Quote Originally Posted by ReluctantSamurai View Post
It occurred to me that the discussion has been focused on how the player handles the economy and research, and so it should. But....

....in reading through a lot of campaign descriptions it seem apparent the the AI hasn't a clue for doing either very well. Factions rise then fall because they run out of food or get crushed by squalor. While this may mimic reality in some small sense (the ebb and rise of empires), it doesn't really present the player with that one faction that's going to be able to kick your ass (for Shogun 1 it was the Hojo Horde; for MTW the Golden Horde; for RTW it was the Romans if you were playing a non-Roman faction, and Egypt for just about everyone else; etc).

So how would you simplify development to where the AI could conceivably handle it, and yet keep it interesting for a player? or would you just have a script for an AI faction to follow?
This is pretty simplistic, but one approach might be to just limit its development. Say, AI cannot build past Level III in in provincial capital slots, and cannot build minor settlements past level II. Moreover, limit AI factions to only one or two military buildilngs overall, and no more than one temple per settlement. With something like this, the AI economy would still be quite suboptimal, but could probably maintain some reasonable level of public order and support some decent-sized army stacks. When I take over enemy settlements, I usually find that the AI has way overbuilt in temples or military buildings, and/or has built a Level IV town & temples without remotely enough food to support.