Hacker News new | past | comments | ask | show | jobs | submit login

I wonder what makes SIMD an improvement here - in the end it all boils down to TTF bytecode and I don't think this comes with SIMD instructions, right?



wasm has SIMD extension https://github.com/WebAssembly/simd/blob/master/proposals/si... . You can use them even in browsers (e.g. Chrome >= 91)


OP is asking how SIMD is beneficial to this specific application. If there's a small neural network involved then evaluating that is probably a good place to use SIMD.


Yes. The program actually includes an ONNX runtime, which uses SIMD to accelerate NN inference.


but... the program ends up entirely compiled and executed as TTF bytecode in the end right, since it's entirely contained in the TTF font ? And TTF bytecode is only the following instructions : https://developer.apple.com/fonts/TrueType-Reference-Manual/... and I don't see anything related to SIMD in there


No that is not what is happening, HarfBuzz have a experimental Wasm shaper, so the font embeds some wasm code that tells Harfbuzz what to output.

https://github.com/harfbuzz/harfbuzz/blob/main/docs/wasm-sha...


oh ok, so it can't work in any system that doesn't use harfbuzz.. that's much less interesting than what I originally thought


It’s not TTF bytecode, but WASM bytecode. There’s an experimental version of HarfBuzz that can run this.




Consider applying for YC's Spring batch! Applications are open till Feb 11.

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: