At 9/17/12 05:10 PM, I-smel wrote:
I got COMPLETELY LOST reading that
That's because he's probably using techno-babble and buzzwords to cover the fact that he's having to hack through his convoluted waypoint pathfinding system instead of building an efficient representation of the walkable spaces in his world :P Either that or he built his nodes in the correctly efficient way but didn't use them right yet.
That's what I made out of his explanation of what he's planning to do. Maybe the picture he's planning to post will prove me wrong.
PSvils, I'm interested to know how you set up your scheduling though. I'm assuming you split up the pathfinding process into several runs such that you don't get any information (as in partial, incomplete paths) before you get the final complete information (the path your algorithm returns)? I haven't thought about this before, as I haven't dealt with pathfinding too much, but it might be interesting to play with the optimisation of accuracy and length of partial path to an endpoint vs algorithm complexity (running time). Might be completely devoid of merit though.
These are interesting topics, and your project sounds fun. You should post structured writings about it, and not just pop up on the forum and start describing something from the middle of your train of thought when none of us have a clue about the background of what you're talking about.