sman1975
Emperor
@askmarcoanything - I made a few design notes for a global WW2 scenario, or even a Pacific Theater scenario, but didn't get very far along. The problem was designing a decent scenario management engine (SME) that twists AI behavior into something resembling rational game play. While it's almost possible to develop one for the Pacific scenario, it's way beyond my expertise/available time to come up with one. Look at this AWAW - it's almost all database changes, with no SME at all, and it's taken 3 months (so far)...
The main problem is the AI simply doesn't behave like a normal player would in an existential conflict. While human players are cranking out infantry, air units, carriers, etc., AI civs will still be building cultural and scientific buildings, trying to win a cultural victory, even though that victory condition has been disabled. There's not a good method for you to "order" an AI unit to move to tile X/Y. There are a couple that you can use, but as these orders are quickly overwritten by other game logic, the effects are ephemeral at best. I tend to look at those methods as "suggestions" more than actual orders. On a small-focused map, like Europe, these flaws aren't so obvious. However, when you put it on a large focus map, the performance of the AI is quite disappointing. It's one of the reasons I haven't pursued these efforts very far.
The main problem is the AI simply doesn't behave like a normal player would in an existential conflict. While human players are cranking out infantry, air units, carriers, etc., AI civs will still be building cultural and scientific buildings, trying to win a cultural victory, even though that victory condition has been disabled. There's not a good method for you to "order" an AI unit to move to tile X/Y. There are a couple that you can use, but as these orders are quickly overwritten by other game logic, the effects are ephemeral at best. I tend to look at those methods as "suggestions" more than actual orders. On a small-focused map, like Europe, these flaws aren't so obvious. However, when you put it on a large focus map, the performance of the AI is quite disappointing. It's one of the reasons I haven't pursued these efforts very far.