05-29-2015, 01:37 AM
@xoft I can't tell whether you're describing cached Dijkstra's algorithm. It would work so long as you didn't have two many mobs and it was never used for pathfinding beyond a certain distance. The problem is that you need ~800KB of information per mob. If you're just caching the cells that are evaluated, then it would be better to move back to the original, cPathFinder/cPath interface. However I suspect that Safwat's planned reverse Dijkstra player field optimization would be more useful.