This is the core of Microsoft’ s announcements on the use of AI around its Windows. This new feature integrated into the flagship OS should considerably improve game rendering while simplifying life for developers. The technology is similar to existing AI upscaling technologies, such as Nvidia’s Deep Learning Super Sampling (DLSS) and AMD’s FidelityFX Super Resolution (FSR). However, with its solution, Microsoft wants to offer an agnostic approach so as not to depend on specific GPU components for AI processing. This is obviously something that, on paper, will appeal to developers.
Super Resolution: a solution to simplify life for developers
However, Microsoft does not dismiss the idea of using hardware to boost the efficiency of its Super Resolution. As with the rest of its AI strategy at the heart of Windows, the software giant suggests that computers equipped with integrated Neural Processing Units (NPUs), such as certain Intel and AMD models, could clearly amplify the effectiveness of this feature. As far as GPUs are concerned, DirectSR is not intended to replace NVIDIA’s DLSS, AMD’s FSR or Intel’s XeSS, but to offer a single code capable of activating a variety of solutions.
At GDC, held from March 18 to 22 in San Francisco, Microsoft obviously emphasized the simplification that this new Windows :
Super resolution is a cutting-edge technique used to increase the resolution and visual quality of games […] DirectSR is the missing link that developers have been waiting for when approaching SR integration, offering a smoother, more efficient experience that adapts to all hardware. (Josué Tucker, Microsoft).
See you after the summer to enjoy Super Resolution
In any case, Microsoft will be rolling out its solution to improve the definition of video game graphics. Expected in the fourth quarter with the release of the 24H2 update, this function, dubbed “Automatic Super Resolution”, will allow you to upscale the definition of your screen when you play, thanks to the capabilities of artificial intelligence. The visual impact of this solution will strongly depend on your on-board hardware.
