Shane Peelar

At the very least, is it possible we could supply Plot with our own allocator? That would go some way towards this goal, but hopefully shouldn't be too intrusive on...

Nice! I will give this a shot and get back to you about how it goes.

#1822

So, thinking about this a little more, to truly get the benefits of an SoA representation for calculations such as `position = velocity*time`, we'd actually need each element of these...

Seems I just can't help myself! Digging around the code a bit, to facilitate this kind of change, we might need to ensure the allocator for the tables are correctly...

Just curious, is there interest in a prototype implementation of this? Or benchmarks to show potential speedup under (good) circumstances? I believe I could mock up a benchmark with the...

Agreed! I'll keep the PoC focused on just a hypothetical custom "Position" and "Vector" component :)

@alice-i-cecile Not quite ready to share my code yet, but my initial benchmarks are promising:  "naive" is the normal Vec3 way of doing position + time*velocity (e.g., AoS representation)...

Err yes... my bad! I'll edit the post (very sleepy right now).

While I'm working on some more in-depth benchmarks, I found this [excellent post from the nalgebra project](https://www.rustsim.org/blog/2020/03/23/simd-aosoa-in-nalgebra/) about different data layouts and their realization in Rust. The post is a...