Hi anpeaco....I had intended to reply last night but it just got too late.
I'm not familiar with DCS World but upon looking it up I find that it is a combat flight simulator game. So I put it in the same category as MS Flight Sim and the other simulator games that seem to have a higher incidence of the problem occurring (Dirt Rally, iRacing, and other racing sims). These simulator games are fast moving and have a LOT going on with respect to environment maps, textures, shadows, and modeling than your typical single shooter games. This means they can require a LOT more VRAM (GPU memory). You have a 3080ti which has, I believe, 12 GB of VRAM available. You would think that would be plenty but maybe not depending on what else you might have running on your system when playing the game. For instance, a 4K HDR monitor requires a whole lot of VRAM.
Here is a link to a thread on the DCS Forums with players discussing the problem of the 3080ti VRAM being over saturated:
3080Ti possibly full vram ? - Virtual Reality - ED Forums (dcs.world)
A lot of possibilities mentioned in that thread and a lot of discussion about decreasing textures and shadows insure VRAM isn't saturated.
So that's an area to start for sure. However, from your post it appears that something triggered the issue. Meaning all was fine and then suddenly one day it wasn't. So the obvious question is "what changed?" Was there any updates?...Nvidia graphics? Windows? Game update?....etc.?
If not as far as you can tell then a couple other possibilities is that it could be an overheating issue. If it starts slowly and gets worse over time it might be your GPU VRAM is becoming too warm and the GPU begins to throttle itself back which then could be causing artifacts. Overclocking of the GPU could also be a culprit if you've done that. The worst case scenario is that your VRAM is starting to fail. Unfortunately it happens.
Here is a link to an article on how to increase your VRAM allocation:
How to Increase Dedicated Video RAM (VRAM) in Windows 10 and 11 (makeuseof.com)
The fact that you are still seeing the problem irrespective of the number of encoder slices you set in the registry (per Pokebert) hints that the problem you are having isn't quite the same as everyone else because changing the 'numslices' attribute has, as far as the reporting to this thread, been a universal fix. So it is likely it is some other issue.
You could also try the following to reduce the workload on your GPU's VRAM:
In game: reduce anything that requires more complex textures and shadows
In the Oculus Debug Tool (be sure to 'run as administrator before changing ODT settings): set pixels per display override to '0';
FOV tangent multiplier: set horizontal to ..9 and vertical to .8; This will reduce the number of pixels that need to be rendered on the outside of the display by 30%. Note that this needs to be re-set every time before you fire up Oculus.
Set the Encode resolution width to 3664 (default for the Quest 2...not sure what it would be for the Pro but if different just set it to '0' to invoke the default.
Set the Encode bitrate to '0' (default...150 mbps (I believe it still is for the Quest 2)
The above settings should help to reduce the GPU's rendering load
Nvidia control panel: Set Texture filtering - Quality to "Quality" (instead of High Quality)
And finally the Oculus PC app....for whatever framerate you have set (90, 80, or 72) set the resolution slider to default ('0'). If you are running 120 try 90 (or even 80) for testing to see if you can get the problem resolved. If it goes away you can then return to the next level up though I suspect 120 would be problematic.
I would also recommend you have the latest Nvidia drivers installed and use Display Driver Unistaller (DDU) to ensure a clean install of updated Nvidia driver (its a free app).
And as I mentioned above I would definitely check for a potential overheating issue.
Oh...I think I'd try leaving the numslices to '4' for testing these other settings.
Good luck and please report back your progress.
