PC reboots when playing games

[DaCypher]

My computer has been rebooting seemingly randomly while I'm in the middle of playing games.  There are no error messages, crashes, BSOD, or anything like that.  I am just playing a game and all of a sudden the screen goes blank and then I see my PC startup messages and Windows is booting up.  This has been going on for a few months now so I've tried a bunch of different things to try to fix it:

 

• Memory scans -- no errors
• Motherboard BIOS -- I upgraded to the latest version of the BIOS for my motherboard in hopes that the various "Improved system stability" items on the changelog would address my problem but to no avail.
• Overheating -- I ran some stress tests and my CPU was running a bit hot so I swapped out the stock Intel headsink/fan with an aftermarket one and now it tops out around 65 C (not great but seems to be within a healthy limits from what I can tell).
• PSU -- From the research that I've done, a faulty power supply can cause this issue.  I couldn't figure out how to test for this so I just went ahead and bought a new one anyways.  Still having the problem though...

 

Despite all of this, I've not made any progress.  Any ideas would be welcome...

 

Here are the specs of my PC:

 

• ASUS P8P67 LE Motherboard
• Intel Core i5-2500K CPU
• Radeon Rx 470 GPU
• EVGA 650 B3 PSU
• 8GB system RAM (I don't remember the specs here since it's been a few years since I bought it, but it was decent stuff and not the super cheap stuff).
• 512GB SSD and 2TB HDD (games are all installed on SSD).

 

Nothing fancy, no overclocking or anything.  Just using factory settings for everything.  I've got the latest drivers for everything as well.  Though I don't think that would matter given the nature of the problem.  It seems like this is a hardware issue rather than software/drivers, right?  I think I've eliminated the PSU from the equation since I replaced it, I doubt the hard drives could cause this, and memory tests seem to show no issues with the RAM.  Could there be an issue with the motherboard, GPU, or CPU itself?  I'm not sure how to diagnose those without just buying replacement parts which I'd like to avoid doing unless I know there's an issue...

 

 

 

 

[sc302 Veteran]

sounds like a gpu issue or your processor is going into a thermal shutdown...but it is probably your gpu doing that.  

 

Monitor your temps.  

 

http://openhardwaremonitor.org/

 

 

Use the logging option then check your temps upon reboot

[JustGeorge]

From what I've read, 2500K will begin throttling @ 95c. Doubtful its the CPU.

 

RX line of GPUs = Space Heaters. Could be the GPU or PSU.

 

Try re-seating GPU and if that doesn't work, swap PSU to test if possible, I'm betting defective PSU breaking under load.

[DaCypher]

Thanks for the replies.  I did actually buy a new replacement PSU.  The one I listed is the new one I bought about a week ago.  The previous one was 620W.  So it seems unlikely that both the old and new PSU would have the same problem resulting in the same reboot behavior, right?

 

I did actually reseat the GPU when I swapped the PSU, but I could try that again.  Are there any software tools to detect issues with the GPU like there are for testing RAM?

[sc302 Veteran]

https://www.custompcreview.com/articles/8-best-tools-stress-test-pc-cpu-ram-gpu/

 

here are a few tools.  but if you use the monitor I linked to earlier, you can get an actual reading up to the point of reboot from your gpu, cpu, motherboard, etc.  You can check to see where the load was as well as where the heat is at the time of reboot as it creates a log entry every second.  Why stress test where you could get an actual read of most traceable system monitors at time of reboot (cpu/gpu/mobo load, fan speeds, temp, voltage)...you have to enable logging to do this as it isn't enabled by default.  It creates a csv with those items and can be viewed via excel in text form (sorry no pretty graphics).

[Mindovermaster Global Moderator]

I don't think his PSU is at fault here. It is ether his GPU or CPU overheating. Do what SC302 said...

[DaCypher]

Ah OK great.  I've got OpenHardwareMonitor running now.  Hopefully it'll shed some light on this issue.  I'll post an update here the next time it happens...  Thanks again everyone for your help so far.

[DaCypher]

OK I been gaming for several hours tonight before the issue occurred again.  This time I was running OpenHardwareMonitor with logging enabled.  The CPU temp stays steady at around 65-ish.  While this is hotter than I'd prefer I think this is well below the cutoff limit (someone correct me here).  The GPU core temp seems to be hovering right at 69.  While it does get up to 70 a couple times through out, I notice that 70 shows up a bit more frequently near the end of the log file.  I wonder if the GPU could have a cutoff at 70/71?

