPU-Z, the temperature goes quite high (peak 97° gpu temp with 100% fan speed)

I think the GPU temp is too high. Look at these reviews, the temp should be less than 70C, because you didn't post which one you had.

Also what PSU did you have?

Recommend to check your GPU fans to see they do work or not. If the fans work and you had a PSU that can handle the r9 280. You should try to re-apply the thermal paste on the GPU.

 

[WaltzWorker]

Could be that since you've been over temping it it is now saying enough is enough... All drivers updated too?I know at some point you may find that the drivers aren't updated anymore - which we find all the time at work lol If you did a stress test/performance benchmarking at some point you may find now that the numbers maybe lower... Not sure how else to help. OH! If your PS is 'old' too - that can also degrade the system - pulling too much power from it... Not sure other than hooking up some stuff to see how much power it actually puts out...

 

[cin19]

450W PSU ( be quiet! System Power 7 ) with r9 280, they are fine, but if it is old, like other member siad, you may get problem, because the whole system will draw almost 400W, overtime you know what happens.

 

[cin19]

You need to test the r9 280 in other PC to see it works fine or not. If it will work fine, that means your PSU need to replace. If the 280 does the same, you need to buy the new GPU, or use the software ( like MSI afterburner) to tweak the voltage or speed etc. It will work sometime.

 

[WaltzWorker]

With just the MB, CPU & Card (can't find exact) but your pushing 400 Watts on the power supply. If you got 400 Watt power supply that could be it... You've mentioned that you reinstalled the OS so it can't be the game files, maybe download and install a fresh copy too? Always update the drivers for everything you can... I usually put the 'updated' ones on a CD/DVD and save... OH!!! This article
What is the safe temperature value for i5 4460?

may help - just above the graphic it says: "Core temperatures above 85°C aren't recommended. "

 

[Nakroma]

With just the MB, CPU & Card (can't find exact) but your pushing 400 Watts on the power supply. If you got 400 Watt power supply that could be it... You've mentioned that you reinstalled the OS so it can't be the game files, maybe download and install a fresh copy too? Always update the drivers for everything you can... I usually put the 'updated' ones on a CD/DVD and save... OH!!! This article
What is the safe temperature value for i5 4460?

may help - just above the graphic it says: "Core temperatures above 85°C aren't recommended. "

I think the GPU temp is too high. Look at these reviews, the temp should be less than 70C, because you didn't post which one you had.

Also what PSU did you have?

Recommend to check your GPU fans to see they do work or not. If the fans work and you had a PSU that can handle the r9 280. You should try to re-apply the thermal paste on the GPU.

Thanks for the answers!
My exact GPU is this one

I have a be quiet! System Power 7 with 450 Watt, which should be enough (it did work for a few years now)

The fans work, thermal paste is a good point, I should do that, since I didn't change it since I build the pc.
However, I'm still not sure if it's actually a temperature issue (altough it is definetly too high). Right now for example, I started up a game and it immediately crashed, at only 58°.

Also, I've noticed Amazon Prime and Netflix crashing sometimes, a few weeks ago, not YouTube and other similiar sites though. The crashes are different, it just freezes the entire screen. Not sure if that's related at all, I just wanted to bring it up, just in case.

 

[igg-games]

450W PSU ( be quiet! System Power 7 ) with r9 280, they are fine, but if it is old, like other member siad, you may get problem, because the whole system will draw almost 400W, overtime you know what happens.

 

[Nakroma]

Reapplying thermal paste dropped the temperature by about 18°! However, that did not fix the issue. I'll probably buy a Asus Expedition or ROG Strix RX570 O4G. With that, I cover both bases (either my PSU is running down and now I use less power, or my GPU broke somehow and I replace it). Does that seem like a good idea?

 

[Nakroma]

It was not the PSU, sadly. Put the GPU in another pc, same crashes. Bought a new one, thanks everyone for the help!