[SOLVED] Help with repeated crashes during gaming

Mar 6, 2021
4
0
10
Hi all,

I am looking for an opinion on a series of crashes I am having on my PC. The PC is home built, and has been having issues more frequently since I refreshed the parts in it a year ago. The PC may go for periods of months without crashes, but lately it has been crashing once or twice per day for the past few weeks. It happens mostly during GPU-intensive gaming, the more intense the game, the more likely it will crash. Crashes usually look like the following:

  • All screens go black and PC immediately restarts (most likely)
  • Audio cuts out and screens sit frozen for a few seconds, or forever until I restart the PC.
  • A bluescreen appears with the message WHEA_UNCORRECTABLE_ERROR or MACHINE_CHECK_ERROR
(a couple times recently, Windows did a CHKDISK automatically after the crash but found no problems.)


My PC components are all very new. Specs are as follows:

  • Windows 10 Home OS, officially sourced and bought (Reinstalled 3 months ago)
  • Intel(R) Core(TM) i7-9700K CPU @ 3.60GHz (1 year old)
  • AMD Radeon RX 5700 GPU (1 year old)
  • 2 x 16 GB of ram (1 year old)
  • 4GB Hard Drive (1.5 Years Old)
  • 256 GB SSD (OS is on this) (1.5 years old)
  • Gigabtye B365M DS3H Micro-ATX Motherboard (1 year old)
  • 700 watt power supply (6 months old)
  • 5 case fans (older, still running fine)

Things I have already tried to fix the issue:

  • Reinstalled Windows 10 completely fresh: This was done in December 2020. This seemed to help a bit but the crashes came back within a couple weeks.
  • Crystal Disk Info identified bad sectors on my old OS drive, I copied my OS installation onto a new hard drive with more storage and removed the bad drive. This was done within the past month. It didn't help at all. Current OS drive has good health according to CDI, 4TB storage drive says "caution". I would rather not replace it unless it absolutely could be contributing to the crashes.
    EDIT: Added screenshot: https://drive.google.com/file/d/1dgZE1I8VCZjjlAAD4xRFR2drKLoDL_cV/view?usp=sharing
  • HWMonitor identified my CPU temperatures were going up to the high 80's, sometimes low 90's during intense gaming. I replaced the CPU fan with a newer model and applied fresh thermal grease. CPU temperatures went down considerably and CPU temp now hovers around mid-60's to low 70's during intense gaming. This may have helped mitigate the issue slightly, frequency of crashes went down. HWMonitor readings seem to be fairly normal.
  • Disabling all unnecessary startup programs and minimizing background processes.
  • Windows is completely up to date, so are all drivers. I reinstalled the AMD experience software a couple times and reset all the settings. I did attempt to update the BIOS last year, but not recently because I fear this may end up bricking my PC if it crashes during the update.
  • I dust my PC out every month very thoroughly. The CPU fan is completely clean, less than 2 weeks old. The GPU should also be fine, one crash I had indicated some power related issue with the GPU but then the ADM software reset the "wattman" settings and I never saw that error again.
  • Running diagnostics for the RAM and the CPU, no issues were detected.
  • Switching the RAM slots
  • Reseating basically everything in the PC (except the CPU, don't want to risk bending pins unnecessarily.)
  • Disabling the "AMD High Definition Audio Device".

Here is a copy of one of the latest crash dumps, occurred while I was playing Slime Rancher today. Sound immediately cut out, all monitors when black. One monitor started flashing shades of green momentarily, then went black again, and the system sat frozen for 15 seconds or so until I powered it off, then started it up again a minute later. I put the crash dump into windbg.exe and ran !analyze -show and !analyze -v. All I have managed to gather is that the error may be related to the CPU and heat. The error has mostly been the same among my collected crash dumps so far, but the process cited has also been "audacity.exe", "valheim.exe" and "blender.exe".

https://pastebin.com/m6BFVrwG

Please help, I fear I'm nearly at the point where I have to start replacing random components to see if that fixes the issue and I really don't have a lot of money to be throwing around. I shouldn't be having problems with this new of a PC, the only thing in it older than a year and a half is the case and the case fans. I'll take most suggestions and provide any additional info you need.

Thank you
 
Last edited:
Solution
Got another crash today, the kernel dump cited "ntkrnlmp.exe" as the image name. Looked it up and saw an article that recommended disabling EIST and C-states in the BIOS, so I did that. Valheim has been running stably for about 2 hours now... hoping this might have helped.

