|
Problem description: My computer is randomly rebooting when I'm playing Player Unknown's Battlegrounds. This started happening a few weeks ago after I moved my desk to another side of the room and a different power point. I stopped using that power point for anything except my computer (and peripherals) and it seemed to go away. But tonight it's happening again, twice in 10 minutes. Both times under load and within 5 minutes of starting a game. Event viewer doesn't show anything and Google sort of indicates that my PSU might be struggling. I'm overclocking my CPU to 4G (have since I bought it in 2010 and have never had an issue) and my GPU to +60 Core and +140 memory (power limit at 100%) in MSI afterburner. What is the best way I can stress test my GPU overclock? Attempted fixes: I've checked temperatures (which I think are OK?) and looked through the Event Viewer. I'm monitoring my GPU temps using HWinfo and the GPU VRM temperature, is that the most accurate? It's quite different from the MSI temperature (higher). Under load that HWinfo temp. gets to 75-80 C but MSI says about 55-60 at the same time. Recent changes: Overclocked my video card slightly and moved my computer to a different power point. I can't remember if I was running an overclock on my GPU before I moved my desk but I think I was. -- Operating system: Windows 10 System specs: My PSU is an EVGA 500W Bronze one (model# 100-B1-0500-KR 500B 500W PSU, 80PLUS Bronze (MAL65528)) which I bought (along with my GTX970) in December 2015. Location: New Zealand I have Googled and read the FAQ: Yes
|
# ? Sep 13, 2017 10:35 |
|
|
# ? Apr 28, 2024 21:53 |
|
Reduce your overclocks back to stock and see if that fixes it.
|
# ? Sep 13, 2017 15:49 |
|
I thought that might be the suggestion. What's the best program for stress testing a GPU? Also which temperature sensor should I be looking at?
|
# ? Sep 13, 2017 19:44 |
|
Since you have stability problem in PUBG you'd play that to test to see if the problem is resolved. You're mostly paying attention to the GPU temperature, the VRM temperature doesn't really matter that much. I mean, the card doesn't work if the VRMs overheat, but unless you've jacked up the voltage for overclocking that is unlikely to be the problem.
|
# ? Sep 13, 2017 21:47 |
|
Turned my overclocks off but haven't played much this week. Was fine for about 2 hours then I got two restarts in 5-10 minutes. Both were in the Beta FPS mode so maybe it's just that? Are restarts a common graphics overclock symptom?
|
# ? Sep 22, 2017 12:41 |
|
System restarts usually point to a CPU overclock issue, the game crashing or the driver restarting points to GPU overclock.
|
# ? Sep 22, 2017 16:58 |
|
Could it also be a failing PSU? My CPU has been overlooked to a modest 4Ghz since I got it back in 2010 and I've never had any issues. It could just be old I guess?
|
# ? Sep 23, 2017 11:10 |
|
Yes, it could be a failing PSU, or even motherboard. Just to confirm, you did set the CPU back to stock and it still happened? It's always possible you just didn't adequately stress test the CPU at 4Ghz and it's really only stable under all workloads at 3.8 with that voltage/cooling, for example. Or stuff's just getting more marginal with age.
|
# ? Sep 23, 2017 18:18 |
|
I never touched my CPU overclock actually because I've never had an obvious issue with it. Might just have to put this down to PUBG being unoptimised and my components getting old I feel.
|
# ? Sep 23, 2017 22:39 |
|
Turn down the CPU overclock and see what happens.
|
# ? Sep 23, 2017 22:51 |
|
I probably won't be able to run PUBG anymore But I'll give it a go and let you know. Thanks for all your help!
|
# ? Sep 24, 2017 00:26 |
|
Reset my CPU overclock and played PUBG for about an hour today with about 6-8 reboots. Almost feels like it is getting worse. This is getting really frustrating.
|
# ? Sep 24, 2017 04:19 |
|
I would look at running a furmark burn in test, if its a power problem to do with the psu/gpu that should sniff it out quickly. Have you ever reseated your ram ( take it out and put it back in)?
|
# ? Oct 4, 2017 12:43 |
|
|
# ? Apr 28, 2024 21:53 |
|
Snakes in a can posted:I would look at running a furmark burn in test, if its a power problem to do with the psu/gpu that should sniff it out quickly. Have you ever reseated your ram ( take it out and put it back in)?
|
# ? Oct 4, 2017 17:33 |