Chromatic

To be honest, this isn't the first time I've been having these issues. I guess you have already noticed, but I had the same problem in November too and we couldn't find a conclusive answer back then either. I already have high performance settings enabled on Windows each time I run xlodgen/texgen/dyndolod. The monitor does shut off by its own still due to the settings used, but I noticed the same issue happening even if I didn't let it shut off anyway. I am looking for it, but haven't noticed such an entry in the Event Log. However: After doing some research in this thread, I decided to install an older driver version like you also suggested. I went back all the way to 22.5.1 which was mentioned here in the past and surprisingly, everything worked fine and lod generation was successful even without a minimal driver install or shutting off AMD related processes. In fact, I ran DynDOLOD two different times with successful lod generation for each one (let me know if you still want me to post the logs anyway). Does this mean that newer AMD drivers are a no-go?

Not possible. 24.3.1 is actually the latest release for my GPU (RX Vega 56) since AMD has put their older cards on a different driver release schedule, as stated here: I can't say I noticed any excessive video memory usage during the entire time DynDOLOD was running, the maximum amount I saw was ~1GB. Latest logs: https://ufile.io/h074r74a No bugreport was generated this time either, the debug log is empty but I included it just in case.

New logs: https://ufile.io/atzdmuam No bugreport.txt was generated this time.

No, only selecting SIworld didn't have any problems and lod generation was successful. RenderSingle=1 didn't make a difference, this time lod generation was stuck at the same spot but no error message appeared. After letting DynDOLOD run for many hours with no progress, I had to close it manually and terminate the process in the task manager. No logs were generated this time.

Got this error during the last DynDOLOD run: I am using the latest (24.3.1) AMD WHQL drivers. Logs: https://ufile.io/0oyqmll5

I've monitored memory usage a bit on my latest run and when the error popped up system memory was ~8,6gb used out of 16 and video memory was ~2,1gb out of 8. As shown in the picture below: Latest logs but without a bugreport file this time, it doesn't seem like it was generated: https://ufile.io/twtc9q4u

No overclocking is done on this system, and the only setting enabled in the BIOS is XMP. I haven't run long stress tests yet but everything seems stable otherwise. I also pretty much closed all other programs during the latest DynDOLOD run except from Vortex and DynDOLOD itself but I still got an error. As for memory usage, I haven't checked it this time but system memory is 16GB and the GPU has 8GB of video memory so I assume it should be plenty with everything else closed. Let me know if you need me to test this further though, I can do another run and closely monitor memory usage and maybe even do some long hardware stress tests if needed. Latest logs: https://ufile.io/dj87ufng

Latest logs: https://ufile.io/iwpf7kqf

New logs as requested: https://ufile.io/7jlxnjar

Using GLDebug=1 (alongside LockTexconv=1) hasn't made a difference, the same error happens again. Logs: https://ufile.io/0lzdeav6

LockTexconv=1 didn't make a difference, DynDOLOD hanged at the same spot like before (I had to properly close the process from the task manager). RenderSingle=1 did help in my latest run since DynDOLOD didn't get stuck this time but now I got this error message on the same spot: New logs: https://ufile.io/ti1yj9xk

DynDOLOD keeps hanging after some time, had to enable real time logging too since no logs were being generated otherwise. Logs are provided here: https://ufile.io/artotucx. I am using a Radeon Vega 56 on the latest GPU drivers (Adrenalin 23.11.1 (WHQL Recommended) with a minimal install.