Literally haven't gotten a crash since I changed these BIOS options. Pretty sure I fixed it. Also saw that lots of other people with this processor model get the same issue. I hope this helps to spread the word a bit!
Hi all,

I am looking for an opinion on a series of crashes I am having on my PC. The PC is home built, and has been having issues more frequently since I refreshed the parts in it a year ago. The PC may go for periods of months without crashes, but lately it has been crashing once or twice per day for the past few weeks. It happens mostly during GPU-intensive gaming, the more intense the game, the more likely it will crash. Crashes usually look like the following:

  • All screens go black and PC immediately restarts (most likely)
  • Audio cuts out and screens sit frozen for a few seconds, or forever until I restart the PC.
  • A bluescreen appears with the message WHEA_UNCORRECTABLE_ERROR or MACHINE_CHECK_ERROR
(a couple times recently, Windows did a CHKDISK automatically after the crash but found no problems.)


My PC components are all very new. Specs are as follows:

  • Windows 10 Home OS, officially sourced and bought (Reinstalled 3 months ago)
  • Intel(R) Core(TM) i7-9700K CPU @ 3.60GHz (1 year old)
  • AMD Radeon RX 5700 GPU (1 year old)
  • 2 x 16 GB of ram (1 year old)
  • 4GB Hard Drive (1.5 Years Old)
  • 256 GB SSD (OS is on this) (1.5 years old)
  • Gigabtye B365M DS3H Micro-ATX Motherboard (1 year old)
  • 700 watt power supply (6 months old)
  • 5 case fans (older, still running fine)

Things I have already tried to fix the issue:

  • Reinstalled Windows 10 completely fresh: This was done in December 2020. This seemed to help a bit but the crashes came back within a couple weeks.
  • Crystal Disk Info identified bad sectors on my old OS drive, I copied my OS installation onto a new hard drive with more storage and removed the bad drive. This was done within the past month. It didn't help at all. Current OS drive has good health according to CDI, 4TB storage drive says "caution". I would rather not replace it unless it absolutely could be contributing to the crashes.
  • HWMonitor identified my CPU temperatures were going up to the high 80's, sometimes low 90's during intense gaming. I replaced the CPU fan with a newer model and applied fresh thermal grease. CPU temperatures went down considerably and CPU temp now hovers around mid-60's to low 70's during intense gaming. This may have helped mitigate the issue slightly, frequency of crashes went down. HWMonitor readings seem to be fairly normal.
  • Disabling all unnecessary startup programs and minimizing background processes.
  • Windows is completely up to date, so are all drivers. I reinstalled the AMD experience software a couple times and reset all the settings. I did attempt to update the BIOS last year, but not recently because I fear this may end up bricking my PC if it crashes during the update.
  • I dust my PC out every month very thoroughly. The CPU fan is completely clean, less than 2 weeks old. The GPU should also be fine, one crash I had indicated some power related issue with the GPU but then the ADM software reset the "wattman" settings and I never saw that error again.
  • Running diagnostics for the RAM and the CPU, no issues were detected.
  • Switching the RAM slots
  • Reseating basically everything in the PC (except the CPU, don't want to risk bending pins unnecessarily.)
  • Disabling the "AMD High Definition Audio Device".

Here is a copy of one of the latest crash dumps, occurred while I was playing Slime Rancher today. Sound immediately cut out, all monitors when black. One monitor started flashing shades of green momentarily, then went black again, and the system sat frozen for 15 seconds or so until I powered it off, then started it up again a minute later. I put the crash dump into windbg.exe and ran !analyze -show and !analyze -v. All I have managed to gather is that the error may be related to the CPU and heat. The error has mostly been the same among my collected crash dumps so far, but the process cited has also been "audacity.exe", "valheim.exe" and "blender.exe".

https://pastebin.com/m6BFVrwG

Please help, I fear I'm nearly at the point where I have to start replacing random components to see if that fixes the issue and I really don't have a lot of money to be throwing around. I shouldn't be having problems with this new of a PC, the only thing in it older than a year and a half is the case and the case fans. I'll take most suggestions and provide any additional info you need.

Thank you
Post a screenshot of CDI for each disk.
 
Mar 6, 2021
4
0
10
Got another crash today, the kernel dump cited "ntkrnlmp.exe" as the image name. Looked it up and saw an article that recommended disabling EIST and C-states in the BIOS, so I did that. Valheim has been running stably for about 2 hours now... hoping this might have helped.
 
Mar 6, 2021
4
0
10
Got another crash today, the kernel dump cited "ntkrnlmp.exe" as the image name. Looked it up and saw an article that recommended disabling EIST and C-states in the BIOS, so I did that. Valheim has been running stably for about 2 hours now... hoping this might have helped.

Literally haven't gotten a crash since I changed these BIOS options. Pretty sure I fixed it. Also saw that lots of other people with this processor model get the same issue. I hope this helps to spread the word a bit!
 
Solution