AMD FSR Redstone Will Support NVIDIA GeForce and Intel Arc GPUs: Neural Rendering Core Not Just for Radeon

AMD’s latest advancements in machine learning (ML) have introduced a neural rendering technology that may extend its compatibility beyond just Radeon GPUs. This innovation highlights the company’s strategic adaptation of ML operations, promising to enhance the gaming experience across various platforms.

AMD’s FSR Redstone: Broad Compatibility for Enhanced Visuals

Announced at Computex 2025, AMD’s FSR Redstone is a groundbreaking Machine Learning suite tailored for developers. This suite enables the integration of neural rendering technologies into video games, significantly improving both visuals and performance. According to a report by 4gamer.net featuring insights from Chris Hall, Senior Director of Software Development at AMD, FSR Redstone leverages the powerful ML2CODE (Machine Learning to Code) framework, which plays a crucial role in this technology.

The primary function of AMD’s ML2CODE is straightforward yet effective: it converts pre-trained neural network models into GPU compute shader code. This process generates optimized HLSL code that can be executed across a wide range of GPUs that support modern shader technologies. Consequently, FSR Redstone requires runtime ML inference, with ML2CODE acting as a bridge that translates the neural rendering core into standard compute shaders. This approach ensures that the resulting shader code can operate on vertices from AMD, NVIDIA, and Intel GPUs, thus offering unparalleled cross-platform support.

AMD's Radeon RX 9070 XT and RX 9070 GPUs — AMD’s RX 9000 GPUs | Image Credits: AMD

FSR Redstone was developed using AMD ML2CODE (Machine Learning to Code), a research project from ROCm. The core part of the neural rendering technology is converted into optimized Compute Shader code by utilizing ML2CODE. This means that FSR Redstone’s neural rendering core can also run on GPUs made by other companies.

At AMD, we use HIP in the development process for many innovative new AI-related technologies. ML2CODE aims to integrate with the most commonly used graphics rendering pipelines, such as Vulkan’s shader language “GLSL”and DirectX’s “HLSL”.

It’s highly likely that the AI cores of the various AI-related functions used in FSR Redstone are developed using HIP code. This is because HIP code can output code optimized for each generation of Radeon GPU, and thanks to this architecture, it can also run on GPUs other than AMD. Regardless of whether this makes sense, if HIP code is converted to CUDA and built with an NVIDIA compiler, it will likely run on an NVIDIA GPU.

AMD’s Senior Director of Software Development, Chris Hall (via 4Gamer)

In a noteworthy revelation, Hall confirmed that FSR Redstone does not necessitate AI acceleration capabilities for its operations. This design choice ensures that all machine learning enhancements can be utilized on older GPU models, as the system optimizes shader code prior to execution—thus eliminating the reliance on AI compute resources during runtime. While older hardware may experience some performance overhead, broad support is anticipated.

This development marks a significant step forward in rendering technology, particularly within AMD’s RDNA framework. Previous iterations like FSR 4 were limited to RDNA 4, leaving earlier generations unsupported. Given that Redstone represents a pioneering ML-based implementation from AMD, it holds the potential to deliver notable performance enhancements on RDNA 3 systems as well.

Source & Images