News for hackers. Stuff that matters.
Re:Exploring SIMD performance improvements in WebAssembly (2... by jcelerier on Wednesday February 02, 2022 @06:34PM (#30185858)
very depressing when for so many use cases even native performance is very very very much not fast enough
Re:Exploring SIMD performance improvements in WebAssembly (2... by kevingadd on Wednesday February 02, 2022 @07:00PM (#30186227)
In many of these use cases native is only "not fast enough" because the code you're running makes very poor use of the cache, pipelining, simd instruction sets, and memory bandwidth
Re:Exploring SIMD performance improvements in WebAssembly (2... by Dylan16807 on Wednesday February 02, 2022 @06:49PM (#30186086)
> when

If native performance is "very very very" not fast enough then that's supercomputer work and it doesn't really matter if WASM is 3x native or 0.3x native. So that context should be where you're the least depressed.

Re:Exploring SIMD performance improvements in WebAssembly (2... by jcelerier on Thursday February 03, 2022 @05:37AM (#30190841)
> then that's supercomputer work

today's laptop work is late 90's supercomputer's work (and it was even more depressing back then).

Re:Exploring SIMD performance improvements in WebAssembly (2... by Dylan16807 on Thursday February 03, 2022 @02:11PM (#30197200)
And today's supercomputer work was impossible in the late 90's.

That doesn't really change my argument. When you're looking at languages that are used for small tasks today, their speed doesn't have much relevance to how vastly bigger tasks are accomplished. And by the time those tasks can be run on a laptop, WASM implementations are going to be much better and we still might not be using it at all for those larger tasks.