Complete.Org: Mailing Lists: Archives: freeciv-dev: June 1999:
Re: [Freeciv-Dev] easy AI is too hard (PR#11)
Home

Re: [Freeciv-Dev] easy AI is too hard (PR#11)

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: beraud <berauda@xxxxxxxxxxxxxxxxxxxx>
Cc: freeciv-dev@xxxxxxxxxxx
Subject: Re: [Freeciv-Dev] easy AI is too hard (PR#11)
From: Peter Schaefer <schaefer@xxxxxx>
Date: Fri, 04 Jun 1999 17:23:50 +0200

beraud wrote:
> 
> Children, let's listen religiously to a professionnal Freeciv looser.
> I mean.... ME. How many times did I tell you tha intelligence on the
> AI is NOT the problem. Even in easy mode, the AI shall not be an idiot.
> But what makes it unbeatable by long time beginners like me (played a
> lot of Civ and CivII games in Chieftain mode, about 30 Freeciv games
> and am still a really bad player. So you can trust me) is it's
> horrible research speed. 
> [snip] 
> It would be a real progress if something could be done at last !
> 
> Alex


Ok, let's try to listen:

One good point that has been made that there is no peaceful AI,
and that some of the AI players should wish for a peaceful victory
(spaceship). Peaceful AI would also conduct diplomacy
(and give / take money).

How about handling the fast researching AI this way:
Easy ai 'cheats' and looks up the total researchpoints by 
the other players, as compared to it's own. If the AI is ahead by more that
2 technologies, it raises tay rate and saves money.

Now what to do with the surplus money ? Maybe have the AI give money to
all other players ( With appropriate messages ).

Think you can do that, Alex ?
-- 
Peter Schaefer - schaefer@xxxxxx, schaefer@xxxxxx
"FermaT, by Software Migrations Ltd, is an industrial strength program
transformation system used to migrate IBM 370 Assembler modules into 
equivalent readable and maintainable COBOL programs, and to help solve 
the year 2000 challenge for IBM 370 Assembler code."

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