Complete.Org: Mailing Lists: Archives: freeciv-ai: September 2003:
[freeciv-ai] Re: (PR#5633) Order of AI activities.

[freeciv-ai] Re: (PR#5633) Order of AI activities.

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: per@xxxxxxxxxxx
Subject: [freeciv-ai] Re: (PR#5633) Order of AI activities.
From: "Christian Knoke" <chrisk@xxxxxxxxx>
Date: Sat, 13 Sep 2003 07:18:58 -0700
Reply-to: rt@xxxxxxxxxxxxxx

On Fri, Sep 12, 2003 at 02:12:02PM -0700, Gregory Berkolaiko wrote:
> On Fri, 12 Sep 2003, Christian Knoke wrote:

> And this is what my patch actually does :)
> > But it is possible to annoy the AI by occupying worked tiles.
> > And it is possible to shoot an AI city empty and enter with goto next turn.
> No it is not, since the AI will buy a new unit and it will grow in the 
> city at the turn change.  Well, a good AI will do that.

Buying units, the AI will run out of money quickly. Mayby it should be able
to issue Goto commands just before end of turn, which are then executed at
the beginning of next turn?

We then revive the problem of double moves, but not really:

Goto's need to be executed in random order. Then, since Goto's and moves are
resolved in the same turn, there won't be such a double move effect. Both,
AI and clients, get a fair chance. When the AI attacks from sea, the client
player still can: move a unit, buy a unit, issue a goto (executed next turn)
with a fair chance that his Goto is executed before the attacking AI's Goto.

But maybe this approach is too complex.


Christian Knoke          * * *
* * * * * * * * *  Ceterum censeo Microsoft esse dividendum.

[Prev in Thread] Current Thread [Next in Thread]