AMD has made the source code of FSR 2.0 (FidelityFX Super Resolution) public, which means any developer can now use frame rate boosting technology in their games.
AMD’s FSR 2.0 is the company’s rival to the Nvidia DLSS, and that’s clearly good news in terms of getting more games up and running with the technology. The small caveat here is that Team Red pushed all the materials (opens in new tab) needed to work with DX12 and Vulkan games, but not DX11 (if you want to go this route you will need to contact AMD directly as Tom’s Hardware (opens in new tab) reports).
Furthermore, support for Unreal Engine 4.26/4.27 (and UE5, for that matter) will arrive soon in the form of a plug-in, promises AMD.
AMD also gave us an idea of how long it might take developers to incorporate FSR 2.0 into their games, that time varies – obviously – but if a title already supports DLSS 2.0, it might only take a few days to implement. Likewise, it will be a quick process for those using the Unreal Engine plugin, as you might expect.
For some games, bringing FSR 2.0 support can be a considerably longer (and windier) path, with AMD putting the longest timeframe in their estimates at over four weeks.
Analysis: The proof is in the upscaled pudding
Interestingly, along with this source code release, AMD took the opportunity to hurl some criticism at Nvidia (and Intel XeSS, which, like DLSS, is AI-powered).
As you may know, with version 2.0, FSR is now quite similar to DLSS, mainly because AMD made the switch to using temporal upscaling, which is how Nvidia weaves its framerate-boosting magic.
FSR 1.0 employed spatial upscaling and there is a big difference in switching to temporal with FSR 2.0, as the former only accesses data pertaining to the current frame in the game – whereas with spatial, previous frames are activated. And that results in better quality results when it comes to the upscaled image (upscaling is how these technologies improve frame rates – in simple terms, letting the GPU run at a lower resolution and then upscaling it to a resolution of target instead of running on that native res).
Now Nvidia will point out that its machine learning (AI) capabilities are in the mix along with temporal upscaling for even better results, but AMD claims in its GPUOpen article that the introduction of AI is simply not necessary to get a good result. high quality image. In fact, AMD argues that machine learning is often only employed on blending past frames to get the enhanced image, rather than actually recognizing objects in the real scene, and so Team Red is basically downplaying the importance of adding the power. of AI.
We certainly can’t take this argument at face value, and Nvidia claims that its AI routines are doing considerably heavier work in terms of producing an improved final image.
Whatever the case – and what’s going on under the hood with this technology is a highly complex subject – it’s quite revealing to see AMD go on the offensive like this. Also, when you look at the results of FSR 2.0 compared to DLSS 2.0, it’s a very close issue of image quality. The proof is in the pudding, as they say, and the upscaling built into supported games so far shows FSR 2.0 giving Nvidia a good run for its money (certainly going for Deathloop and God of War as tested by Tom’s Hardware).
Note that both FSR 1.0 and 2.0 can be used together in a game if the developer wants, giving the player the choice of either one – and this is important if the player in question doesn’t have a good enough GPU for FSR 2.0 (which has more stringent hardware requirements). This comes with the obvious caveat that the quality of FSR 1.0 results is substandard, but it’s still nice to have the choice. Hitman 3 is an example game that is getting support for FSR 1.0 and 2.0.
AMD’s openness towards FSR and assurance that it covers a wider range of bases and different graphics cards – including Nvidia models – is one of the biggest strengths of its frame rate boost solution compared to DLSS, that specifically requires an Nvidia RTX graphics card.