I searched the forums but somehow could not find a good answer.
I upgraded my GPU from a 3080Ti to the 4090 FE as I wanted to dive into the world of VR.
Is there a way to make the looks better, especially the terrain?
Generally the settings are all set to ultra already and I get a solid smooth VR experience with 75 FPS. (I fly the Cessna 152 only though and not above cities but only above landscape).
How can I make the video output more clear? When watching YouTube videos of some content creators their footage looks way more crisp than what I see through the HP Reverb G2.
My CPU is a 5800X3D if that matters.
Would increasing the render scale improve the visuals?
You can experiment with higher setting in OpenXR Tools for WMR. You can go higher here, FPS will drop, quality will increase. You can also enable DLSS and go even higher with OpenXR Tools for WMR.
If you have Reverb G2 - it has small sweet spot. You can experiment with putting the lenses closer to your eyes by using shallower gasket (you can start experimenting without a gasket, and if this will work for you, you can buy a shallower one).
I have a 5800x3D and a 4090, like you. You can get the sweet spot much, much sharper by simply setting OXR to 160% and using DLSS Quality mode. The difference in the scenery is incredible. I use motion reprojection, and with those settings it will lock at 30fps no problem with our CPU and a 4090.
If you are using the Toolkit you can add a bit of fps headroom by enabling foveated rendering preset of quality wide. You can also set OXR back to 100% and use resolution override of about 4,200 x 4,200 or a bit more in the Toolkit. Lowering the FOV setting to about 94% will also sharpen the sweet spot even more.
Thanks a lot for those inputs! I will try step by step making adjustments. Setting OXR to 160% will be my first step in order to make some improvements.
Does it then really still need DLSS Quality mode? Like, do they have to go together?
With my settings, yes, DLSS Quality is needed to keep a decent FPS. TAA looks better, but the framerate hit is substantial. Try yourself and see what works best for you.
5800X3D and 4090 here too. I really don’t like NVidia DLSS DLAA or DLSS quality or balanced. The Cockpit looks so much better in TAA and 100% OXRWMR. I feel the issues lie in mixing tasks between 3-4 apps. Windows own OpenXR for WMR, OXToolkit and MSFS settings AND Nvidia control panel.
Who does what and who does it better is key and enabling one and not the other apps to do that same thing.
Anisotropic filtering can be set X4,X8, X16 in the Nvidia control panel or in the MSFS VR panel
Render Scale can be set in OXR for WMR, Toolkit or MSFS.
Sharpening can be done in all 4 apps. and I believe you can sharpen the sharpening and mess everything up.
There’s no clear set of instructions on which to choose, how to avoid conflicts or overlap, or over-riding behavior.
CAS, AMD Contrast Adaptive sharpening, in toolkit vs AMD FidelityFX sharpening in MSFS… are they the same thing, which does a better job? which uses less resources? Do you set 70-90% CAS in Toolkit and MSFS AMD FidelityFX sharpening to Zero?
Which application handles Render scale better, and how does it relate to world scale in MSFS?
It’s a bunch of nebulous terms that don’t give clues to their interplay or overlap, or interference.
MSFS 2020 VR interface seems to include every feature except resolution override or render scale.
There should be a wiki of sorts to split the job on GPU-Apps for VR and MSFS settings.
I can tell you that after dropping $2K on 5800X3D and 4090… It bugs the heck out of me to see my wing edges jiggle and the top of a building or mountain wobble. Yes, I can fly at 350kts and look straight down at the ground, 250FT above the ground, and it’s smooth… I don’t mind 30FPS MR. Which App do you choose to control MR? Which is better? OXR WMR, Toolkit or MSFS? all 3 can “do it”.
There has to be a way to prioritize which device/app/tool does which job to maximize FPS/VRAM/CPU/GPU and VR headset performance.
I’m about to try locking frames at 30 and then 45 using stock nvcp, what is the advantage of using back edge sync in rivatuner, I saw you mentioned that in another post
Main reason I like using RTSS is because it actually works in real time, not like NVCP settings where you need to restart for it to work.
Back edge sync is just the recommended setting for the frame rate limiter getting the smoothest motion, no idea what exactly it does technically. I’m not sure whether it’s really any better than front edge sync!
)
