Tech

How far can we push computing?

Read full article at source ↗

Based on a feed excerpt; full article content has not been acquired. Source evidence retrieved 19/10/2026, 09:20:25 UTC.

Briefings

MIT Technology Review reports that while Moore's Law has stalled and Dennard's Law has ended, fundamental physical limits remain far below current data center scales. Feynman's 1959 lecture suggested there was 'plenty of room at the bottom,' yet the transistor's size reduction has halted. Key facts include the 1959 Caltech lecture by Richard Feynman, the 1985 win by Tom Newman for the 25,000x text shrinkage, and the 2026 hyperscaler spending of over $700 billion on data centers. The article notes that scaling laws are engineering observations, not fundamental laws, and that quantum mechanics and general relativity define true physical limits.

Key points

• Richard Feynman's 1959 lecture at Caltech argued there was 'plenty of room at the bottom' for computing, suggesting no physical law prevented building the entire Encyclopedia Britannica on a pinhead.
• The transistor size reduction has stalled, evidenced by the 2026 hyperscalers spending over $700 billion on infrastructure, which buys performance by brute force rather than scaling.

Context

This article analyzes the transition from the 20th-century information age to the 21st-century intelligence age, contrasting Moore's Law with the current slowdown in computing density and energy efficiency.

Limitations

Evidence is limited to a single article from October 2026. Specific dates for the 1959 lecture and 1985 text shrinkage are provided, but the exact year of the transistor's first successful integration is not specified in the text.

Generated by Bittrees-hosted qwen35-2b sha256:aaf42c8b7c3c on 19/10/2026, 09:24:19 UTC. Based on publisher-provided feed evidence; may not cover the full original article. Check the source for details.

View archived briefing on IPFS
bafybeigw3brvj3fssohhdjhg3kr2qtv4eljwxld7vqnn7a4jnzqhvvuxc4