Invasion_priority vs pts_desire in descr_campaign_ai_db

[bitterhowl]

Hi fellow modders. During my investigations on CAI I find that pts_desire parameter is avoided by vanilla and most of mods. As well as most_desirable_faction. While reading tutorials about CAI I met invasion_priority and thought it's a parameter for faction desire to invade. But in my Westeros ironborn, who started on few islands, ignored Lannisters who are their neighbour and faction to outlve - by any invade_priority and straight order to invade on particular turn and even after few scripted sieges they were passive.

But everything has changed with pts_desire 400 on is_faction_to_outlive. Lannisters are smashed without any scripts or direct orders. More to say - I want ironborn to became vultures who attack weakened factions - and with military_balance 0.4-0.5 and increasing pts_desire it works on a whole map.

So I want to share this and clairify for myself difference between them.

[z3n]

Interesting, thanks for sharing.

I believe it is avoided sometimes as there was something wonky with how the scores are added. They go up without going down ever, so it makes it harder for factions to switch interests. But it has been a while since I looked at it closely. Very interesting find however about distance!

[bitterhowl]

Engine counts pts_desire anyway by internal algorythm and this factor seems to be important. Faction_to_outlive is just an example, we can ignore it and use more sensible indicators. Military_balance, production_balance, num_enemies - anything you think to be valid.

I think we can help AI to collect more strenght on one main direction by checking if target is most_desirable or not. If yes - we set proactive invade tactics if no - opportunistic or no invasion at the moment. Same for defend - maybe I'm wrong but AI need no deep defence strategy on main attack direction. So for most_desirable faction defend probably should be minimazed.

[Piterx93]

Engine counts pts_desire anyway by internal algorythm and this factor seems to be important. Faction_to_outlive is just an example, we can ignore it and use more sensible indicators. Military_balance, production_balance, num_enemies - anything you think to be valid.

I think we can help AI to collect more strenght on one main direction by checking if target is most_desirable or not. If yes - we set proactive invade tactics if no - opportunistic or no invasion at the moment. Same for defend - maybe I'm wrong but AI need no deep defence strategy on main attack direction. So for most_desirable faction defend probably should be minimazed.

Gathering in one direction does not depend on the choice of invasion target.

[bitterhowl]

I continue my studies and have to say that using +/- pts_desire I can direct AI to choose target according my needs without any campaign_script coding.
In my mod for Westeros I got only one catholic fsction, so this is example is a hypothetical one - but. what do you think about extremely raising pts_desire for catholics to excommunicated faction? I think thisway we can make papal relations much more sensible to player.

In my tests I add more invade_priority on most_desirable target and I think I see this method rather effective.