|
|
This work is licensed by Raphaël Moll under a Creative Commons License.
|
|
| Color Theme: | Gray | Blue | Black | Sand | Khaki | Egg | None |
|
|
Visited 9 times by Google, last 2012/05/19 10:10
Visited 22 times by Yahoo!, last 2011/10/26 01:05
|
|
|
|
Index: Home | What Is Izumi | Misc Links | Random Thoughts | Too Much To Read | The Rant Vault | Quotes Dev: Projects | Ideas For Dev | Nerdkill | Rig | Hint
$Id: NerdkillIII.izu,v 1.1 2005/11/05 21:55:57 ralf Exp $
20050415
The goal of this project is to perform a real live test on OpenGL vs DirectX 3D. As such the purpose here is more to learn than to produce a real game. The focus is obvious on technology, API and design patterns rather than on gameplay. We're not talking about anything close to a commercial-grade game here.
There is no guarantee that this project will actually be finished (assuming it's actually started first) nor that it will be publicly available in any form at any point. No guarantee means it may happens or may more likely not happen.
As explained in Net Droids Notes, it is obvious that
if one wanted to release an actual game, the efforts should obviously
focus on the game design first (gameplay, etc.) and then implement
using existing frameworks.
On the other hand, from a learning point of view it seems more interesting
to me to focus on tackling with a minimal implementation of the various
components needed for such a game, including but not limited to rendering,
scene management, interactivity and so forth.
20050512
If you expect anything meaningful from reading this page, you came to the wrong place. This is a "research" work that is not expected to actually end in any tangible implementation. Think of it more as a scrapbook of random notes centered on a given theme.
Related project: NetDroids Notes
(Notes are given in chronological order.)
A number of high level modules are obvious:
That's the bulk of it. What is missing?
Obviously my target is a Windows host using .Net 1.1. The implementation should have modules that abstract the underlying APIs. That is encapsulate all API-specific stuff in modules or external libraries then base a portable framework on top of that. The Nerdkill module design worked well for that.
Ideally a number of derived ports could be interesting, such as MacOS X or Linux with OpenGL, OpenAL, Mono and maybe GTK# or Cocoa. I won't actually go there anyway but I should not make the framework too dependant on a single host. This is why I'm interested in trying to support at least a couple of APIs in the first place such as OpenAL vs DirectSound or OpenGL vs DirectX 3D.
|
|
This work is licensed by Raphaël Moll under a Creative Commons License.
|
|
| Color Theme: | Gray | Blue | Black | Sand | Khaki | Egg | None |
|
|
|
|