How mobile and specialization broke homogeneous compute into TPUs, NPUs, and more
blelbach · x · 2026-09-17
Compiler veteran blelbach recaps how compute went heterogeneous: the homogeneous world broke when image classification and mobile took off, and the push for more scaling brought specialization — hence TensorCore, TPUs, TMA, NPUs, and ANE. It's a golden age for hardware architects, at the expense of software people who must build a programming model for every three-letter-acronym accelerator.
More from Infra
- Jensen Huang says safety-testing data centers will become AI's third compute demand pillar — rohanpaul_ai · 2026-09-17
- Jensen Huang: a 1GW NVIDIA AI factory costs $50-60B but generates ~$50B in annual rental revenue — rohanpaul_ai · 2026-09-17
- Ilya warns neoclouds' weak cybersecurity invites rogue AI agents to hijack compute — Miles_Brundage · 2026-09-17
- AMD's free AI Developer Program: $100 cloud credits, Discord access, hardware raffles — wkmyrhang · 2026-09-17
- After AWS me-central-1 loss, dev jokes about explaining the outage to Codex weekly — andersonbcdefg · 2026-09-17
- The RAM Crisis Is Only Just the Beginning as AI Demand Squeezes Supply — perelin · 2026-09-17