Edited February 21, 2018 by DaCypher

[Mindovermaster Global Moderator]

You try newer/older drivers? I took a look around, and it might be a driver issue...

[+virtorio MVC]

Check the Event Viewer to see if there's anything of interest happening or going wrong at the time of the reboot.

 

There's a Windows 10 update that came out last week that caused one of the machines in my office to reboot (no blue screen, just restart as if I'd hit the Reset button), so you might want to give that a try if you're out of other ideas.

[DaCypher]

This has been going on for several months and I have tried a variety of drivers. Maybe I should try some really old drivers?  I'm going to try the drivers from ASUS site rather than the latest ones from AMD.

 

There's nothing interesting in the Windows event viewer.  It just says the system started after an unclean shutdown and the previous events are totally normal (very similar to if you'd just pull the power while the system is running).  It's got to be some kind of hardware problem that causes this...

[sc302 Veteran]

9 hours ago, DaCypher said:

OK I been gaming for several hours tonight before the issue occurred again.  This time I was running OpenHardwareMonitor with logging enabled.  The CPU temp stays steady at around 65-ish.  While this is hotter than I'd prefer I think this is well below the cutoff limit (someone correct me here).  The GPU core temp seems to be hovering right at 69.  While it does get up to 70 a couple times through out, I notice that 70 shows up a bit more frequently near the end of the log file.  I wonder if the GPU could have a cutoff at 70/71?

Cutoff/shut down is at around 90. Well within limits.   Is there a minidump file?

[winlonghorn]

3 minutes ago, sc302 said:

Cutoff/shut down is at around 90. Well within limits.   Is there a minidump file?

I am betting that this is a driver or update issue as others have mentioned above. I can't imagine what else it could be if you swapped for a new power supply and your temps are well within their limits as you stated. 

[sc302 Veteran]

If it is happening with the same game, it could even be the game. 

 

Minidump would point to a definitive driver/dll causing the issue.  If it is being created. 

[domboy]

Crazy as it sounds, I actually had a failing power button cause random shutdowns/reboots on my PC a year or so ago. I thought it was overheating, or bad power, or failing motherboard etc. But it ended up being the stupid little button. I doubts that's your issue as this was an older case, but I thought I'd mention it on the odd chance...

[farmeunit]

I had similar issue when going from Win10 1703 to 1709.  Had to revert back to 1703, so still there for now.  I have an RX580.  I tried multiple versions of older drivers along with newest.  Only rolling back to 1703 fixed it. 

[JustGeorge]

I’m surprised your 470 runs that cool. My RX480 hit 80-83c under load. Never malfunctioned though. 

 

As sc302 mentioned, if it’s a specific game that’s rebooting then it’s probably the game itself causing it. Multiple games makes it being hardware, driver or OS. 

 

Have you used a program called Display Driver Uninstaller to completely delete the old gfx drivers before installing new ones?

 

http://www.guru3d.com/files-details/display-driver-uninstaller-download.html

 

You’ve eliminated PSU already and your temps aren’t the problem. Can you roll back to pre FCU Windows to eliminate OS? Do you remember if the problem started after a major update? 

 

 

Edited February 21, 2018 by slamfire92

[+Mirumir Subscriber¹]

4 hours ago, sc302 said:

Cutoff/shut down is at around 90. Well within limits.   Is there a minidump file?

For older cards - yes, but the threshold is lower for newer cards imho.

 

My GTX770 used to shut down at 80-85 degrees. One of its two stock fans malfunctioned  

[Dysphoria]

Yeah my first guess was going to be restart due to high temps on CPU or GPU, but it seems like both are running at acceptable temps especially the GPU... It should be getting warmer than that unless you are having your case open while running the tests.

 

You might have already done this but when troubleshooting I try to bring everything to default settings, meaning reset BIOS to factory default, remove any overclocking you might be doing to your CPU, RAM and GPU. Overclocking introduces number of computational errors and it can cause freezes, reboots or BSODs.

Edit: missed that part...

Quote

Nothing fancy, no overclocking or anything.  Just using factory settings for everything. 

If that didnt fix it, try restoring to older versions of Windows updates and AMD driver sets...

 

If none of it works, back up important docs and do a fresh install of windows, that might eliminate software issues or bugs introduced through other third party applications and tweaks.

 

End game is get a new GPU and see if issue persists, most places have 30 day return policy... you can actually try that first and see if it's a hardware issue...

