"The AI model powering DLSS 5 is a single unified model. Same model for every game. It's not trained per-title, per-face, or per-object type."
Why are Nvidia claiming developers have full control over the artistic intent of DLSS 5 output when it uses a unified model that isn't trained per game?
Surely this would mean DLSS 5 output would converge to the same type of visuals across games, removing artistic intent/touch. Just as we see from current gen AI tools
Giving developers true, full artistic control would surely mean having specific models for each game trained only on that studio's assets?
I think a lot of people are starting to realize “better graphics” are separate from artistic vision. Games use so many tools to achieve their intended look- it’s not a one size fits all solution and never will be. It’s why those unreal engine 5 Mario remakes look like total shit
Nvidia's DLSS 5.
This is the ugliest, trashiest AI-slop tech from Nvidia by a long shot.
It even changes some of the facial features and makes them look like the images created by Twitter grifters who post, “I made Grace hotter, Western devs can't stop me.”
Ewwwwww. #Nvidia