factual

Handwritten SIMD assembly far outperforms compilers

Handwritten SIMD (single instruction multiple data) assembly delivers 10x-50x (sometimes 62x) speedups over scalar code and is multiple times faster than compiler auto-vectorization or intrinsics, contradicting the common belief that good compilers can match handcrafted assembly; this matters because FFmpeg/dav1d run on order-of-magnitude a billion CPUs where every cycle counts.

factualpending

Speaker

Kieran Kunhya

Evidence Quote

“the key thing that we do differently in FFmpeg is we don't use any abstractions or any major abstractions on top of that”

Source

Lex Fridman Podcast #496 - FFmpeg— Lex Fridman Podcast
Created: 6/13/2026, 12:26:55 AM

Your notes

Loading notes...