Just wanted to sound people about about a change I propose to make in case there are violent objections...
The background is that I have found issues with the use of pioneers and colonists, in that, if you don't pe-prepare the ground for them (pre-creating improvements for food production) they tend to immediately found into starvation. One can readily argue that it is up to the player to make the necessary pre-improvements to avoid this. However, my counter-argument is that doing so puts the AI at a big disadvantage, since it has no clue there is any need to do this, and no logic for doing so. Not only would it be quite a large effort to add such logic, but we'd also have to change the way settler units are defined, since currently there are no tags that the AI can use to determine what a city starts off with (including population), and thus have a basis for decision making about what pre-preparation might be needed. This is because those units are defined (in terms of their differences from ordinary settlers) not by XML, but by Python actions when they found cities.
For the above reasons, what I propose to do is change the city founding functionality to check for the resulting city's food-storage-after-growth percentage, and add that much food at the time of founding. This represents the pioneers/colonists bringing some food with them (which I think is reasonable), and should serve to provide a sufficient buffer against starvation that the AI is able to build enough food prodcution in the new city before losing population (I'll check and if necessary adjust the buildings to ensure there is SOME storage included, but I think they provide granaries already so I believe that's already the case).
Opinions...?
The background is that I have found issues with the use of pioneers and colonists, in that, if you don't pe-prepare the ground for them (pre-creating improvements for food production) they tend to immediately found into starvation. One can readily argue that it is up to the player to make the necessary pre-improvements to avoid this. However, my counter-argument is that doing so puts the AI at a big disadvantage, since it has no clue there is any need to do this, and no logic for doing so. Not only would it be quite a large effort to add such logic, but we'd also have to change the way settler units are defined, since currently there are no tags that the AI can use to determine what a city starts off with (including population), and thus have a basis for decision making about what pre-preparation might be needed. This is because those units are defined (in terms of their differences from ordinary settlers) not by XML, but by Python actions when they found cities.
For the above reasons, what I propose to do is change the city founding functionality to check for the resulting city's food-storage-after-growth percentage, and add that much food at the time of founding. This represents the pioneers/colonists bringing some food with them (which I think is reasonable), and should serve to provide a sufficient buffer against starvation that the AI is able to build enough food prodcution in the new city before losing population (I'll check and if necessary adjust the buildings to ensure there is SOME storage included, but I think they provide granaries already so I believe that's already the case).
Opinions...?
