Complete.Org: Mailing Lists: Archives: freeciv-dev: September 2001:
[Freeciv-Dev] Re: find_the_shortest_path()
Home

[Freeciv-Dev] Re: find_the_shortest_path()

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: rf13@xxxxxxxxxxxxxxxxxxxxxx
Cc: Gregory Berkolaiko <gberkolaiko@xxxxxxxxxxx>, freeciv-dev@xxxxxxxxxxx
Subject: [Freeciv-Dev] Re: find_the_shortest_path()
From: "Ross W. Wetmore" <rwetmore@xxxxxxxxxxxx>
Date: Tue, 18 Sep 2001 20:31:43 -0400

It is actually poor gaming strategy to make algorithms too rigorously
correct at doing non-critical tasks.

One of the biggest problems with AIs is that they tend to repeat 
themselves obsessively, and such predictability both gets them into
deadends and makes playing against them pretty boring.

Just a warning against things like trying to overload a function to
do everything possible exactly right ... not always a smart move :-).

Cheers,
RossW
=====

At 02:37 PM 01/09/18 +0200, Raimar Falke wrote:
>On Tue, Sep 18, 2001 at 11:34:30AM +0100, Gregory Berkolaiko wrote:
>> It is needed and it is used in do_unit_goto / find_a_direction
>> But it all can be improved, e.g. you can recalculate path if you uncover
>> (change) in the territory or an enemy unit...
>
>Since find_a_direction also calculates a desirability/fitness/negative
>cost it is possible to include this in the main algorithm (as a
>callback) and have so only one best path.
>
>       Raimar
>-- 
> email: rf13@xxxxxxxxxxxxxxxxx
> "I haven't lost my mind - it's backed up on tape somewhere."




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