I saw an issue and an example in the AssemblyScript repo.
The example combines BABYLON.SolidParticleSystem with WASM / AssemblyScript, aims to accelerate SPS.buildMesh().
So what’s going on now? Is it pending to observe more time 'cause still have bugs or it’s ready to shift to the main branch after 4.0?
Because my project heavily depends on SPS to reduce draw calls and memory usage, so could I use the code in the example include the wasm file in production environment?
The WASM SPS was just a prototype, a partly rewriting of the SPS code in order to check how WASM could speed up computations on big loops or big data array. It’s not planned to be merged in BJS core, and AFAIK there’s no plan to port BJS parts to WASM up now.
I understand.
Actually, I also have extended SPS a lot in my project, like add vertex data directly and other functions, so I don’t mind keep on improving my modified SPS. If it works, i will add this feature to my project.
Could you share more information about the test, so i could save lots of time.
So could you please tell me according to your test, does WASM accelerate greatly when build mesh and update particles? And is there any trouble side effect brought by WASM?
Thank you again!
you’re very appreciated