Supply chain insiders and sources close to TSMC have revealed that NVIDIA is making significant progress in developing its next-generation GPU microarchitecture prototype, codenamed “Feynman.” This new architecture is set to follow in the footsteps of the “Rubin” and “Rubin Ultra” series. Recent reports indicate that NVIDIA has opted to skip the N2 series process and instead directly utilize TSMC’s more advanced A16 process for manufacturing the relevant chips. Mass production is expected to commence in the second half of 2028.