DLSS and FSR will soon have a new competitor for the attention of gamers and developers. There appears to be a new effort in Redmond from Microsoft spooling up. The Washington-based tech giant listed two separate positions, one for a Senior Software Engineer, and another for a Principal Software Engineer of Graphics. Both of these are now being attributed to an attempt at Microsoft’s own resolution upscaling effort.
The listing explains that the positions are intended to “ensure that all games run beautifully at the highest resolutions and frame rates.” It also mentions that it will involve “working in a friendly, collaborative environment on technology related to graphics such as the DirectX API, graphics driver, Machine Learning, shader compiler, PIX tooling, or silicon feature development.”
The push by DLSS and FSR has been pretty uneven, and it has left space for more games to expand as more options come to market. The idea is cool, but has taken time to see widespread use. The usage of machine learning algorithms to develop the most efficient techniques to handle resolution upscaling in games is really helpful. This will deliver the best possible performance on the high-end displays that are slowly becoming more prevalent. You can see the end-result of these efforts in a trailer for DLSS in Rainbow Six Siege down below.
AMD has also jumped into the ring with their own FidelityFX to handle upscaling on team red GPUs. This technology is billed as AMD FidelityFX Super Resolution (FSR) and has shaken up the game a bit. AMD’s FSR and Nvidia’s DLSS have been jockeying for position for some time now. Most games usually use one or the other at first, and may patch in the other if there’s demand.
This has sort of left the door open for future efforts, and it appears MS is the first to jump at the chance. The biggest hurdle so far has been performance hits. Microsoft has a major chance to bring a new alternative that reduces the load rendering 8K images puts on machines. If the company can push the boundaries of machine learning and make a true improvement, they’re primed for success.
Machine learning-based resolution upscaling still has plenty of room to grow. As the technology improves, so too will the output in gaming and production PCs. As more and more displays move into 4K, 8K resolutions and beyond, demand for the best quality images linearly increases too. What Microsoft manages to do with the opportunity remains to be seen.
