Space Engineers saying GPU is overheating

I have read that some newish intel CPUs have a performance issue that requires the CPU to be under clocked.The fault often reports the problem with the GPU, but it was discovered, by a game dev team that the fault was with the CPU. Some of the CPUs had to be replaced as the fault was permanent, but do not take my word for it, I would suggest you read up on intel news, forums, videos, etc. ASAP.

According to Wikipedia Intel has extended the warranty period on your CPU.

https://en.wikipedia.org/wiki/Raptor_Lake#Two-year_warranty_extension_for_certain_models

The problem is affecting Gen 13 and 14 processors running 65w and over.

Hello Erik,

thank you for reaching our forum with this problem that you are experiencing. We are sorry that it is happening to you in a first place.

Sadly, the error "HRESULT: [0x887A0005]" that you are having the logs, suggest that there is actually problem rather with your HW, then actually with the game.

The problem/error mentioned is a hardware-related issue which usually caused when the GPU is overheated/overclocked or PCI port malfunctions. To resolve this you could try the following steps;

Check GPU/system temps/voltages with something like HWMonitor - if temperature issues, check GPU and system fans are clean and running

Switch GPU driver settings to high-performance (i.e. not quality/high quality)

Update to the latest GPU drivers (or try clean installation)

Update your OS

If the GPU/system is over-clocked, revert to stock clock speeds

It is also possible that the port/card is just not seated properly.

If the issue still occurs after performing these steps, try the following workarounds as well;

Verify your game files on Steam - right-click on Space Engineers in Steam Library → Properties → Installed Files → Verify Integrity of Game Files

Delete SpaceEngineers.cfg file (it means your game configuration will be deleted) - you can access your SpaceEngineers.cfg by typing %appdata% into your Windows search bar and you will be redirected to the hidden Roaming folder. After that just follow: \Roaming\SpaceEngineers.

Delete ShaderCache folder - same access as SpaceEngineers.cfg

Also, try disabling all other displays except one or try to change the multi-monitor setup.

Please let me know, if anything here above helped you with the problem. If not, please provide me new logs from the game from the current time.

Thank you in advance.

Kind Regards

Keen Software House: QA Department

I had a post about this almost two months ago. I didn't get it resolved unfortunately. You can go take a look at my post, there are tons of comments and other log files there, some with additional details and things to try. It did try a lot of things and maybe one will work for you.

I still think SE has an issue with 4090 cards. A lot of the crashes with similar logs within the last year or two have mentioned they are using a 4090. I hope you have better luck getting this resolved, because I would really like to play the game without crashing.

OP:

When you say "It's saying my GPU is overheating" - what, exactly, is "it?" What file/process is creating the error, and in what exact/specific form & manner is it being displayed or otherwise communicated? (Dialogue with window/process image name, error string pulled from log, SE's own crash report dialogue, one of your hardware monitors or injected/sideloaded overlays, etc)?

Both of you:

Do you have your GPU power profile set to max for SE? And are you limiting framerate through the driver itself (control panel or nvinspector), or through RTSS (or other)? What are you seeing for GPU temps (ideally from Afterburner) both at idle and at the moment your crash occurs? Are your GPU fans set to be always-on even at low temps? Are you running default Tdr values? Do you have your OS graphics preferences set to "High Performance" (which should be your 4090) for SE, and do you have your integrated GPU (if present and/or if you've unlocked it on the KF chip) disabled in the bios? GFE / Shadowplay are disabled already for testing, I presume?

Note:

The error string HRESULT: [0x887A0005] in this case is not significant to the troubleshooting process. That's a generic DX code that (in practice) tends to get thrown whenever a rendering process loses its ability to communicate with the GPU for any reason - which could be (and most commonly is) caused by what's called a "hard" driver crash... which in turn can be caused by almost anything. What is clear from the log posted is that SE itself doesn't "know" why it's crashing; this is a (very) strong indication of a configuration or other lower-level problem rather than a 1st-order or higher-level app problem.