 

 

[Mindovermaster Global Moderator]

You could try to stress test your GPU, too. Kick the game issue out of the park.

[Brandon H Supervisor]

no bulging/bad caps that you can tell on the mobo is there? one of the only other things i can think of that hasn't been covered already

 

as others have said, grab another hard drive (so you don't have to kill your current install) and try loading a fresh copy of windows 10 or even try 7 to see if the reboot issue persists

[sc302 Veteran]

1 hour ago, Mirumir said:

For older cards - yes, but the threshold is lower for newer cards imho.

 

My GTX770 used to shut down at 80-85 degrees. One of its two stock fans malfunctioned  

According to the reviews in testing that I read up on, my numbers are probably accurate.  But there is no official source (that was quickly found) giving out max temps for that particular card.  

 

If you care:

http://www.legitreviews.com/xfx-radeon-rx-470-4gb-black-edition-video-card-review_184908/12

 

 

[theefool]

I know a couple mentioned memory dumps.  Have you checked to see if c:\windows\memory.dmp exists?  This is of course default for 8 through 10.  If Windows 7, check the c:\windows\minidump folder.

 

 

[DaCypher]

First of all, thanks again for all the replies, definitely some good ideas being thought up here...

 

@sc302 & @theefool -- no minidump files (in either location; I'm running Windows 10 1709)

@sc302 & @slamfire92 -- this has been happening in more than one game, unfortunately...

@domboy  -- interesting suggestion; I suppose I could test by temporarily disconnecting the power button from the motherboard and see if the problem occurs?  Maybe the additional fans running during gaming might cause vibration resulting in the faulting power/reset buttons triggering?

@farmeunit & @slamfire92 -- unfortunately I cleaned up the 1703 update after things seemed to be running smoothly for a couple months so I don't have an easy way to go back.  That said, I've actually had this problem for quite a while.  At least a year from what I can remember.  I just wrote off the issue previously as buggy games.  Lately I've been playing a lot of Assassin's Creed Origins for the past few months so it's clear this is not an isolated incident or particular to a certain game (and it's really bugging me since I'm playing more frequently).

@slamfire92 -- I haven't tried the display driver uninstaller, probably not a bad idea, I'll give that a go and try an older driver...

@Euphoria -- Case is closed so maybe I'm just lucky with the lower GPU temps?  The case is not that great for cooling, but I bought it because it was nice and quiet...  ASUS has a tool for adjusting various settings including over/underclocking the GPU.  I haven't touched any of the settings but maybe I should try underclocking it a bit?  Interestingly it also has a setting for desired temp of the GPU which it defaults to 60 (clearly it fails since it gets up to 70).  I wonder if I should try dropping that down and maybe it will more aggressively kick the fans on?

@Brandon H -- nothing obviously wrong on the motherboard from what I can tell (no bulging capacitors or anything; at least to my untrained eye).  2nd hard drive install of Windows is not a bad idea, but does seem like a bit of work for something that I am convinced must be hardware related...

 

Hopefully I didn't miss a reply from someone...

 

To summarize some of the ideas (and for my own notes on what to try next):

1. Power button -- try disconnecting the power and reset buttons
2. Display Driver Uninstaller -- cleanly uninstall Radeon drivers and go back to an earlier version.  Sometimes games complain about older drivers so I'll have to play with that a bit.
3. ASUS GPU TweakII -- play with the settings in this app (more cooling; over or underclock the GPU)
4. Fresh install of Windows
5. New GPU

 

[DaCypher]

Quick update: I decided to game a bit before trying any of the suggestions (can't a guy just have some fun ).  This time instead of rebooting the video just kind of died: monitor showed no signal as if I had unplugged the video cable.  I still had audio from the game and my keyboard and mouse seemed to still be responsive (changes in audio suggested I was still interacting with the world despite being unable to see anything).  Ultimately I had to hard reboot to get back into Windows again.  I do actually believe I've seen this behavior in the past but I think I just wrote it off as buggy games in the previously.  This is suggesting the issue is with the GPU itself.  Since this has happened in the past (probably over a year ago if memory serves...) then it seems unlikely to be fixed by falling back to older drivers.  It seems now my only real options are #5 (new graphics card) or maybe #3 (maybe playing with the settings in the ASUS utility will somehow resolve this?).

 

Here's a thread on the AMD forum that seems to have a bunch of people with a similar issue:

https://community.amd.com/thread/204905