Complete.Org: Mailing Lists: Archives: freeciv-ai: March 2004:
[freeciv-ai] Re: [Freeciv-Dev] Discusion about ai client side
Home

[freeciv-ai] Re: [Freeciv-Dev] Discusion about ai client side

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: Guillermo Lopez Alejos <100030179@xxxxxxxxxxxxxxx>
Cc: freeciv-dev@xxxxxxxxxxx, freeciv-ai@xxxxxxxxxxx
Subject: [freeciv-ai] Re: [Freeciv-Dev] Discusion about ai client side
From: Raimar Falke <i-freeciv-lists@xxxxxxxxxxxxx>
Date: Wed, 10 Mar 2004 20:16:32 +0100

[ Your mailer sends weird chars for begin-quote and end-quote. ]

On Tue, Mar 09, 2004 at 08:22:04PM +0100, Guillermo Lopez Alejos wrote:
> Hi,
> 
> I'll try to explain myself.
> 
> First diagram "Before the change (now)" or current architecture
> 
> This diagram is an abstraction of the current FreeCiv architecture
> (well... the architecture I think it has).
> AI players are built in the server code (so they can access a lot of
> information they must not have).
> The "Client" block in the diagram is the "core client".
> Human players see what is happening in the game and input their commands
> trough the "User Interface".
> 
> Second diagram "After the change"
> 
> When I say "change" I mean how the game architecture is going to look
> like when I finish coding. 

> My main objective is to prepare the game to be able to receive
> orders from human players or from other programs (this would allow
> the replacement "on the fly" of an leaving human player with an AI
> program). I'll have to define a data interchange interface to allow
> AI programs to communicate with the "core client".  The way I
> planned to do this is putting a layer (or "wrapper", but forget this
> "wrapper") onto the "core client".

People did this (adding an interface for an external program)
before. Please take a look at the archives.

> But now I'm starting to understand what agents are; pieces of code
> that perform tasks. Well, now I need to know more about agents,
> specially where the real "intelligence" is going to go.

The real "intelligence" is emergent. (at least this is my idea/hope
here) Or to put it into other words it is distribute across all the
agents. If you look at the server AI (in ai/*) you will also notice
that there is no single real "intelligence" somewhere. At least I
didn't found it.

> It's vital to my particular project (final career project) to understand
> exactly how things works now, to know what the developers (all of you :)
> has in mind and vice versa.

We have a lot in mind but not so much time to code it all ;)

        Raimar

-- 
 email: rf13@xxxxxxxxxxxxxxxxx
 "This is Linux Country. On a quiet night, you can hear Windows reboot."


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