Re: About GtkGLExt / glChess

[Alan Horkan <horkana maths tcd ie>]

On Thu, 31 Aug 2006, [ISO-8859-1] Andreas Røsdal wrote:

> Date: Thu, 31 Aug 2006 22:44:28 +0200
> From: "[ISO-8859-1] Andreas Røsdal" <andrearo pvv ntnu no>
> To: games-list gnome org
> Subject: Re: About GtkGLExt / glChess
>
> Corey Burger wrote:
> > n 8/31/06, Andreas Røsdal <andrearo pvv ntnu no> wrote:
> >
> >> There have been many good suggestions for new games in gnome games so
> >> far. One of the games that I like is glChess
> >> (http://glchess.sourceforge.net/). It depends upon OpenGl, using the
> >> GtkGLExt library. Note that glChess does not require hardware support to
> >> be playable with OpenGl, it can also be played with software as well.
> >>
> >> OpenGL is perfectly suited for game-development.

Words such as perfect can be very subjective.
SDL seems simpler than using OpenGL directly but I would not be an
expert on the sujbect.

> >> OpenGl would add the extra graphic-appeal to gnome-games that some
> >> people have been missing, and that we possibly could see more games with

Gnome games has kept to a more limited selection of games, cards and
puzzles mostly rather than more dramatic 3D or arcade games.  If people
are expecting even more eyecandy than all the SVG support has allowed it
would seem more like a problem of marketing and managing expecatations.

> >> OpenGl for GNOME in the future. Many Linux distributions already ship
> >> with several OpenGl games by default (eg. Neverball and Tux Racer).

It would certainly be cool if a few of these applications gathered
together into their own collection.  Perhaps gnome-games could encourage
the creation of such a seperate collection?

> >> How do people feel about adding GtkGLExt as a library dependency on
> >> gnome-games, if glChess would be selected as a new game?

It seems like a big step.  I'm wary of any of the proposed new
technologies given the risk it entails.  Using Guile seemed like a good
idea at the time but more developers would always be good.

Gnome at large has adopted Python so it does seem like a relatively safe
choice as far as it goes.

> > I like the idea of OpenGL, but I have big concerns about thin clients
> > and other memory tight environments. Thus, I suggest that glChess have
> > a non-GL mode that is enabled by default, a standard 2D chess board.
> > Create a view menu which allows users to change to OpenGL mode on the
> > fly. Thus we can pimp the new stuff without pissing of our users (and
> > our large deployments, most of whom are on some sort of thin client)
> >
> Take a look at:  http://bobthegnome.blogspot.com/  The main developer,

Having the developer on board is a very good thing and should definately
be one of the first steps in this process.

I appreciate his honesty and sense of humour but note carefully the
following comment:

"This is some of the WORST written code I've seen.
I can understand why you morons ceased working it!"

http://glchess.sourceforge.net/?q=node/21

Given the long term maintainance issues Gnome Games faces are part of the
motivation of getting rid of a game and adding another.  Perhaps some kind
of a code quality review would be advisable before adopting any a game?

> Robert, at least seems very interested in making this work, and has
> suggested a 2D / 3D solution.  :)

Unlike most of the current gnome games I am certain any good chess game
will need to deal with the loading and saving of Game files.  A standard
File menu will definately be needed, and I'm surprised the game doesn't
provide more than two top level menus.  If bob is interested I would be
happy to provide him with more feedback on the intereface irrespective of
how things progress in terms of gnome-games (assuming I can get the game
installed and built but I'm surethink Andreas would not have proposed it
if he had any difficultly building it).


Mise le meas

-- 
Alan Horkan