This is an interest check primarily.
I have my game partak here:
https://itunes.apple.com/us/app/partak/id639879646?mt=8
You control a blob of colored particles and try to absorb the other blobs of particles. This game rests on a custom path finding system that currently has no contemporary in the app store. It is basically an implementation of Dijkstra’s alghorithm optimized through a quad tree. You case a see video it running here with some Debug.DrawRays to visualize what it is doing:
This pathfinding system has a lot of constraints. But it has a unique benefit over the other options in that it can path find many many agents simultaneously to a single point with very little overhead. On an iPad 2 the bigger constraint is not actually the pathfinding system, but rather the overhead of rendering 5000 seperate agents.
It could pathfind more agents, but an iPad 2 could not render that many thing simultaneously.
Basically in this system the overhead is all in creating a pathfinding ‘Gradient’, which in my game it can still do this 60 times a second an iPad 2, which occurs entirely in a seperate thread. But once the gradient is calculated, basically pathfinding on the gradient is extremely low over head, 5000 agents on iPad 2 is not an issue at all. But you could get only at most 60 new gradients/end points a second. So each agent cannot have it’s own unique endpoint, although they do find their own unique path.
I’m just curious, if I were to package this for something to go on the asset store, would anyone be interested? What kind of expectations would you have on this system? What would you want to use it for?