Is there any sophisticated game AI engine that can save time on AI implimentation and provides strong and clever AI? The game genre is strategy, so AI should be strong. The game in some aspects is similar to the starcraft. So we can use it as good example. As for me SC2 game AI is is very weak. When I played it, I quickly found very simple strategy line with help of it which could win in 100% of cases for any race. After that it was not interesting to play. So I want stronger AI, and AI that can learn. I don't like deterministic algorithms. I can accept some good end state machine algorithm, but good implimentation of such algos requare long time and testing. And all of such algos in stratgies are week and not interesting once the player has found weaknesses.

I can try to impliment AI by myself using neural networks, genetic algo or something like that.. but that will requare even more time.

So I am interesting in some ready sophisticated game engine, that would be easy to train and adjust under myself. I looked and found only http://alive.sourceforge.net/ ... it not looks very populat, fast and sophisticated.

(The game is openGL/C++. I have enaph free processor resources for AI)


closed as too broad by congusbongus, Philipp, bummzack, Josh Mar 18 '15 at 16:22

Please edit the question to limit it to a specific problem with enough detail to identify an adequate answer. Avoid asking multiple distinct questions at once. See the How to Ask page for help clarifying this question. If this question can be reworded to fit the rules in the help center, please edit the question.

  • \$\begingroup\$ What type of ai are you looking for? What purpose does it have to serve? Are you looking for a logicaly strong ai then you might want to look into a logic language such as prolog (and it's derivative GOAL) or are you looking for a simple expert system? Or do you want it to be self learning (neuronet, evolutary)? \$\endgroup\$ – Thijser Mar 18 '15 at 9:08
  • \$\begingroup\$ @Thijser For strategy like starcraft, I think self learning is better for this ganre... \$\endgroup\$ – Brans Mar 18 '15 at 9:14
  • \$\begingroup\$ Can you add that to your question? Try to explain what you are trying to do, how complex is the ai going to be? AI is a large subject and there are many techniques for you to chose from each carrying it's own up/down sides. Also try to provide information on your enviroment (have you made the rest of the game yet? Are you working in unity? Are you working in openGL/C++? How fast do you want the AI to work? Is it deseriable that the AI is non-determanistic? ) \$\endgroup\$ – Thijser Mar 18 '15 at 9:22
  • \$\begingroup\$ Hello and welcome to GDSE. While your intentions are good, this kind of questions don't really have place in here, because answers would be opinion based and thus, out of the scope. I addition, you show zero effort by yourself, you have not tried anything, not googled anything, studied anything, you just expect someone to do it for you. I flagged this as offtopic, under "what tech to use". Those kinds of questions are outside the scope of this site. Visit our help center for more information. PS: At minimum, show what you have tried, looked at, what language etc. \$\endgroup\$ – Katu Mar 18 '15 at 9:26
  • 4
    \$\begingroup\$ I doubt that there is such a framework. The bulk-work of writing a strong game AI is not reinventing the low-level grunt work which frameworks could cover for you. It is writing a specialized expert system which understands how your game works, which is highly game-specific. And real-time strategy in particular is a genre which is very unsuitable for standard machine-learning algorithms like neural networks, because the link between action and consequence is often a very indirect one. \$\endgroup\$ – Philipp Mar 18 '15 at 12:50

Browse other questions tagged or ask your own question.