I have noticed that the magic of DLSS doesn’t work wuite well with camera’s DOF effects. I wrote Nvidia and they tell me that the problem is that DLSS happens in the same stage as TAAU in the pipeline (which happens one step after DOF: Screen Percentage with Temporal Upsample | Unreal Engine Documentation ), so Depth of field becomes quite weird… and it’s a quite important effect!
I have been trying to not use DOF and make my of Post Process material to simulate the DOF in a later stage, but the result is not near to the quality of actual DOF.
Please, do you have any iade about a workaround for this?
About the last part you are right, but about the rest… I think DLSS looks even better than any other AA. And of course it performs much more better, being a game-changer; I don’t know how you can say the opposite.
I am in the same boat, downloaded the latest DLSS and the depth of field seems to be gone when switching it on. I wonder if I am missing some settings with console variables?