Jump to content
  • Advertisement


This topic is now archived and is closed to further replies.


Training Sets for a Fighting Game

This topic is 6586 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

Im working on a 2D Mortal Kombat style fighting game. The Art I have is original and the engine is just about complete. But Im kinda stuck with a certian dilema. From the 2 games that I have written in the past. Most of the AI I used was basically pattern style AI affected by rand() and player distance. Even though I could pull of an in depth pattern style AI script for the fighting game, maybe its just me but then it really wouldnt be intelligent. I wont mention any names but there are a couple of fighting games that used such AI. Anyway I have heard about "Training Sets" but have found little to no info on it. Correct me if Im wrong but I belive it works this way. As the agent is affected in the game world by the player it records all of its experiences (success and failures) into a .dat file. Then it continues to save this data and eventually recalls it to figure out what do and check and see if it had experienced it before. Thus it would know what worked and what didnt. (That was the sky high over view of what I know about it) If anyone knows about this or where I can find specific info on it. I would appreciate it. Also if you recommend a better AI approach for the fighting game please suggest it thanks!!! a_insomniac If my compiler was voice activated with the voice of my mother-in-law....there would be no errors!

Share this post

Link to post
Share on other sites

  • Advertisement

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

GameDev.net is your game development community. Create an account for your GameDev Portfolio and participate in the largest developer community in the games industry.

Sign me up!