Graphics Cards

[VanillaGorilla]

Computer Specs

GPU: MSI TRIO 6800 XT
CPU: Ryzen 3700x
Mobo: ROG STRIX X570-F GAMING
Ram: XLRB 32 GB 3600
PSU: Be Quiet pure power 1000 watt Gold
Storage: 2 x Samsung 1TB 980 evo

Ok so from what I can tell my computer is powering off due to over heating when I play certain games its only started happening recently and I have no idea how to fix it.

I do have 3 monitors with one running at 1440p / 144 hz and two running at 1080p on 60 hz.

For cooling on the machine I have an AIO water cooler for the CPU at the front of the case, two 120mm coolermaster fans at the top of the case and one at the rear.

From what I can tell using GPUZ my temps never go over 80+ for either the GPU or the CPU.

However when I play certain games for instance CSGO it will randomly black out all three screens and take forever to reboot unless a manual reboot is performed and the timing is random as in I could play for 2 hours or for 10 minutes before it happens, yet I can play New world or valorant for 6 hours straight with no issue.

I have cleaned out the pc for Dust, I have updated and rolled back graphics drivers, done Bios and all other driver updates.

I have tried running the games on lower graphic settings and switching off all the "extras" in AMD Adrenalin.

Any insight into this would be great

[Skrybe]

Have you tried stress tests to see if heat is the cause? I don't mean games, try something like Furmark for GPU, SuperPi for CPU and MemTest to test the RAM. Just be warned they generate a ton of heat so if you have problems you are likely to get crashes and there is always the risk it'll damage your hardware.

I'd start with CPU, since it's possible that your CPU is thermal throttling which keeps the heat below 80 but may not prevent every crash. A friend was having slowdowns and random crashes and it was the culprit.

Fire up something like HWInfo then start SuperPI. You'll be able to see the Thermal throttling section for your CPU in HWInfo and that'll confirm it.

Not knowing what the AIO cooler is it's impossible to say whether it's adequate for your CPU. It might be unable to deal with the amount of heat generated. Depending on how long it's been installed the coolant might be low, or have air bubbles trapped in it. Or the heatsink/paste between the CPU and the cooler might have a problem. Too much or uneven paste on the CPU can cause overheating. If the heatsink has been on the CPU for years then it's probably worth removing it and replacing the paste with new stuff. It can dry out and lose effectiveness over time.

Try the CPU first, if it's fine then move onto stress testing the video card with Furmark and watching the temps in AMD's performance monitoring. If that's ok then run MemTest and see if the memory is ok.

[VanillaGorilla]

Hey Thanks for the response!

The AIO is a CoolerMaster MasterLiquid ML240L V2.

I ran some stress tests you suggested for furmark the temps for the gpu only got to about 64c average and 87c junction.

As for CPU on super pi it only got to 48c.

It only happens with certain games and the timing is random which makes me think it isnt heat related.

[Skrybe]

Ok, assuming you ran those tests for more than a couple minutes, with those temps it's not heat related. Which rules out the AIO cooler as a problem too. A few more questions to see if we can pin something down;

What AMD drivers are you using? Video and chipset.

What bios are you running on your mobo? I had super hard to pin down random crashes related to fTPM on an Asus DarkHero board. The 4403 bios fixes (apparently) those fTPM issues. If you're running an older bios it might be worth trying the update. If it's really old there are a couple newer BIOS updates that include "stability fixes" based on their notes.

https://rog.asus.com/motherboards/rog-strix/rog-strix-x570-f-gaming-model/helpdesk_bios/

You mentioned CSGO, are there any other games getting crashes?