TechTechPotato
Ian Cutress on chips and data centers, often with first-hand vendor interviews
11:47The superconducting-chip claim: change the temperature range, not the process node
Move the logic into the liquid-helium range and replace transistors with superconducting switches — that is Snowcap's bet: tape out RISC-V first, then claim that 28nm-class logic can deliver 5nm-class performance.
11:10AI Chip Shipments Are About to Jump an Order of Magnitude: From One Million Units to 15 Million
Broadcom's president walks through an AI chip roadmap that runs from a million units toward 15 million. The ‘beast’ is the largest package: eight stackable compute dies plus 16 HBM. Four of the five leading frontier labs are working with them.
31:10OpenAI's Own Chip Is 100x Ahead? That's a Self-Chosen Benchmark on a Dedicated Stack
OpenAI's first in-house chip, Jalapeno, beats the GB200 by 100x on a benchmark OpenAI picked — but that result comes from a dedicated chip running dedicated models on a dedicated software stack; the real moat is system integration and pricing power.
21:44Intel Drops EMIB for UCIe: 1.2GB of Cache Behind 256 Cores
Diamond Rapids is expected in 2027 with up to 256 P-cores, 1.2GB of last-level cache and 16 channels of DDR5; it replaces EMIB with trimmable compute building blocks and a UCIe interconnect, and the huge cache is its answer to the AI era.
What Etched's chip really solves may be bandwidth, not compute
Etched welds the transformer into hardware with 0.45-volt ultra-low-voltage transistors and direct cross-chip SRAM access, but the host suspects the problem actually being solved is bandwidth, not compute.
23:20Cerebras hits 4400 tok/s on a single chip: inference speed is the moat
Cerebras pushes WSE 3.5 to 4400 tok/s per chip, and the CS4 rack packs three wafers into a 120kW cabinet. The core argument: inference has to be split into prefill and decode, and speed is the moat.
30:20Custom Silicon Isn't About Saving Money, It's About Escaping Nvidia's Generational Grip
The real motive behind hyperscaler custom silicon is not cost but control: not being held hostage to Nvidia's generational cadence, while using sheer scale to squeeze down networking and compute costs. The Broadcom tax is cheaper than the Nvidia tax, but only the largest players can afford to play.
16:08AMD Buys Talus: Burning AI Models Into Silicon, 17,000 Tokens a Second
AMD has acquired Talus, a startup that builds custom chips for one specific AI model each and reaches 17,000 tokens per second. It marks a turn in AI inference toward non-programmable, model-specific silicon.