Caporegime
FidelityFX Super Resolution, using DirectML.
https://www.hardwaretimes.com/amd-f...eon-boost-coming-in-spring-adrenaline-update/
Ever since the RDNA based graphics cards were launched, AMD has been working on the software side of its GPU Eco-system. The Radeon Software Adrenaline 2020 Edition was one of the largest revamps of the control panel in years. Radeon Boost which is essentially a driver-centric implementation of Variable Rate Shading was one of the key features of that updates. Other notable features were Radeon Chill, Enhanced Sync, Anti-Lag, and Integer Scaling, all in one window.
It looks like Team Red is working on another major update to Radeon software, slated to land this spring. One of the primary additions will be the inclusion of the DLSS-alternative otherwise known as FidelityFX Super Resolution which will leverage the DirectML API to upscale the image with a minimal loss in quality. Unlike NVIDIA’s implementation, this is a software-based technique that only requires the presence of Microsoft’s DirectML API to work.
https://www.hardwaretimes.com/amd-f...eon-boost-coming-in-spring-adrenaline-update/
Ever since the RDNA based graphics cards were launched, AMD has been working on the software side of its GPU Eco-system. The Radeon Software Adrenaline 2020 Edition was one of the largest revamps of the control panel in years. Radeon Boost which is essentially a driver-centric implementation of Variable Rate Shading was one of the key features of that updates. Other notable features were Radeon Chill, Enhanced Sync, Anti-Lag, and Integer Scaling, all in one window.
It looks like Team Red is working on another major update to Radeon software, slated to land this spring. One of the primary additions will be the inclusion of the DLSS-alternative otherwise known as FidelityFX Super Resolution which will leverage the DirectML API to upscale the image with a minimal loss in quality. Unlike NVIDIA’s implementation, this is a software-based technique that only requires the presence of Microsoft’s DirectML API to work.
