No, they could not. AMD cards don’t have any of the hardware to execute the same or similar operations. Executing the same code without tensor cores to accelerate them would tank performance, which is the entire reason you get less performance gain for far worse image quality with FSR in the first place.
The literal only thing AMD’s hardware is competitive at is raw traditional raster performance, because Nvidia has better designs that leverage hardware features to accelerate portions of the ray tracing and upscaling workloads much more efficiently.
AMD is trying to copy hardware features with software, and it’s not comparable.
You are not logged in. However you can subscribe from another Fediverse account, for example Lemmy or Mastodon. To do this, paste the following into the search field of your instance: [email protected]
No game suggestions, friend requests, surveys, or begging.
No Let’s Plays, streams, highlight reels/montages, random videos or shorts.
No off-topic posts/comments, within reason.
Use the original source, no clickbait titles, no duplicates.
(Submissions should be from the original source if possible, unless from paywalled or non-english sources.
If the title is clickbait or lacks context you may lightly edit the title.)
Since it is open source they could push their optimizations upstream and having both companies working on a uniform solution is better for everyone
(I guess not better for Nvidia’s monopoly since they have worse cards)
No, they could not. AMD cards don’t have any of the hardware to execute the same or similar operations. Executing the same code without tensor cores to accelerate them would tank performance, which is the entire reason you get less performance gain for far worse image quality with FSR in the first place.
The literal only thing AMD’s hardware is competitive at is raw traditional raster performance, because Nvidia has better designs that leverage hardware features to accelerate portions of the ray tracing and upscaling workloads much more efficiently.
AMD is trying to copy hardware features with software, and it’s not comparable.