AMD Unveils New FidelityFX Super Resolution 3.0 Details at GDC 2023

Perspective: During last year’s Ryzen 7000 series launch, AMD announced updates to its FidelityFX Super Resolution upscaling solution. Team Red doubled down on their promise at this year’s Game Developers Conference with more details on the benefits, challenges, and technologies involved in the development of FSR 3.0.
During development of the FidelityFX™ SDK presentationAMD’s Jason Lacroix gave attendees a more in-depth preview of the scaling tool’s features and capabilities. The next iteration of FidelityFX Super Resolution (FSR 3.0) will give users a 2x increase in frame performance. compared to the existing FSR 2.0 solution. Unfortunately, AMD did not give a release date for the technology.
Team Red says it can already meet the requirements using currently available solutions and a concept known as frame interpolation. Frame interpolation is used in upsampling technologies to synthesize pixels and frames between generated frames or images. This technique is critical for increasing performance at higher frames per second without sacrificing overall image quality.
AMD engineers are confident that, if implemented successfully, they will be able to more than double the frame rate while achieving a one-to-one sampling to interpolated pixel ratio. A successful implementation of the solution also ensures that there are no feedback loops in the generation process, so any artifacts or other inaccuracies will only appear once and will not be taken into account in future frames.
The path to achieving the desired goal of doubling workforce productivity is not an easy task. According to the presentation, before AMD can successfully implement FSR 3.0, there are several challenges to be overcome. The Red team did not shy away from discussing these issues, highlighting potential issues related to color fixing methods, non-linear motion interpolation, and increased interpolation requirements for post-processing and user interface elements.
AMD Scaling Solution is an open source software technology that can be used across multiple generations and brands of graphics hardware. The solution has an uphill battle against Nvidia’s Deep Learning Super Sampling (DLSS) hardware technology. DLSS 3.0which has been available for adoption and use since the release of the RTX 4000 series graphics cards last October, uses proprietary Tensor hardware cores to support its scaling solutions.
FidelityFX Super Resolution is an open source software technology that can be used by developers across multiple generations and brands of graphics hardware. The solution has an uphill battle with Nvidia’s hardware DLSS. Deep Learning Super Sampling 3.0, available since the release of the RTX 4000 series graphics cards in October, relies on proprietary Tensor hardware cores to support scaling.
While AMD’s solution is more than a generation behind Nvidia’s latest, the commitment to an open source solution could be a hallmark for developers looking to adopt a scaling methodology. The upcoming FSR 3.0 technology aims to provide developers and users with excellent overall performance, reduced latency, and a more streamlined integration path for new games and games already using FSR 2.0. In contrast, Nvidia’s DLSS 3.0 is exclusive to the latest line of 4000 series GPUs.
If AMD FSR 3.0 continues the company’s trend to support legacy and even competing hardware while delivering on the promise of improved visual quality and frame delivery, then Team Red will be able to close the growing gap in the battle to adopt scaling technologies.
Source link