Hi, I was wondering how can you implement AI so that my agents not just track and find the shortest path to the goal but also implement some strategic rules, for ex. the agents decide that instead of track me down, it waits in a certain position to caught me later or trying to find path behind my back. I''m kinda new on this matter and the only thing I''ve learn so far is A* and for AI behaviour I use simple state machine. Any help will do just fineee.. Thanks.

Try this. You make a special buffer you fill with zeroes. Then, each time your enemy moves, he leaves behind him some clue (number). For example 10. And then, with each game cycle, you decrease the numbers in the map (if it isn''t 0 alreayd). That makes the enemy leave a "path" behind him that disappears with time and you can easily follow it or learn, how far he is. Hope this helps

*Grumbles about an "Internal Server Error" eating his post*

I would suggest reading the book "AI Game Programming Wisdom", looking at William van der Sterren''s website (http://www.cgf-ai.com), and looking at the site http://ai-depot.com/. All of those are good places to look for information on tactical terrain analysis (making bot know where to wait for an ambush, when they should snipe, how to execute squad maneuvers, etc)

"The Requested Information Is Unknown Or Classified" -Anonymous