Fsr frame gen was just released. Dlss frame gen wasn’t perfect either at release (even now it still has ui issues).
Fsr frame gen for me looks pretty impressive for a rushed release. I’ll need to see how it evolves and if amd can solve the antilag+ latency with frame gen and enhance smoothness.
I also want to see how amd can enhance fsr upscaling image quality, as currently the worse image quality compared to dlss frame gen is because of fsr beeing less good than dlss upscaling.
For consoles, well it’s another reason for devs to create unoptimised games, while giving the 60 fps console players could “finally” experience, and want with the curent gen.
However on another side it’s also a way to get better smoothness (well see), at at negligible (for console players) image quality loss.
Most console players play on a TV, pretty far from it. So quality won’t affect them much.
Guarantee you the FSR-DLSS gap will be filled shortly after AMD has competitive AI coprocessors on their cards. People say a lot about all the training DLSS does on NVIDIAs cloud blah blah but the real reason it’s better is because it runs on hardware that is otherwise idle and so can just do more without eating into latency or performance. It’s the same reason XeSS on Intel GPUs is better than FSR.
What would really be impressive is if AMD can get FSR to leverage the AI cores on all three cards. If the goal of being “open” is trying to nullify NVIDIAs advantage then that would go a long way to killing DLSS as a point of distinction.
Yes and no.
Fsr frame gen was just released. Dlss frame gen wasn’t perfect either at release (even now it still has ui issues).
Fsr frame gen for me looks pretty impressive for a rushed release. I’ll need to see how it evolves and if amd can solve the antilag+ latency with frame gen and enhance smoothness.
I also want to see how amd can enhance fsr upscaling image quality, as currently the worse image quality compared to dlss frame gen is because of fsr beeing less good than dlss upscaling.
Here is the hardware unboxed analysis of fsr 3 : https://youtu.be/jnUCYHvorrk?si=RXUZMBJXLelM-h2O
For consoles, well it’s another reason for devs to create unoptimised games, while giving the 60 fps console players could “finally” experience, and want with the curent gen.
However on another side it’s also a way to get better smoothness (well see), at at negligible (for console players) image quality loss. Most console players play on a TV, pretty far from it. So quality won’t affect them much.
Guarantee you the FSR-DLSS gap will be filled shortly after AMD has competitive AI coprocessors on their cards. People say a lot about all the training DLSS does on NVIDIAs cloud blah blah but the real reason it’s better is because it runs on hardware that is otherwise idle and so can just do more without eating into latency or performance. It’s the same reason XeSS on Intel GPUs is better than FSR.
What would really be impressive is if AMD can get FSR to leverage the AI cores on all three cards. If the goal of being “open” is trying to nullify NVIDIAs advantage then that would go a long way to killing DLSS as a point of distinction.