Basically, I would like to create some amazing JSettlers AI with numerous levels of difficulty and that don't behave predictively.
Having looked at the initial code, I know this won't be an easy task as there is extremely tight coupling between behaviours and objects.
I intend to start by doing some basic refactoring including breaking large methods into smaller methods with a more focused task as well as doing this for the classes themselves. Adapting the code to use the Strategy Pattern for different AI components of the game will mean developers can drop in individual AI components for testing.
I would also like to create a mechanism to run AI vs AI tests, so we can see the effectiveness of the new strategies when put up against existing strategies.
Lots of work to do!
Log in to post a comment.
Sign up for the SourceForge newsletter:
You seem to have CSS turned off.
Please don't fill out this field.