Quote Originally Posted by Davor
I think it also has to do with this 'spoil' below, but you’re probably aware of it already.
I don't think so because ASFAIK these settings only effect the priority of the next building (eg. first market place, second temple) or unit (first horse archer next spearmen) to be bought by the AI and not the tendency to either spend the last mne on a market place or a squadron of horse archers.

You want to say that with your script a faction never got in red? Your sentence seems to imply the opposite: AI is smart enough, as long as it doesn’t lose important cities, to run a positive treasury?
That seems to be true for the factions that do not get money help at all and by this are not objects of the script (Rome, AS, Karthage etc). So, it doesn't have anything to do with my script, but shows that the AI is able to deal with the thougher economy in EB; provided it is running a faction that is about stable in economics when the game starts.

The help is for those factions that are either for some time in trouble or won't ever recover at all (for example the Arverni and Aedui, who never have a balanced economy when run by the AI).

5000 seems too much money per turn/per town to me. Imagine SA with 20 towns going slightly (-100) into red. This would give them a 100.000 minae boost. At least, this is how I understood your script, partly because I understood you this way, and partly because of my wrong interpretation of the RTW money counting.
Yes, and that is the trick: they don't get more money (this turn) if the money help has brought them into the green on any settlement turn start. I was certain that the programm works this way, because the AI will check its towns only once per turn; and when it constructs new buildings or raises new units the treasury must always be updated before the next settlement's turn starts.

I was on the other hand concerned that the AI is now able to spend the 5,000 on every town; but no: the AI isn't able to (or seems not to be) spend more money than it has.


The 5,000 are not a number proved by hard scholary examinations and scientific proofed tests. It was about the first number that came to my mind when I thought on a summ that could help the one-town factions out of the debth. It is about the negative you'll make every turn with factions like Sweboz, Pontos or Hay when you don't do anything.


NOTE1: maybe increasing the average unrest for the player-cities could also increase the overall difficulty of the game as the player would have to leave more troops in cities when campaigning.
May be.

On the other hand, troop numbers are not the problem of the AI but the usage of troops. May be the tests with the Alex.exe in other thread will show some improved battlefield a stratmap behaviour of the AI?