Complete.Org: Mailing Lists: Archives: freeciv-dev: June 2003:
[Freeciv-Dev] Re: (PR#3387) Path finding and ZOC
Home

[Freeciv-Dev] Re: (PR#3387) Path finding and ZOC

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: undisclosed-recipients: ;
Subject: [Freeciv-Dev] Re: (PR#3387) Path finding and ZOC
From: "ChrisK@xxxxxxxx" <ChrisK@xxxxxxxx>
Date: Mon, 2 Jun 2003 05:12:09 -0700
Reply-to: rt@xxxxxxxxxxxxxx

On Mon, Jun 02, 2003 at 04:51:28AM -0700, Gregory Berkolaiko wrote:
> 
> On Sun, 1 Jun 2003, ChrisK@xxxxxxxx wrote:
> 
> > On Wed, Feb 12, 2003 at 11:36:44AM -0800, Christian Knoke wrote:
> > 
> > I think Pathfinding should make reasonable assumptions about reality to
> > yield good results. So, yes, IMHO it should remember former city locations
> > and use it, since cities do not vanish that often.
> 
> Client remembers cities' locations (because you can see them displayed, 
> right), in a sort of dummy city struct.  I think it should also remember 
> the occupied flag in such cities.

I'm not sure with this. At least the AI leaves its cities often unsentried in
the beginning. Then you see the empty cities with your ships. Later on, when
we attack them, the AI puts a unit in them.

We don't want the truth. We just want to guess right most often. And when we
guess wrong (assuming the city is occupied but it is not) the harm is
smaller as if it goes the other way round (and our unit is blocked by ZOC).

Christian

-- 
Christian Knoke     * * *      http://www.enter.de/~c.knoke/
* * * * * * * * *  Ceterum censeo Microsoft esse dividendum.



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