News AMD FSR 3 With Fluid Frames Promises 2x FPS Uplift Over FSR 2

[Admin]

AMD just announced its next-generation version of FidelityFX Super Resolution known as FSR 3. AMD neglected to share many details about the new tech, but it gave away that FSR 3 will be using frame generation technology like Nvidia's DLSS 3.

AMD FSR 3 With Fluid Frames Promises 2x FPS Uplift Over FSR 2 : Read more

 

[cryoburner]

So AMD might need a hardware solution for FSR 3 to work.

It seems unlikely that you would need specialized hardware for frame generation to work, though having access to it might improve performance at a given quality level. Even without dedicated hardware, as long as the frame generation calculations take significantly less time to perform than rendering a new frame, the feature could be beneficial.

It sounds a lot like the frame-doubling used to reduce performance drops on VR headsets, like "Asynchronous Spacewarp" for Oculus, a feature that they added through a software update six years ago. That just looks at the previous frame's depth and motion vector buffers and uses that information to deform the frame to simulate a new one, shifting objects to their new, estimated positions. There can be artifacts in some scenarios though, since missing regions behind moving objects need to be filled in, and I wouldn't expect either Nvidia or AMD's solutions to look perfect either, but it could be an effective way to make low frame rates feel a lot smoother.

Maybe Nvidia thinks they can sell more graphics cards by artificially restricting the feature to 40-series hardware, but I see little reason why such a feature couldn't be enabled on all cards.

 

[bit_user]

It seems unlikely that you would need specialized hardware for frame generation to work, though having access to it might improve performance at a given quality level.

IMO, that depends a lot on how intrusive they want to make it. If you can make it at least as intrusive as TAA, then the computational burden should be relatively light.

Even without dedicated hardware, as long as the frame generation calculations take significantly less time to perform than rendering a new frame, the feature could be beneficial.

Yeah, but it would be even better if it didn't eat into the compute budget for rendering, hardly at all. Because, as good as it might be, actually rendering new frames should remain the preferred option. So, if it's implemented as a non-intrusive feature, then having some hardware motion extrapolation engine would be very beneficial.

It sounds a lot like the frame-doubling used to reduce performance drops on VR headsets, like "Asynchronous Spacewarp" for Oculus, a feature that they added through a software update six years ago. That just looks at the previous frame's depth and motion vector buffers and uses that information to deform the frame to simulate a new one, shifting objects to their new, estimated positions. There can be artifacts in some scenarios though, since missing regions behind moving objects need to be filled in, and I wouldn't expect either Nvidia or AMD's solutions to look perfect either, but it could be an effective way to make low frame rates feel a lot smoother.

A cool thing you could do, if you make the feature sufficiently intrusive, is to have the game engine fill the more significant holes by rendering them. As I say that, I'm reminded somewhat of Microsoft's ill-fated Talisman project.

https://en.wikipedia.org/wiki/Microsoft_Talisman

 

[rexsgk]

This Article forgot to mention one part of RDNA 3 Hardware and That is, their Hardware "AI Accelerators", AMD mentioned that their FSR 2 doesn't support Machine Hardware AI Accelerators, but tight lip on FSR 3 details but yet their card performance boosted the frames from 60fps to 112fps on FSR 3. Just want to throw it out there.