← All stories
AI & Tech

Data Movement Costs Dominate Logic Operations by Factor of 7 to 1

Dwarkesh Patel Podcast · Reiner Pope – Chip design from the bottom up · May 22, 2026
Data Movement Costs Dominate Logic Operations by Factor of 7 to 1
Dwarkesh Patel Podcast
Dwarkesh Patel Podcast
Reiner Pope – Chip design from the bottom up
"So this circuit I've described here, almost all of the cost, like 7/8 of the cost is in the reading and writing the register file. And only a tiny fraction of the cost is in the logic unit itself."
Pope demonstrated through circuit analysis that in traditional CPU and GPU cores, 87% of the transistor area and energy goes to moving data between registers and compute units, with only 13% performing actual calculations. This fundamental inefficiency was the primary motivation for introducing specialized tensor cores and systolic arrays in modern AI accelerators.
From this episode
Dwarkesh Patel Podcast
Dwarkesh Patel Podcast

Reiner Pope – Chip design from the bottom up

August 3, 2026 · 5 Egleze moments
Read episode summary and key points →

More moments from this episode

AI & TechMaddx Developing Split-table Systolic Arrays to Bridge GPU-TPU Architecture GapAI & TechNVIDIA Changed Performance Specs After Acknowledging Quadratic Bit Width ScalingAI & TechGPUs Are Arrays of Tiny TPUs with Different Trade-offsAI & TechFPGA Energy Efficiency Not as Superior to ASICs as Commonly Assumed
More stories More from Dwarkesh Patel Podcast