Question PC restarting while playing games and CPU stress testing

Tomascari

Honorable
Sep 20, 2014
13
1
10,515
0
Hi everyone,

I just built my first PC and it works like a dream... except it occasionally shuts down and restarts while gaming.
The specs are (all brand new components):
CPU: Intel i5 11600k (with Vetroo V5 Air cooler)
Motherboard: Asus ROG Strix B560-I Gaming Wifi
RAM: 2x8GB Corsair Vengeance (3200MHz DDR4)
GPU: Gigabyte RTX 3060 Eagle OC
Storage: Crucial P2 1TB M.2 NVMe
PSU: Silverstone SX-700 G

I've tested on 3 games so far. Rocket League has worked fine every time. However, Watch Dogs: Legion and Star Wars Jedi Fallen Order both crash after about 15 mins to an hour. Both have done so twice. When the system crashes, it simply freezes for a few seconds and then completely shuts off (not even case power light stays on) and then restarts itself.

Things I have tried so far:
  • Drivers are fully up-to-date.
  • Monitoring temps. Nothing seems to be overheating and definitely hasn't exceeded 70C when it has crashed during gameplay.
  • CPU stress testing. This reproduces the crash and restart every time (run about 6 times now). I have been using Prime95 first with the blended test and then with the small FFTs tests. Small FFTs shut it down more quickly (5-30mins in) but each blended test shuts down well within an hour. Also doesn't appear to bring up anything useful in Windows Event Viewer (just "critical: The system has rebooted without cleanly shutting down first. This error could be caused if the system stopped responding, crashed, or lost power unexpectedly.")
  • GPU stress testing. First tested on Uingine Superpostion - no problems at all. Then ran for an hour on FurMark and again there were no problems (may test again to see if I just got lucky). GPU temp never exceeded 71C.
  • MemTest64. Ran this twice which brought up no errors.
  • Updated the BIOS. This made no difference in Prime95.
  • Reset XMP in the BIOS. This made no difference in Prime95.
To me this suggests there is an issue with power delivery to the CPU. But the GPU stress testing suggests that the fault may not be with the PSU itself (let me know if this may be wrong). I feel it's unlikely to be caused by the CPU itself so that leaves the motherboard. I don't really know what else to try. I don't really have any components I can swap in but I may be able to try the PSU from my 10 year old desktop (very dusty and may be underpowered).
It may be worth knowing that I had a similar problem with my old desktop up until I stopped using it about 5 years ago. However, I believe that was BSOD but mainly occured while gaming as well. I don't remember the details though.

Has anyone got any suggestions on how I could troubleshoot any further? Is it possible something as simple as a dodgy fan is tripping something maybe?

I appreciate any help anyone can give, cheers.
 

Lutfij

Titan
Moderator
If the system reboots when taxed regardless of a game or a stress testing app, it's due to either;
a| thermal issues, overheating...can be any component in the build, even an overheating VRM on the motherboard or chipset
or
b| your PSU is incapable of delivering necessary power to the entire system

Seeing how you have 700W at your disposal, perhaps try and swap out the PSU for a donor unit to power up the build. Might it be possible to remount the CPU cooler to rule out an improper contact/mount? To also note, perhaps try and breadboard the build and inspect each part. Which two slots are the rams populating? A BIOS version would help us understand if you've got anymore BIOS updates pending.

Version for your OS? Did you fabricate your bootable USB installer using Windows Media Creation Tools?
 

Tomascari

Honorable
Sep 20, 2014
13
1
10,515
0
If the system reboots when taxed regardless of a game or a stress testing app, it's due to either;
a| thermal issues, overheating...can be any component in the build, even an overheating VRM on the motherboard or chipset
or
b| your PSU is incapable of delivering necessary power to the entire system

Seeing how you have 700W at your disposal, perhaps try and swap out the PSU for a donor unit to power up the build. Might it be possible to remount the CPU cooler to rule out an improper contact/mount? To also note, perhaps try and breadboard the build and inspect each part. Which two slots are the rams populating? A BIOS version would help us understand if you've got anymore BIOS updates pending.

Version for your OS? Did you fabricate your bootable USB installer using Windows Media Creation Tools?
Thanks for your response.

I may be able to try my old PSU but other than that, I haven't got easy access to another.

One thing I forgot to mention was running HWinfo during the CPU stress tests. For the most part, the CPU maxed out about 75C but interestingly, on the last couple of runs, it was hitting 100C. It was crashing in either case and when gaming, the CPU was typically about 60C but still crashing and rebooting. Nothing else got very high, VRMs for instace stayed under 45C.

The motherboard is mini-ITX so only has 2 RAM slots (both filled). And am now running BIOS version 0503 which is the latest on their website (board came with 0403 installed originally).

I'm running Windows 10 installed through a USB stick. I simply got the download from the MS website last week and put it on the USB following the instructions they give. When it was installed I kept runing the updater until there was nothing left. Currently on OS build 19042.964, version 20H2. I do have one optional update though which is 'Intel - net - 22.50.0.7' driver update.
 

ASK THE COMMUNITY

TRENDING THREADS