0

I have so far had 4 instances of my PC randomly crashing. As far as I can tell nothing overheated and there was no common action prior to those crashes that would be a clear cause of the issue.

The first thing that happens is that my display would show something like this while the sound on my headphones suggests activity for at least a few more seconds until I just get some weird scratching noise and everything just goes black, despite all lights/fans/etc. in my PC still running.

On the first 3 crashes any attempt at rebooting resultet in no display signal to the monitor unless I tried to fix the problem that I could not identify in some way. What I've done so far is: Crash #1: Clean out PC interior. Unplug the power and hold down the power button for a minute. PC successfully booted afterwards. Ran /chkdsk c: /f /r & Windows 10 Memory Diagnostic Tool with no results. Crash #2: Removed RAM and switched on PC, resulting in continuous short beeps, supposedly signaling a correctly functioning mainboard. Cleaned RAM with a rubber and put it back in. PC successfully booted, no further checks were run. Crash #3: I took the picture referenced above and reset the CMOS with a screwdriver on the 2 respective pins. PC successfully booted afterwards. Ran WhoCrashed resulting in

On Thu 17.08.2017 13:03:46 your computer crashed crash dump file: C:\WINDOWS\memory.dmp This was probably caused by the following module: Unknown (0xFFFFF8031EC07560) Bugcheck code: 0x116 (0xFFFFBE8A1DFC8010, 0xFFFFF80E8574F7F8, 0x0, 0xD) Error: VIDEO_TDR_ERROR Bug check description: This indicates that an attempt to reset the display driver and recover from a timeout failed. A third party driver was identified as the probable root cause of this system error. Google query: VIDEO_TDR_ERROR

(Despite there being 4 of those crashes so far, this is the only crashlog that matches time and type of the issue after checking with WhoCrashed. None of the other 3 crashes seem to have registered with the system logs.) Now this led me to believe that my GPU could indeed be somehow involved since I had some (supposedly) unrelated troubles with my GPU driver in the past. While trying to play games like League of Legends or Dungeon Defenders 2 I would be unable to launch the actual game due to DirectX errors in some form. Back then my research led me to this issue, with fixed my problem for me by running an older driver until recently, when AMD supposedly fixed the issue and I have since been able to freely update my drivers again. Problem here is, though, that there hasn't been any driver update coinciding with the beginning of the crashes on thursday.

I am at a genuine loss of ideas on what to check or do next since I seem to be unable to identify any cause of the issue or even reproduce it on purpose, as there is until now no consistent time of using the PC until a crash occurs nor a specific activity being carried out when it does.

For further information I have uploaded a recent DxDiag but am not allowed to post a third link due to my low reputation. My graphics card is a MSI TwinFrozr R7950 and the older (working) driver used before was 17.3.3 while I am now using 17.7.2.

Thanks in advance for everybody trying to help

Improve this question
9
  • 1
    dropbox.com/s/o6lmt4fsjdf1fty/DxDiag.txt?dl=0 Missing DxDiag
    – Adrian Zielony
    Commented Aug 19, 2017 at 13:45
  • You will have to configure your system to generate a dmp file instead of automatically rebooting, then use windbg to analyze the crash, to determine the cause. Your DxDiag report isn't helpful in a case like this.
    – Ramhound
    Commented Aug 19, 2017 at 15:42
  • @Ramhound, generating a dump file is not an alternative to automatically rebooting. Automatically rebooting is an alternative to keeping the BSoD on the screen until someone hits reset or power-cycles the machine. If the machine is configured to generate a memory dump you get one whether or not it's configured to automatically reboot after the crash.
    – Jamie Hanrahan
    Commented Aug 19, 2017 at 22:52
  • Adrian: WhoCrashed found only your \windows\memory.dmp file because Windows only ever keeps the most recent one. However you should have a collection of "minidump" files in \windows\minidump . These have far less information than the memory.dmp file but a collection of them can often give clues. Automatic tools like WhoCrashed can only rarely give a good answer - even experienced debuggers can't solve every dump, even full memory dumps. But if all the minidumps also point to the video driver it's a good indication of a hardware issue with the GPU.
    – Jamie Hanrahan
    Commented Aug 19, 2017 at 23:00
  • @RickBrant I know that. Easier to say what I said to get the information required to help then explain all that in detail
    – Ramhound
    Commented Aug 20, 2017 at 2:16

1 Answer 1

Reset to default
-1

From what i can understand, this seems to be a hardware or driver related issue, since the monitor gets no signal after your crashes and from the fact that you are getting some messy screens.

If you are running windows, simply changing the driver or reinstalling it won't do much, since windows will install a temporary generic GPU driver until you install it again. I had such an experience where I had a similar screen artifacting issue, and reinstalling the drivers didn't help, I had to use system restore for it to finally work.

Another thing you should check is that if you have your GPU overclocked then consider turning that off as it may cause instabilities and crashes.

Also, if you think you can trigger the crash, then try doing so from safe mode and see if that works. If it also crashes from safe mode, then your problem is not likely related to the GPU.

I hope this helps.

Improve this answer
3
  • As I said, I have no clue as to how I could intentionally trigger the crash. I have also never overclocked any part of my PC, so unless that can happen on accident, it should not be the cause. What I've done so far and failed to mention in my initial post was boot into safe mode and uninstall/reinstall my driver with DDU.
    – Adrian Zielony
    Commented Aug 20, 2017 at 10:24
  • Seems that you did everything i would've done and i can't think of anything else. Now one more thing i would try, would be to change the PCIE slot that the GPU is on, if the problem is from hardware, then perhaps this could help trouble shoot and narrow down where the problem is. Not much help i can offer other than that, my apologies.
    – FarisHijazi
    Commented Aug 21, 2017 at 11:08
  • I have started testing with a different GPU and will proceed testing with a different PCIe slot and different RAM, depending on results. If anyone has another hint on how I could force this crash or what more to check, I'd appreciate that very much.
    – Adrian Zielony
    Commented Aug 31, 2017 at 21:11

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .