Guys, It’s a huge topic that I tried to discover by myself with a lot of attempts. I didn’t succeed in C# for some reason. Some of implementations obviously contains bugs, some of them works with fancy and not suitable for VL datastructures, some of them doesn’t works for no reason. VL tries to hit me for recursion and other attempts to “do the right”. Actually, I’m not too mature to do that kind of thing.
There are a lot of algorithms that may and sometimes should be implemented with spatial data structures and optimizations with them. For example, I think drawing nodes in VL may be corresponding with this type of tasks — adding to graph, “intersecting” this graph with render region, exclude evaluation outside the “render view”, and why not to add more “renders” in that case? Physics? Cellular automatas? Huge parallel calculations with OpenCV? Even compute shaders. A lot of serious computational algorithms need optimization like that, not only custom generative image processing.
R-tree, Quadtree, Octtree, k-d trees. I’ve had an experience with Rust and I was really surprised how much of beautiful implementations are in Rust and how easy to use them. I’m even tried to write wrappers, but it’s all looks and feels ugly. Actually, I had couple hybrid tools Rust/VL because of that.
It would be interesting to hear your opinion. Are there any solutions that I’m not aware of?