Open MaxGraey opened 5 years ago
I saw this: https://hacks.mozilla.org/2019/03/fast-bump-allocated-virtual-doms-with-rust-and-wasm
Library | Optimized? | Milliseconds |
---|---|---|
Angular 2 | Yes | 2371 |
Dodrio 0.1-prerelease | No | 2181 |
But Angular is not fastest framework. It will be great see comparison with vanilla js (as baseline) and stdweb .
FYI I created an implementation of that benchmark for wasm-bindgen
that can be used as a point of reference/starting point: https://github.com/krausest/js-framework-benchmark/tree/master/frameworks/keyed/wasm-bindgen
It looks like wasm-bindgen
is pretty fast! Currently it's ranked 4th https://krausest.github.io/js-framework-benchmark/current.html
Another benchmark worth mentioning is UIBench. https://localvoid.github.io/uibench/ I like this one because it's easier to implement and run benchmarks. Here's a wasm-bindgen
implementation I created as a point of reference:
Yes, pure wasm-bindgen
as well as stdweb
pretty fast, but frameworks with vdom like yew (which use stdweb under the hood) is one of the slowest frameworks compared to JavaScript based
The wasm-bindgen
benchmarks are ports of the vanilla JS implementations so they show the overhead costs vs JS.
Also there's nothing inherently wrong with VDOM. ivi uses VDOM and it's pretty fast. I imagine a Rust port would perform similarly, adjusting for overhead.
Looks like you've seen all the benchmarking that currently exists.
Working on an implementation for js-framework-benchmark, and will also see if that turns up any more interesting corners of perf to look at. Thanks for your patience :)
Looking forward to getting your PR. Let me know if you run into problems with the benchmark and I‘ll try to help.
Any updates on this? I'd love to see how dodrio performs in benchmarks vs other wasm projects and javascript frameworks.
It will be great have some one on this site: https://stefankrause.net/js-frameworks-benchmark8/table.html