Psyringe
Scout
A recent game of mine reinforced my belief that the current AI's decision-making with regards to colonies doesn't work very well. Here's what happened:
Zara Yaqob has expanded from his mainland into an archipelago. At 530 AD, he manages to found Judaism is one of his mainland cities. In 610 AD, he also founds Hinduism, which happens to be founded on the largest island in the archipelago. Zara Yaqob has three cities on this island, and none of them has been reached by Judaism yet.
At about 1500 AD, Zara Yaqob decides to release the cities on this island as a colony, led by Alexander. At this point, Judaism still hadn't spread to any of those cities. On the other hand, Hinduism had spread to very few cities on Zara Yaqob's mainland (if at all).
In 1520, Alexander revolts and breaks free from Zara Yaqob. He gets involved in a few wars, and in 1690 he ends up capitulating to Hammurabi, remaining his vassal for the rest of the game.
What's odd imho is Zara Yaqob's decision to release these three cities as a colony. It was foreseeable that Alexander would switch to Hinduism (he didn't even have any city with Judaism) and that the relations would then deteriorate due to the diplomacy modifiers. But the decision wasn't only bad politically, it was bad financially as well. By giving up a holy city, Zara Yaqob robbed himself of a very good potential source of revenue (especially considering that I was playing on a giant map with 400+ cities, so each holy city was a potential financial powerhouse). Also, Zara Yaqob had about 8 other cities on other islands (though no more than two on each), so I'm not sure whether he really gained that much maintenance cost relief by releasing these three. In any case he'd been better off releasing other cities or (if it's not possible to release less than three cities as a colonies) to give one or two cities away to competitors, instead of giving a holy city to a colony which was bound to break free.
I'm not sure whether this is worth doing something about it, the situation I describe is very specific and probably won't come up often, and generally improving the AI's decisions about colonies might be a can of worms that you don't want to open at this point. But I thought it wouldn't hurt to at least report my findings. Feel free to ask for further details or to point out errors when I made some.
Oh, and in case it matters: The game was played with BetterAI 0.41.
Zara Yaqob has expanded from his mainland into an archipelago. At 530 AD, he manages to found Judaism is one of his mainland cities. In 610 AD, he also founds Hinduism, which happens to be founded on the largest island in the archipelago. Zara Yaqob has three cities on this island, and none of them has been reached by Judaism yet.
At about 1500 AD, Zara Yaqob decides to release the cities on this island as a colony, led by Alexander. At this point, Judaism still hadn't spread to any of those cities. On the other hand, Hinduism had spread to very few cities on Zara Yaqob's mainland (if at all).
In 1520, Alexander revolts and breaks free from Zara Yaqob. He gets involved in a few wars, and in 1690 he ends up capitulating to Hammurabi, remaining his vassal for the rest of the game.
What's odd imho is Zara Yaqob's decision to release these three cities as a colony. It was foreseeable that Alexander would switch to Hinduism (he didn't even have any city with Judaism) and that the relations would then deteriorate due to the diplomacy modifiers. But the decision wasn't only bad politically, it was bad financially as well. By giving up a holy city, Zara Yaqob robbed himself of a very good potential source of revenue (especially considering that I was playing on a giant map with 400+ cities, so each holy city was a potential financial powerhouse). Also, Zara Yaqob had about 8 other cities on other islands (though no more than two on each), so I'm not sure whether he really gained that much maintenance cost relief by releasing these three. In any case he'd been better off releasing other cities or (if it's not possible to release less than three cities as a colonies) to give one or two cities away to competitors, instead of giving a holy city to a colony which was bound to break free.
I'm not sure whether this is worth doing something about it, the situation I describe is very specific and probably won't come up often, and generally improving the AI's decisions about colonies might be a can of worms that you don't want to open at this point. But I thought it wouldn't hurt to at least report my findings. Feel free to ask for further details or to point out errors when I made some.
Oh, and in case it matters: The game was played with BetterAI 0.41.
