Nvidia Develops 'Feynman' GPU Microarchitecture to Succeed Rubin and Rubin Ultra

Nvidia has begun advancing prototype development of its next-generation GPU microarchitecture, codenamed "Feynman," which will succeed the Rubin and Rubin Ultra series.

The Feynman GPU will leverage backside power delivery technology on the A16 process, combined with multiple advanced packaging solutions, including SoIC-based 3D chiplet stacking, CoWoS-L 2.5D integration, and next-generation panel-level packaging. Initially, multiple chiplets are planned to be vertically stacked within a single package via SoIC to shorten data transmission paths between chips and reduce latency.

Such custom memory may employ base dies with integrated specific logic functions, such as memory controllers or packet processing units, enabling some data to be pre-processed before entering memory to improve overall transmission and computational efficiency.

Interconnect technology will be key to further expanding performance on the Feynman platform. Nvidia plans to introduce co-packaged optics, deploying optical interconnects between GPUs to reduce latency and power consumption pressures associated with traditional copper interconnects.

These plans remain at an early stage based on supply chain information; specific product specifications, timelines, and actual deployment are subject to confirmation by Nvidia and TSMC.