AMD FSR Redstone arrives December 10

AMD has teased Redstone, a new FSR platform with details due December 10. It builds on FSR 4's Artificial Intelligence and machine learning super resolution and introduces neural radiance caching, Artificial Intelligence and machine learning-based ray regeneration, and frame generation.

AMD’s general manager and senior vice president Jack Huynh posted a teaser on X announcing Redstone, the next iteration of the company’s FSR platform, with a fuller reveal scheduled for December 10. Redstone is described as building on the Artificial Intelligence and machine learning super resolution introduced with FSR 4, signaling a continued emphasis on hybrid rendering techniques that blend traditional rasterization, path tracing, and learned components.

The Redstone platform introduces three headline technologies. Neural radiance caching relies on a machine learning model that continuously learns how light interacts inside a scene to predict and store indirect lighting. According to the announcement, this cached prediction aims to reduce the performance cost associated with ray tracing by supplying plausible indirect illumination without tracing every ray in full detail.

Redstone also adds Artificial Intelligence and machine learning-based ray regeneration, which the company likens to NVIDIA DLSS 3.5 Ray Reconstruction. That system uses a neural network to regenerate pixels that path tracing could not accurately resolve, with a particular focus on improving reflections when super resolution is applied. Finally, the platform includes an Artificial Intelligence and machine learning-based frame generation model that evolves AMD’s interpolation approach from FSR 3. This model integrates temporal and spatial awareness to generate interleaving frames with greater accuracy and enhanced image quality. AMD notes that this method does not aim to match the 2x frame-rate doubling associated with NVIDIA’s RTX 40-series Ada architecture and instead prioritizes visual fidelity.

55

Impact Score

Marvell unveils ultra low power 2 nm dense sram with major efficiency gains

Marvell is detailing performance data for its 2 nm custom sram ip, showing sharp gains in power, area, and bandwidth density over standard dense sram offerings. The company positions the architecture as a strategic edge as logic scaling continues to outpace memory in leading semiconductor nodes.

Contact Us

Got questions? Use the form to contact us.

Contact Form

Clicking next sends a verification code to your email. After verifying, you can enter your message.