Hacker News Reader

pretentious7on Plane: Per-user backends for web apps

Thanks for the link! It's actually rendering to Dom huh... Interesting.

pretentious7on Plane: Per-user backends for web apps

Yep! (It's annoyingly nontrivial to do with gpu acceleration, but we'll have a guide out as soon as we figure it out)

(I work at drifting in space)

pretentious7on I want to give Oracle money but I can't

I have also been having some rather inexplicable issues with paying oracle, it's really strange.

pretentious7on Near, the emulator developer of bsnes, higan, and more, has died

But there's a lot of eccentrics whose life is the internet. To disconnect for them is essentially the same as solitary confinement.

Insofar as privacy, compartmentalization, and good opsec, near did all of that, doesn't help when the mob goes after friends and family.

pretentious7on Near, the emulator developer of bsnes, higan, and more, has died

1776 hosting is run by the KF site admin.

pretentious7on Maybe you don't need Rust and WASM to speed up your JS (2018)

https://blog.feather.systems/jekyll/update/2021/06/21/WasmPe...

Here's some benchmarks. But in summary, for small simple hot loops, chrome is just as fast using js, but for firefox wasm gives big performance improvements.

Of course, if you can use SIMD instructions, then wasm will win, but that's less fair.

pretentious7on Maybe you don't need Rust and WASM to speed up your JS (2018)

Does it still 404? Sorry I made some mistake with it. Theres a link in the post to an interactive demo.

The reason I said fastest is cause I wrote the hot loop with SIMD instructions. I'll be adding in period checking and stuff in a couple months time, will be sure to refer your work then, thanks.

pretentious7on Maybe you don't need Rust and WASM to speed up your JS (2018)

https://blog.feather.systems/jekyll/update/2021/06/21/WasmPe...

We did that! Yea chrome basically keeps right up, of course not accounting for SIMD

pretentious7on Maybe you don't need Rust and WASM to speed up your JS (2018)

(https://blog.feather.systems/jekyll/update/2021/06/21/WasmPe...) OP was very useful for us in optimizing our js vs wasm benchmarks. We were wondering what a very simple parallel problem like mandelbrot rendering would show about browser jits vs their wasm compilers.

Our conclusion was that wasm was consistent across browsers whereas js wasn't. Further, If you can use simd, wasm is faster. Also, v8 is way faster than spidermonkey.