AMD recently published a new version of its FidelitySDK with FSR 4 upscaling and FSR 3.1.5 frame generation support. However, during the process, the GPU manufacturer accidentally published the full FSR 4 source code on GitHub (seemingly unintentionally). Before AMD took it down, some media outlets and X users managed to record screenshots of the files, including Videocardz.
The files reveal that AMD was working on a second version of FSR 4, capable of operating through the int8 numerical format. Videocardz' screenshot shows duplicate FSR 4 files, with a set focused on FP8 and the other int8 support. The only reason you'd want to make another version of your upscaler that runs on a different numerical format is to support more GPUs.
