julien-lebot / ict207-project Goto Github PK

The project for OS X is missing, Shay's code runs fine on OS X 10.5.7, so a project file for it should be committed.
New code should reflect this duality, when putting platform dependent code, let's not forget to include it with #ifdefs.

A decoupling between the window and the input system is required for clarity.
The Model View Controller or the Delegate pattern should apply.

If you would like to work on this issue, put a comment with your name, and your proposed solution (if different from the one described in this issue).
This issue will be opened for 1 week before it is closed for resolution.

We have a console (actually multiple broken implementations).
The console works in a CLI but not with OpenGL at the moment. This issue aims at solving that problem.
The console's issue will require the issue number 11 to be solved first, no solution will be accepted before solution 11 is closed.

One will need to integrate the existing console code with the 2D GUI and input code.
If you would like to work on this issue, put a comment with your name, and your proposed solution (if different from the one described in this issue).
This issue will be opened for 3 weeks before it is closed for resolution.

We need a particle system. I have some sample code involving shaders (particle system on the GPU).
This is very important as we'll use particles for the guns (amongst other things).

If you would like to work on this issue, put a comment with your name, and your proposed solution (if different from the one described in this issue).
This issue will be opened for 2 weeks before it is closed for resolution.

Our planet currently displays lots of artifacts.
It is not certain what is the source of the problem, but it might or might not be to do with the shaders.
There is no current proposed solution to that problem.

If you would like to work on this issue, put a comment with your name, and your proposed solution (if different from the one described in this issue).
This issue will be opened for 2 weeks before it is closed for resolution.

The Design Document for this project is non-existent, this should be fixed ASAP.

We need a GUI interface, both in 2D (overlays) and 3D (3D cockpit).
The interface should be able to display some simple forms (rectangles, circles etc...) as well as the more complex 3D cockpit and its HUD.
Ideally the interface will be loaded from a file. (.gui ?)

If you would like to work on this issue, put a comment with your name, and your proposed solution (if different from the one described in this issue).
This issue will be opened for 2.5 week before it is closed for resolution.

We don't have sound yet. We decided to use OpenAL, and I have lots of sample code for that.
The ideal sound library will use OpenAL positional capabilities, play PPM and OGG vorbis sound, include a background track and effect class.
NB: The sound library should very well be multi threaded especially if decoding OGG vorbis.

If you would like to work on this issue, put a comment with your name, and your proposed solution (if different from the one described in this issue).
This issue will be opened for 2.5 weeks before it is closed for resolution.

Frustum culling will be required some time soon as we'll be displaying a lot of triangles on screen.

If you would like to work on this issue, put a comment with your name, and your proposed solution (if different from the one described in this issue).
This issue will be opened for 3 weeks before it is closed for resolution.

The camera should implement some form of LOD along with the models.
This is especially important as we'll have lots of far-away shots.
It will also help to have high res close up shots for the models and lots of low res models in the background.

If you would like to work on this issue, put a comment with your name, and your proposed solution (if different from the one described in this issue).
This issue will be opened for 2 weeks before it is closed for resolution.

We don't have any physics nor networking for now.
Physics and networking are very tied together as this website mentions:
http://gafferongames.com/

So this issue will cover both topics: one need to implement as a first step the RK4 integrator and client side physics described in the article. (DO NOT just copy and paste the guy's sample code as we'll require lots of modifications compared to his implementation)
Then as a second phase one will need to implement the server and the server side physics.

If you would like to work on this issue, put a comment with your name, and your proposed solution (if different from the one described in this issue).
This issue will be opened for 3.5 weeks before it is closed for resolution.

At the moment it is not possible to display a cube map / skybox, but some ground work has been done toward that goal.
It should be just as simple as looping through a list of files and loading them with the appropriate texture target.
A better solution (on top of the previous one), would be to load the cube maps from a DDS files, so modification to the current DDS loader would be allowed and required.

If you would like to work on this issue, put a comment with your name, and your proposed solution (if different from the one described in this issue).
This issue will be opened for 1.5 week before it is closed for resolution.

SDL is used in Shay's code just for the sound, it can be replaced by our own sound engine, and really we do not want a depency on such a huge library just for sound.
Changes to Shay's code that do not include the sound part should be made soonish.

Shay's world textures conflict with our own textures.
Therefore for testing, the displaying of Shay's world has been disabled and it must be restored when this issue is closed.
The proposed solution is to replace all Shay's world texture handling with our texture manager.

If you would like to work on this issue, put a comment with your name, and your proposed solution (if different from the one described in this issue).
This issue will be opened for 1 week before it is closed for resolution.

There is absolutely no animation in any of our models, nor is it described in the file format.
A remedy for that should be found so as to permit some form of interactivity; static worlds ain't fun.

If you would like to work on this issue, put a comment with your name, and your proposed solution (if different from the one described in this issue).
This issue will be opened for 3 weeks before it is closed for resolution.

Unit testing shows that hand coded SSE code causes memory access violation, intrinsic asm works fine on the other hand, so it's not about memory alignment.
Intrinsic asm takes precedence over hand coded SSE (as it is coded anyway), but it should be considered to fix it eventually.