Generally speaking its a bad GFX that is at fault.
It crashes when it overheats, which is why it runs for about 2 minutes before the crash.
If you can furmark your gfx for over an hour, then remeber that furmark is nowhere near the GPU load that unreal puts on it.
You can try stuff like occd at your own risk (since you can fry the pc with it like they say in their release/install process) to manually test the load and see at what point it fails progressively.
It’s a job/skill in itself.
Unless you are 99% sure your gfx is not at fault - by running some other ue5 project for a few hours or similar - then just assume you need a new one…
Btw, assemby references are all baked on whomever’s sytem was used to bake them. So a link to a d drive is not at all an indication of a problem.
You can try building the engine from source - if the DLLs used are also built along they’ll be referenced to your local setup. If they arent, then their symbols would point to wherever the system of those who packaged them pointed.
Probably, google up the name of the cpp file to find some version of it and get an idea of what it is.
In the case of
DXGI_ERROR_DEVICE_REMOVED with Reason: DXGI_ERROR_DEVICE_HUNG
Its pretty clear isnt it.
The gfx was removed because it hung.
Ergo, either a bad card, or bade code that is causing it to loop forever.
Since others were (at that time) able to run the project just fine, then the GFX was likely at fault.
Or the driver - remeber that bad drivers can cause the same issue without the hardware necessarily being bad.
Either way, if you get a device hung error, your device is at fault. One way or another. Drivers, instability, incompatibility, whatever the actual cause of it may be.
(Including but not limited to Power Load. If your psu can’t keep up with the power needed instead of powering down your system it will disable the item requesting the power or cycle it on/off).