As the title said, this error is really annoying.
My PC is 5900X+3080+32G running Windows11.
While running the project, my VRAM usage is only 40-50%.
I never get this error in UE4.26 or 4.25
It could happen even in the Default Third Person Template Project.
This error appeared so randomly. it could happen anytime(Add a node, connect a node, play the game, compile, create a folder, etc.)
I don’t think it’s a hardward problem?
Hope this can be fix in the next version of UE5
I got this problem when MSI Afterburner was running in the background. Perhaps that is the problem?
I dont have MSI Afterburner or any other GPU software
Did this start happening after you switched to windows 11?
Update: After I switch the RHI from DX12 to DX11, it didn’t crash for a long time.
Don’t really know why or how DX12 causing the problem.
Dang. I tried switching to DX11 and no luck. still instant crashes almost.
Oddly, a opened a project that isn’t related and left it open to test … same crashes.
It appears that once UE crashes from this, something in the program is corrupted or damaged and continues to crash. OR my 4 month old 3090 GPU is faulty and dying. My guess is its UE.
Do you have any other tips? I can’t open my projects anymore and have tried on 2 computers.
Unfortunately, there seem to be several of us affected by the same issue going on several months now. For example: Runtime D3D Crash of Ue5-ValleyOfTheAncients.
I haven’t seen any reliable solutions, and I frequently build the engine from source to see if there’s any fix incorporated in the 5.0 branch. Nothing has stopped it so far.
In my case, it started with migration to Windows 11, but some people seem to be affected on W10, too. So far everyone that’s listed their hardware seems to be on NVIDIA GPUs.
I am using AMD R9 390, no crashes with Unreal.
I am having random crashes, whether I use Unreal, or play league of legends, or just while browsing.
System restarts without any BSOD, and then sometimes keeps power cycling.
Sometimes it will just boot again and then crashes at the login window.
I have reinstalled windows 10, unplugged all parts in my cpu, and checked them one by one.
Did stress testing with AIDA64,
increased the GPU fan speed with MSI afterburner while running uniengine heaven and other benchmarks, without any crashes. Furmark increases the GPU temp above 85, so I do not use that ( even at 100% FAN speed).
Double checked the voltage of PSU with different software, Working fine.
I am at my wit’s end.
Sometimes like now it is working fine for hours and sometimes random restart with a continuous power cycle( keeps restarting and shutting down before the system boots).
I believe some Windows update is causing an error with AMD drivers.
or my GPU is going bad.
I will test with another gpu as soon as I get my hands on it but till then it feels like a windows problem.