Thanks, still a work in progress. Let me know what you think. The holiday has put a kink in my time table. Hopefully I will get something new up soon...

I would have thought there would be more interest than there has been. Maybe when it is done.

I think you'llfind there is more interest in this than your number of replies on this thread suggests. I, for one, am looking forward to it.

I am anxiously awaiting Part III.

Im definatly reading it so far. put part 1 and 2 on my site this week ( just have to add teh links is all.)

Definitely continue!

How is this coming along, anyway?

I have been a bit distracted with other stuff recently. Also the algorithm I was using in part II doesn't work for complicated paths. One could alway break out Dijkstra's Algorithm, but it seems a bit too terminator like. I want something a bit more fuzzy. Honestly with work I haven't had much of a chance other that to see that something a bit more complicated in order.

What is "Dijkstra's Algorithm"? I was under the impression a simple A* was what most people went with.

Edit> I just checked it out on the internet. I'd go with A*, it is a little more efficient and flexible. A pain to write in c-script though (you'll have to use single dimension arrays to store all that data). Perhaps you could do that side of it in the SDK and include it as a plugin with the tutorial? Dijkstra's Algorithm doesn't appear to be much less data intensive.

Dijkstra's algorythm was used in router paths if I remember right (dear god, my memory scares me). A* is the best to use for games because it is very accurate and also very fast compared to the other options. Precalculated paths using A* can increase performance by not doing the calculations in-game (which leaves you with a static, unchanging path table). Another option is to go with D* (Dynamic A*), it's a touch slower but allows much more dynamics.

-Rhuarc

I may just do that and add a fudge factor to force AI to not take the optimal path.