Some days ago, I experienced some odd lockups in Doom Eternal in Super Gore Nest, googled it, lots of people apparently had a similar issue. Worked around the problem by approaching a specific map area from another angle and by disabling intro video and bethesda login. Didnd't give it much further thought.
Then I wanted to try Portal RTX and the game would lockup and crash a few seconds into the menus becoming accessible. Having crashed Portal like 10 times I decided to try Cyberpunk, now that crashes every time too. I can get into the menus, but starting the in-game benchmark results in game crash maybe 10-20 seconds into the benchmark.
One odd detail: the benchmark seemed to be running way better than it should - on these same settings, normal would be 68-75 fps in the early parts, but right now I am seeing 90-100fps, which is way too unexpectedly good for Path Tracing at 4k on a 4070 TI SUPER.
Another possibly related VERY odd detail: sometimes MSI Afterburner flashes core and memory clocks values that make no sense whatsoever, such as 10000Mhz memory clock, which should be outright impossible for any GPU to have ever existed. This is without any overclocking or tuning of any sort applied.
Tried Quake Remaster, this doesn't crash, so apparently it's only the heavier modern games that have a problem.
Checked CPU and GPU temperatures, no issues there, GPU reaches 75c at the worst.
Tried disabling Hardware-accelerated GPU scheduling - no effect.
Tried mildly downclocking both the cores and the memory of my 4070 TI SUPER and slightly decreasing the power limit - no effect.
Tried a whole bunch of different driver versions, doing a DDU clean every time:
551.86 - Current latest driver, seemed to work for a few weeks, problem started yesterday
551.61 - Driver version I previously used without issues, problem persists
Tried a few older versions that seemingly fix similar issues for a lot of people:
546.65 - same problem as above
537.58 - doesn't recognize GPU, probably too old for 4070 TI SUPER
Reinstalled Windows 11, disabled every possible overlay (Steam, MSI Afterburner, Discord, etc), no effect. Then all of a sudden the problem seemingly went away for a week or so. Only to return again today and currently Cyberpunk often crashes before I even get to the main game menu. Tried "Prefer Maximum Performance" in NVIDIA 3d settings as well as tried enabling NVIDIA debug mode, but neither had an effect.
How do I properly debug this?
Then I wanted to try Portal RTX and the game would lockup and crash a few seconds into the menus becoming accessible. Having crashed Portal like 10 times I decided to try Cyberpunk, now that crashes every time too. I can get into the menus, but starting the in-game benchmark results in game crash maybe 10-20 seconds into the benchmark.
One odd detail: the benchmark seemed to be running way better than it should - on these same settings, normal would be 68-75 fps in the early parts, but right now I am seeing 90-100fps, which is way too unexpectedly good for Path Tracing at 4k on a 4070 TI SUPER.
Another possibly related VERY odd detail: sometimes MSI Afterburner flashes core and memory clocks values that make no sense whatsoever, such as 10000Mhz memory clock, which should be outright impossible for any GPU to have ever existed. This is without any overclocking or tuning of any sort applied.
nvlddmkm - Event 0
The description for Event ID 0 from source nvlddmkm cannot be found. Either the component that raises this event is not installed on your local computer or the installation is corrupted. You can install or repair the component on the local computer.
If the event originated on another computer, the display information had to be saved with the event.
The following information was included with the event:
\Device\Video5
Error occurred on GPUI100
Tried Quake Remaster, this doesn't crash, so apparently it's only the heavier modern games that have a problem.
Checked CPU and GPU temperatures, no issues there, GPU reaches 75c at the worst.
Tried disabling Hardware-accelerated GPU scheduling - no effect.
Tried mildly downclocking both the cores and the memory of my 4070 TI SUPER and slightly decreasing the power limit - no effect.
Tried a whole bunch of different driver versions, doing a DDU clean every time:
551.86 - Current latest driver, seemed to work for a few weeks, problem started yesterday
551.61 - Driver version I previously used without issues, problem persists
Tried a few older versions that seemingly fix similar issues for a lot of people:
546.65 - same problem as above
537.58 - doesn't recognize GPU, probably too old for 4070 TI SUPER
Reinstalled Windows 11, disabled every possible overlay (Steam, MSI Afterburner, Discord, etc), no effect. Then all of a sudden the problem seemingly went away for a week or so. Only to return again today and currently Cyberpunk often crashes before I even get to the main game menu. Tried "Prefer Maximum Performance" in NVIDIA 3d settings as well as tried enabling NVIDIA debug mode, but neither had an effect.
How do I properly debug this?
Last edited:
