quote:
Originally posted by evader:
How does the AI decide what Intellagence projects it does
No one knows but the programmer.
We've been requesting an external control file -- 'AI_Intelligence.txt" on the same design lines as "AI_Research.txt" -- but we've not heard any encouraging sounds from the source. Apparently we're stuck with the internal processes of the AI for the forseeable future.
There is one thing you can do with the text files, though. Split intelligence into three seperate areas. Leave intelligence facilities and counter-intelligence under "Applied Intelligence" and put 'espionage' (gather information operations) under a seperate Espionage Category and Sabotage (actual attacks that are supposed to do damage) under a new Sabotage Category. Now, leave the AIs alone that you think are too 'honorable' or just not capable of sabotage. They will defend themselves much better with no other intel projects but counter-intelligence. The 'xenophobic' races are really probably incapable of understanding another society, for example, and also ought to be very hard to penetrate. But for the races you think out to be 'crafty' and flexible enough to use sabotage, add the sabotage area to their research file. They will then have sabotage and no espionage projects, so they will be much more concentrated in their attacks.
[This message has been edited by Baron Munchausen (edited 22 October 2001).]
AI and Intelagence
Linear Mode
