AMD says FSR 2.0 will work on Xbox and these Nvidia graphics cards

Last week, AMD promised new technology that should allow you to blast your games to higher resolutions or boost their frame rates without requiring Nvidia’s fancy GPU machine-learning hardware, like DLSS. rented from Nvidia. Now, at GDC 2022, he’s revealing how the new FidelityFX Super Resolution 2.0 actually works – and that it’s also coming to Microsoft’s Xbox games consoles.

While AMD says it can’t actually say when Xbox game developers might take advantage of FSR 2.0, it says it “will also be fully supported on Xbox and will be available in the Xbox GDK for developers recordings use it in their games”.

And it also gives the community a list of the two AMD and Nvidia GPUs you can expect FSR 2.0 to run on – if you have an Nvidia GeForce RTX 1070 or higher, the company suggests, you might be able to enjoy FSR 2.0 at least on a 1080p monitor, similarly way you with an AMD Radeon RX 590, RX 6500 XT or better.

What we’ve been asking ourselves since day one is: what’s the problem? How can AMD almost double the frame rate of a demanding game like Loop of death, at 4K-equivalent resolution with the kind of image quality it showed us last week, all without dedicated machine learning cores like Nvidia’s DLSS?

The answer is complicated, but a short version is that it can not unless you have a relatively powerful graphics card to start with.

Although the FSR 2.0 algorithm is remarkably fast – under 1.5ms in all of AMD’s examples – it still takes time to execute, and it takes longer on lower-end GPUs where AMD admits freely that some of its optimizations don’t work as well. good.

In that sub-1.5ms period, FSR 2.0 does all sorts of things, though – AMD says it replaces a full temporal anti-aliasing pass (getting rid of a bunch of jagged edges from your game) by calculating motion vectors; reproject frames to cancel jitter; create a “deocclusion mask” that compares one frame to the next to see what moved and what didn’t in order to cancel out ghosting effects; lock thin elements in place like barely visible edges of stairs and thin wires; prevent colors from drifting; and whole-image sharpening, among other techniques.

Unlike FSR 1.0, this requires some work from the game developer, so it’s not something all game developers will benefit from – but AMD confirms that both Death Loop and Speak (a technology storefront that will also use Microsoft’s DirectStorage) will benefit.

AMD says games that already support Nvidia’s DLSS should be easy to set up, with just a few days of work to integrate, and games running on Unreal Engine 4 and Unreal Engine 5 will have a plugin to make it work. Games that already use temporal antialiasing also have a development advantage. But if a developer hasn’t built their game with some of these things in mind, AMD says it could take four weeks or more of work.

We’re still waiting to try FSR 2.0 for ourselves to see what it actually looks like in motion, but if the developers take advantage of it, the quality seems to be much better than FSR 1.0.

Comments are closed.