taichi VS Halide

taichi – Productive, portable, and performant GPU programming in Python

ETH Zurich is using it for their physics sim courses, University of Utah is using it for simulations (SIGGRAPH 2022), OPPO (they make smart devices running Android), Kuaishou uses it for liquid and gas simulation on GPUs. Lots of GPU accelerated sim stuff.

https://www.taichi-lang.org/

https://www.researchgate.net/publication/337118128_Taichi_a_...

https://github.com/taichi-dev/taichi

Check our the realease note (https://github.com/taichi-dev/taichi/releases) for more improvements.

I've recently started using Taichi (https://taichi-lang.org/) for numerical codes and the fact it doesn't try to trick you into thinking it's numpy is a nice "feature". ;)

Taichi v1.4.0 is released! See what's new: - Taichi AOT, along with a native Taichi Runtime library: Native applications can now load compiled AOT modules and launch Taichi kernels without a Python interpreter. - Taichi ndarray: An array object that holds contiguous multi-dimensional data to allow easy data exchange with external libraries. - Dynamic index: Use variable indices whenever necessary on all backends without affecting the performance of those matrices with only constant indices. See deprecation and more improvements in the release note.

Oh, then if you're not already tied to any particular VFX software, I might as well recommend Taichi again.

It's an upgrade of a basic version. See changes to the source code here: https://github.com/taichi-dev/taichi/pull/6851

If CPU/GPU execution speed is the goal while simultaneously code golfing the source size, https://halide-lang.org/ might have come in handy.

This is a task where Halide https://halide-lang.org/ could really shine! It disconnects logic from scheduling (unrolling, vectorizing, tiling, caching intermediates etc), so every step the author describes in the article is a tunable in halide. halide doesn't appear to have bindings for golang so calling C++ from go might be the only viable option.

Probably would have been much easier to do 15 times in https://halide-lang.org/

The idea behind Halide is that scheduling memory access patterns is critical to performance. But, access patterns being interwoven into arithmetic algorithms makes them difficult to modify separately.

So, in Halide you specify the arithmetic and the schedule separately so you can rapidly iterate on either.

It is not the sorting per-se which was improved here, but sorting (particularly short sequences) on modern CPUs with really the complexity being on the difficulty of predicting what will work quickly on these modern CPUs.

Doing an empirical algorithm search to find which algorithms fit well on modern CPUs/memory systems is pretty common, see e.g. FFTW, ATLAS, https://halide-lang.org/

Halide https://halide-lang.org/

It doesn’t apply in this case, but in general if you really want the best vectorization I would suggest using https://halide-lang.org instead of trying to coerce your compiler.

If we drop the "APL" requirement, wouldn't Halide fit your criteria for the third?