HBM4 controllers eat ~16% of Nvidia's Rubin die, sparking optical interconnect debate
BenBajarin · x · 2026-10-02
SemiAnalysis estimates HBM4 controllers and PHYs take up roughly 16% of Nvidia's Rubin compute die — a surprising share of leading-edge silicon spent interfacing with memory instead of computing. Ben Bajarin responds that optics could help two ways: moving memory off-die for near-memory HBM, and scale-in optical interconnects that free up the I/O shoreline of the die.
Related event: SemiAnalysis: Custom NVHBM Could Free Up ~25% of Die Area on Nvidia Feynman(2 posts)→
More from Infra
- Mobile AI Factory: agents auto-build on-device SLMs with 19x less GPU time — pmttyji · 2026-11-16
- Stratechery's Fictional Obituary: Imagining Jensen Huang's 2026 Exit From Nvidia — Gregory_C_Allen · 2026-11-16
- vLLM serves DeepSeek V4.1-Flash at 954K tokens/s per GPU, $38.96/GPU-hour on GB300 NVL72 — bookwormengr · 2026-11-16
- Colossus 2's 1.2M B200s could generate ~$400B/year in tokens at DeepSeek Flash rates — bookwormengr · 2026-11-16
- BIS issues rule blocking remote access to US AI chips via third-country data centers — teortaxesTex · 2026-11-16
- Qdrant 1.20.1: 50% fewer cold-storage reads, ~2x faster quantization, agent-ready setup — qdrant_engine · 2026-11-16