Hey,
I have some older games that I want to record.
I bumped supersampling via nvidia inspector to improve quality.
Of course by forcing supersampling I'm causing the application to render in much higher resolution.
So capture/encode is being also performed in such huge resolution as well.
In order to fix that, simply I need to resize captured video stream.
Programs like Dxtory fix that issue automaticaly using simple option.
It's called Force Down Sampling
When you set that, Dxtory forcibly down scales it's video capture based on your settings, preventing errors associated with a bug that sometimes happens with downscaled video capture when capturing directly from the GPU's video buffer
Now, when you do 4xSGSSAA and start to record like that, Your fps will suck.
Can you implement that function like dxtory does?
Sparse Grid Supersampling realy improve quality, but you can't record like that, with Dxtory it's possible even on slow GPU.
Thank you.