(6.0 beta 4) Display disconnecting and fans maxxing out - Resolved itself on its own.

[Zloth2]

I'm getting this issue in Beta 6, but I suspect this is an issue with my machine, so I'm posting here first. At a certain point, about 4 minutes after loading a save, my display will get disconnected and the fans in the PC will suddenly crank up to max. They stay that way until I push the power button to cause it to shut down. (I've only given it a couple of minutes to come back.)

I've got a save I made to do some experimentation on large freighter speeds vs. medium freighters. I load the save up (based on a custom start that's only a minute or two old), tell the medium freighter to head to dock at another station, then I fire up the engines on a Helios-E and try to go to the same station. I did this half a dozen times with no issue. Then, in Hewa's Twin II, the monitor switched off and the fans cranked up. The music and sound effects stopped, so I presume the game crashed.

I rebooted, tried again, and this time it crashed soon after getting into Company Regard. I suspected a heat problem, so I gave it 15 minutes, loaded the game again, tried again from the same save, and crashed again at the same spot in Company Regard. I tried yet again, this time with NVIDIA's driver tool to display the temperature. GPU was at 80C just before I went through the gate to Company Regard. A few seconds later, the game crashed in the same spot as before. It seems more like something is coming into view that the game can't render. But I've been going through, doing roughly the same path from the same save several times with no issue.

Here's a video of the run I did checking temperatures:
https://youtu.be/SPQrmZMFyYM

I verified my game files but nothing new got downloaded. I tried running without recording video and still crashed at the same spot.

Event Viewer shows an NVIDIA OpenGL error warning just before the crash:

Code: Select all

The description for Event ID 2 from source NVIDIA OpenGL Driver 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: 

The NVIDIA OpenGL driver has not been able to initialize
a connection with the GPU. This might be due to out of memory error,
an exhaustion of system resources or too many graphical applications running.
The application will not render correctly.
(pid=45$

The message resource is present but the message was not found in the message table

That's followed by this error from the kernel:

Code: Select all

Faulting application name: dwm.exe, version: 10.0.19041.746, time stamp: 0x6be51595
Faulting module name: KERNELBASE.dll, version: 10.0.19041.2546, time stamp: 0xe8e9ac9b
Exception code: 0xe0464645
Fault offset: 0x000000000010fd12
Faulting process id: 0x2828
Faulting application start time: 0x01d950ad9cb7617e
Faulting application path: C:\Windows\system32\dwm.exe
Faulting module path: C:\Windows\System32\KERNELBASE.dll
Report Id: 59de296c-8e0b-4ae1-acae-423c56ae5c55
Faulting package full name: 
Faulting package-relative application ID: 

I'm not sure what to do with this one. Crashing to the desktop or locking up is one thing, but not sending data to the monitor and blasting the fans!? What does that!?

[Imperial Good]

I rebooted, tried again, and this time it crashed soon after getting into Company Regard. I suspected a heat problem, so I gave it 15 minutes, loaded the game again, tried again from the same save, and crashed again at the same spot in Company Regard. I tried yet again, this time with NVIDIA's driver tool to display the temperature. GPU was at 80C just before I went through the gate to Company Regard. A few seconds later, the game crashed in the same spot as before. It seems more like something is coming into view that the game can't render. But I've been going through, doing roughly the same path from the same save several times with no issue.

80 degrees Celsius in X4 is unusual unless you really have the setting cranked with a powerful CPU. I recommend checking if the GPU active cooling fans are still spinning and the air flow path to/from the GPU has not become clogged with dust or pet fur. Some cases are just horrendous for air flow, and if you own such a case you might need to consider removing some panels to see if that makes a difference. You can also try turning down the visual settings of X4 and using a frame rate cap to limit how hard the GPU is working, hopefully lowering its temperature. Although reducing GPU usage is not really a solution (you want the performance you pay for), it can help confirm if thermals are a potential cause.

I verified my game files but nothing new got downloaded. I tried running without recording video and still crashed at the same spot.

There is potentially an inverse relationship between how complex a sector is and how hard the GPU works. The more complex a sector is, the more likely you will be CPU bottlenecked and so the less hard the GPU works due to the CPU not being able to feed it render instructions fast enough. Like wise, the simpler the sector is, the less likely you will be CPU bottlenecked so the harder the GPU works due to the CPU being able to instruct it to render fames more frequently. As such it is entirely possible that some sectors may be more prone to triggering hardware issues than others. It is also possible that the GPU workload varies enough with some sectors due to the different unique objects and backgrounds present in them to make them more likely to expose instability than others.

Event Viewer shows an NVIDIA OpenGL error warning just before the crash:

X4 uses Vulkan. As such an OpenGL error would likely be another application crashing.

but not sending data to the monitor and blasting the fans!? What does that!?

Unfortunately locking up and GPU fans blasting to 100% is often a sign of a catastrophic GPU failure with the GPU going into a "safe mode" with fans at full speed. Given that your GPU is reporting 80 degrees then it could be related to thermals, such as the voltage regulators feeding the GPU die with power overheating and so either not regulating power well enough, or in more expensive GPUs they might shut down entirely to prevent damage. I recommend reducing the GPU operating temperature to see if that helps.

It is entirely possible the GPU is starting to fail and so is becoming less stable at higher temperatures. The more frequently and easier it becomes to trigger such system lockups, the more likely this is to be the case. Running the GPU cooler may be able to delay this, but you might not be able to use the full designed performance capabilities of the GPU anymore and it may eventually fail anyway.

[Alan Phipps]

Perhaps upload and link a DxDiag and your VulkanInfo (link) in case we can spot something.

[Zloth2]

I'm using a 3080 with fairly high graphics settings. It should be able to handle 93C, 80C isn't a big deal. I don't really see how a heat issue could result in crashing at such a precise time repeatedly, either.

Whatever the case, the issue seems to be gone now. I tried simply running the game from the game start in Grand Exchange that gives you the HQ right off and left the game there for half an hour with no issue. Then I buzzed around the main highway loop for a while. Then I went out to Turquoise Sea. No issues. I tried my problem save again and did my standard run. It worked just fine again, and did so twice.

Whatever it was, it seems to have vanished just as strangely as it appeared. <shrug> Thanks for the help!