265
'The biggest speedup I've seen so far' — FFmpeg devs boast of another 100x leap thanks to handwritten assembly code (www.tomshardware.com)
submitted 1 year ago by to c/opensource@programming.dev
18 comments fedilink hide all child comments
 

But admit this boost is only seen in 'an obscure filter'.

you are viewing a single comment's thread
view the rest of the comments
[–] 37 points 1 year ago (1 child)

AVX512, SIMD

It's not just "handwritten assembly", it's all intrinsics, again. The reason a lot of tech that needs to use fast matrix algebra (or fast numeric math in general) tries to use the same small set of libraries, tightly optimized to use those optimized instruction sets.

  • permalink
  • fedilink
  • source
  • hideshow 1 child comment
    •