Complete.Org: Mailing Lists: Archives: freeciv: February 2000:
[Freeciv] Re: [Freeciv-Dev] A questionnaire to the mailing list
Home

[Freeciv] Re: [Freeciv-Dev] A questionnaire to the mailing list

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: ludo <ludovic.crapet@xxxxxxxxxxxxxxxx>
Cc: freeciv@xxxxxxxxxxx
Subject: [Freeciv] Re: [Freeciv-Dev] A questionnaire to the mailing list
From: Nicolas BRUNEL <brunel@xxxxxxxxxxxxxxxxxxxx>
Date: Wed, 23 Feb 2000 19:49:05 +0000 (GMT)

hello and be welcome,

   I have attached the answer Claus Leth Gregersen wrote to my mail. 
I wish you to success in your studies.

bye,    


Hi.
Well Peter Unold and i have always been interested in game development.
At the time we wanted to make a commercial multiplayer game.

Before we started working on our game we wanted to test some ideas in
practise.

At the same time i found something called openciv,
it looked quite interesting but after several hours of trying to compile it
i gave up.
(All those libraries, and i didn't have root permission, it was a  real
pain)

So all of a sudden the idea of doing a basic civclone was born.
Our commercial game was intended to be written in c++,
but at the time c++ compilers for unix wasn't that great, and the set of
features
generally supported wasn't very big so we decided to do it in plain c.

We wanted to test somethings yes, but at the same time we wanted to do
something
that was generally playable on any unix workstation, and since we already
had some
experience with releasing c code for unix (Peter was the original author of
a mud client
called TINTIN,  which was pretty popular back in the early 90's), we decided
to stick with
plain c and plain rough ugly athena/xaw.
I still think it was the right choice, i firmly freeciv would have died with
us,
if it has been written in any other language, or with any other widget set.

Time has changed gtk/kde are both quite good solutions today,
and  especially writing in c++ would be just fine aswell.

The original plan was just to make a _very_ basic civ. map/cities/units we
were most interested
in how it would work, how to overcome problems like turning basically modal
windows like
tech selection into something that wouldn't block the game.
Some considerations about how much bandwidth we could consume without
killing the game etc...

A couple of weeks had passed, and we actually had something running.
We were quite satisfied with what we had build, and we decided to work some
more on it.
And we convinced one of our friends Allan to do the graphics and the
encyclopedia.
Problem all along was that we never did design anything except the list
structure, as we didn't
think we would work on it for much time. So we redesigned some of the edges
as we went on
but some basic structural problems remains.
But i'm not certain which parts could be done much clearer in c anyway.

horrible things in freeciv:

The package sending/receiving routines are very awful, you have to change
6-7 different places
to add a new package or adjust an existing one.

client/server architecture:

Client and server should be 1 executeable, especially now we have the 1
player + AI mode.
This way AI could be done without working on the structures directly, but
through the same
interface as the client. (the idea of client side AI as people refer to it
today i think is very wrong.)

And everything should be rewritten to c++ ;)

We didn't base it on civnet or anything, just on a
civ1 manual + civ2 manual + civ2 tips'n'tricks book i bought .

At some point in the development cycle, it turned from actually developing
and became
bugfix after bugfix and  replying to clueless people who was trying to
compile a program for their first time.
And i guess we grew very weary and very tired of it. (and people like Manuel
Gutierrez Algaba).

I was positve surprised to see that it actually succeded as a project after
we ran screaming away.
For a couple of months it looked like the development was dead, but then
things started rolling.
Today must say that freeciv looks great, not many bugs left, and not many
missing features.
I am really glad that people has generally stuck with the original rules, i
couldn't imagine
anything more horrible than introducing cyberninjas macdonalds wonder or
flying penguins..

I haven't understood why noone has done the isometric view, take a look at
great nations
it looks really good and their old look was not very nice, rotating the
tiles 45 degrees does wonders.
I think that and the lack of sound is what scares some people away now.
There is no doubt that freeciv
is the best multiplayer civ that exists, but still i put my money on civ II
in single player:
The AI in freeciv might be better,
but the AI in civ2 plays more "civilized" and each race got unique traits
(i guess culture is when a nation does a stupid thing, even though it knows
it's a stupid thing.)


/Claus Leth Gregersen.









[Prev in Thread] Current Thread [Next in Thread]
  • [Freeciv] Re: [Freeciv-Dev] A questionnaire to the mailing list, Nicolas BRUNEL <=