Tiramisu
Warlord
In the recent years I have been working on my own AI flavors mod, which has required quite a lot of work, but I also learned a lot. Currently my AI mod is not published while I am still testing/improving it, but I thought I could share some of my personal insights.
AI Settlers: The AI has a very high flavor values for settlers (in the "Units" tables of the database), which makes them settle even during phases, when they don't really need them. As a result they may build settlers too early and in the later stages of the game the AI is likely to keep building settlers, which results in a lot of trashy cities.
In order to solve this problem, I have reduced FLAVOR_EXPANSION for settlers and significantly increased this flavor type for
ECONOMICAISTRATEGY_EARLY_EXPANSION and for
ECONOMICAISTRATEGY_EXPAND_LIKE_CRAZY
Notice that the flavor value for settler units is an absolute value, while for the economy strategies are relative percentage value (where Flavor = 100 means 100% increase of the base flavor value). So whenever the AI decides it requires to expand, it will build settlers much more likely now instead of building settlers too mindlessly. Firaxis made the mistake to increase the percentages at around 10-20% for the specialization of AI strategies, but in order to see differences you need to use values well above that.
Also notice that there are two tables, where you can change the flavor percentages.
1. AIMilitaryStrategy_City_Flavors and
2. AIMilitaryStrategy_Player_Flavors
In my case I chose the first table, because obviously AI cities are most important for regulating settler productions. The second table is probably more for general AI strategies.
Another thing I noticed is that the Leader table has some interesting columns like DoFWillingness, which obviously controls how likely the AI will declare war instead of sitting on a huge pile of military units.
In addition there is also the Boldness column, which apparently controls how far the AI is willing to settler from its original spot. So a high boldness means the AI is more likely to settle on good spots far away, but if the boldness is too high it will place its cities in dangerous places until its empire is stretching so wide that it becomes hard to defend.
Interestingly, a high FLAVOR_EXPANSION for the AI strategy and high boldness can lead to very early large AI empires even on Emperor difficulty, although this kind of strategy is usually too risky.
I have found out a lot more useful things about the AI flavor system, but for now that is it from me.
AI Settlers: The AI has a very high flavor values for settlers (in the "Units" tables of the database), which makes them settle even during phases, when they don't really need them. As a result they may build settlers too early and in the later stages of the game the AI is likely to keep building settlers, which results in a lot of trashy cities.
In order to solve this problem, I have reduced FLAVOR_EXPANSION for settlers and significantly increased this flavor type for
ECONOMICAISTRATEGY_EARLY_EXPANSION and for
ECONOMICAISTRATEGY_EXPAND_LIKE_CRAZY
Notice that the flavor value for settler units is an absolute value, while for the economy strategies are relative percentage value (where Flavor = 100 means 100% increase of the base flavor value). So whenever the AI decides it requires to expand, it will build settlers much more likely now instead of building settlers too mindlessly. Firaxis made the mistake to increase the percentages at around 10-20% for the specialization of AI strategies, but in order to see differences you need to use values well above that.
Also notice that there are two tables, where you can change the flavor percentages.
1. AIMilitaryStrategy_City_Flavors and
2. AIMilitaryStrategy_Player_Flavors
In my case I chose the first table, because obviously AI cities are most important for regulating settler productions. The second table is probably more for general AI strategies.
Another thing I noticed is that the Leader table has some interesting columns like DoFWillingness, which obviously controls how likely the AI will declare war instead of sitting on a huge pile of military units.
In addition there is also the Boldness column, which apparently controls how far the AI is willing to settler from its original spot. So a high boldness means the AI is more likely to settle on good spots far away, but if the boldness is too high it will place its cities in dangerous places until its empire is stretching so wide that it becomes hard to defend.
Interestingly, a high FLAVOR_EXPANSION for the AI strategy and high boldness can lead to very early large AI empires even on Emperor difficulty, although this kind of strategy is usually too risky.
I have found out a lot more useful things about the AI flavor system, but for now that is it from me.
Last edited:
