AMD Makes FSR 2.0 ‘FidelityFX Super Resolution’ Technology Official: Offers Similar or Better Image Quality Than Native Resolution With Higher-Performance Across Several AAA Games
But before we spill a bit more detail on FSR 2.0, to set the scene let’s take a look back at FSR 1.0 first. What exactly is AMD FidelityFX™ Super Resolution 1.0? FSR 1.0 is our best-in-class open source spatial upscaling solution for producing high-resolution frames from lower resolution inputs. Designed for high performance and easy integration, it uses a collection of cutting-edge algorithms with a particular emphasis on creating high-quality edges. Over 80 games and engines have already adopted or are adopting FSR 1.0 right now. The cross-platform codebase allows availability on PC across multiple vendor GPUs, as well as consoles and mobile gaming. It has also just become available in the latest Radeon Adrenalin driver as Radeon Super Resolution³, bringing the benefits of FSR 1.0 to games that don’t natively support it.
Spatial upscalers have a lot of advantages, but they can have some limitations too.
FSR 1.0 requires a high-quality anti-aliased source image, which is not always available without making further changes to code and/or the engine. Upscaling quality is unavoidably a function of the source resolution input. So with a low-resolution source, there is just not enough information with a spatial upscaler for thin detail.
This is where FSR 2.0 comes in! Introducing AMD FidelityFX Super Resolution 2.0! We’re very excited to announce our brand-new open source upscaling technique: FidelityFX Super Resolution 2.0! FSR 2.0 is the result of years of research from AMD and is developed from the ground up. It uses cutting-edge temporal algorithms to reconstruct fine geometric and texture detail in the upscaled image, along with high-quality anti-aliasing.
Delivers similar or better than native image quality using temporal data. Includes high-quality anti-aliasing. Higher image quality than FSR 1.0 at all quality presets/resolutions. Does not require dedicated Machine Learning (ML) hardware. Boosts framerates in supported games across a wide range of products and platforms, both AMD and select competitors.
No Machine Learning Hardware Required and Broad Support Like FSR 1.0, FSR 2.0 does not use Machine Learning (ML) in its upscaling algorithm. While ML is one vehicle to solve a number of problems, it is not a requirement to achieve good quality upscaling. Therefore, FSR 2.0 does not require dedicated ML hardware which means more gamers can benefit from it. FSR 2.0 will work across a wide range of products and platforms, and like FSR 1.0, that includes select competitor graphics cards. Take a look below at the 1:1 single frame comparison comparing native 4K with FSR 2.0 Quality Mode. Deathloop FSR 2.0 (Quality):
I want to know more already!
We’ll be sharing more about FSR 2.0 in due course, but don’t miss our GDC talk on March 23 10.30 am PT where engineers Colin Riley and Thomas Arcila do a deep dive explaining exactly how FSR 2.0 works in great detail. You also won’t want to miss AMD’s Lou Kramer and Arkane Lyon’s Gilles Marion talk either: A Guided Tour of Blackreef: Rendering Technologies in Deathloop which starts at 9 am PT just before. Not able to do GDC? The videos for both presentations will be available here on GPUOpen shortly afterward, along with all our other GDC sessions plus some extras, so keep a close eye out! You can also find out a bit more from the official AMD FSR 2.0 announcement as well. DISCLAIMERS AND FOOTNOTES
