I would say “terrain,” but I don’t think I want to use Unity’s built-in Terrains for this.
What I am trying to avoid is setting every single vertex’s position every single frame when there are only a handful of them that are actually being moved. I think being able to do this is make-or-break for the feasibility of a new game concept I’m kicking around.
I think I should have an array (dictionary) of vertex numbers vs world space x-z coords. Then another of vertex numbers vs an array of all the numbers of that vertex’s closest neighbors. (Which is probably harder to calculate than it is to say.)
Am I on the right track though?
The procedural examples all iterate through every vertex in a temporary array and then set the mesh’s vertex positions (or colors) all at once. Is there some reason why you “have to” do it like that?