Go on Silicon— AI Vision and Human Insight in Chip Design
M37—AlphaGo’s otherworldly move—looks ahead by tracing the evolution of algorithms, data, and models.
M78—Lee Sedol’s “divine move”—discerns the way forward through architecture, experience, and systems engineering.
When these two modes of thinking converge on silicon, this technology blog becomes more than a record of ideas: it is an in-depth reflection on GPUs, AI computing, and chip design.
No articles for this year
“There are a thousand Hamlets in a thousand people’s eyes.” The engineering world is no exception. Every engineer has a preferred GPU. Compute throughput? Bandwidth? Utilization? In my design philosophy, the efficient use of every piece of silicon is the first thing I look for in a GPU.
-M37
Bandwidth trade-offs between Immediate-Mode Rendering and Tile-Based Rendering.
-M78
11th Floor, Building S1, 889 Bibo Road, Pudong New Area, Shanghai
