Re: ...and the AI's decision-process around resignation

[Hans Ronne <hronne at pp dot sbbs dot se>]

>Well, yes, if you have (for example) 2 strong players and one weak
>one, it probably is more interesting if the weak player doesn't resign
>(although it might not matter a huge amount, as the weak player will
>get conquered shortly at least in a game in which one captures
>cities).  But if you have 1 strong player, and any number of weak
>players, it really gets tedious if the weak ones don't resign.
>Especially since this tends to be late in the game when there are a
>lot of units to move.

Well, I think the second case also is more challenging if you keep the weak
AIs around. Fighting a single AI is almost never fun (= difficult), be it
strong or weak. Just tedious. It is situations where you muster all your
strength against one AI only to have another one fall you in the back that
make this kind of games interesting. Even if the second AI is no more than
a petty annoyance, its untimely attack can frequently derail your battle
plan.

>So rewrite the resignation algorithm, sure.  Maybe something based on
>number of can_build units would work?  Or even offensive_worth and
>defensive_worth (although it would be hard to deal with whether units
>are occupants and thus can't be directly seen).  Just make it harder
>(or impossible) to resign?  Doesn't seem like an improvement, IMHO.

I think the best solution might be to enable the ai-may-resign setup option
in all games, but still leave it on by default. This way, you can always
play the game the way you prefer. There are some other options too, that
really should be enabled (but not necessarily set to true) in all games,
e.g. world-seen and sequential. Perhaps also see-all. If I have time, I
will go through all game modules before the release and fix things like
this.

Hans

Hans Ronne

hronne@pp.sbbs.se