- Add FP8 NR models for RTX 40 and 50 series GPUs on the D3D12 path: "FP8 2xPass-in-one" and "FP8 3xPass-in-one", and "FP8 2xPass-in-one v2" and "FP8 3xPass-in-one v2" with other weights. They give the looks of NVFP4 2xPass-in-one and 3xPass-in-one. Like the NVFP4 models they carry all their own weights and take none from the NR runtime, so they work with any NR runtime that offers the original model (preset 1), including 310.8.0 and 310.9.16.0, and give the same picture on each.
- The FP8 model files are in OptiScaler/nvfp4/fp8. A model's weights are loaded only when a pass uses it.
- The FP8 models have not yet been tested on an RTX 40 series GPU.
- Suggested settings for the 2xPass-in-one and 3xPass-in-one models (NVFP4 and FP8): Cinematic, with Local structure lowered to 0.5 and Local tone lowered to 0.8.
- Add NR presets: save the current NR settings under a name, then load, overwrite, rename or delete them ("NR presets and calibration" in the menu; files in the nr-presets folder beside OptiScaler).
- Add a calibrator: on a held frame it renders a saved preset, then adjusts the edit pass's Model intensity, Local structure and Local tone and the Detail and Colour strength until NR's picture is as close to the preset's as it gets. Direct3D 12 only.
- "Enable skin settings optimisation" lets the calibrator adjust the skin and environment sliders and Highlight guard as well.