Pathfinding In Pico-8 скачать в хорошем качестве

Pathfinding In Pico-8

This set of demos show off my pathfinding system I developed for my project, Bun Run. The method I developed I later found out to be called a Dijkstra's Map. The system is simple enough that its token footprint is miniscule and such that agents in the environment do not need to individually path to a goal. Instead, a series of layered goal grids are generated per end goal, that can be sampled at any node for the smallest traversal cost, then the pathing grid for the nearest goal is selected and the agent follows that grid to its end point. In this method, agents move from areas of high value nodes, to low value nodes, with zero representing the end goal. My last example shown demonstrates that a large number of entities can reference the pathing grid and head to the goal without any perceivable lag.