I have to choose a project to work on for the final year of my CS degree. I would have about 5 months to work on it. I'd love to try to find something that would let me learn more about 3D game development. I've been reading about shader programming for the last couple of weeks, but I can't think of any concrete project to work towards, and I'm already finding it difficult to understand even the fundamental techniques. Experience/skill-level: I've only ever programmed basic 2D games like Pong, Tetris clones, and part of a platform game. Please, could anyone suggest any ideas for a project that could be accomplished in this time period? I already asked at college about making a 3D game, but I was advised that it would probably be too much work for a one-man project (The project is just one fairly small component of the final grade) Although I've already mentioned that I'm interested in learning about graphics programming, I'd welcome any suggestions about any area of game dev! I'd really appreciate any suggestions! (I'm so stressed about this...)

Accurate modeling of sound propagation in an environment on a GPU [think ray-tracing]?

Geometry optimization for GPU-based visibility queries?

Procedurally generating plants/terrain on a GPU?

A full-on 3d game is do-able, but it won't likely be academically interesting [at least not within that time frame, as it'll likely have to be pretty bare-bones]. There are tons of things to do if you like graphics programming though. All the above fits well within that category.

Quote:Original post by Drigovas
A full-on 3d game is do-able, but it won't likely be academically interesting [at least not within that time frame, as it'll likely have to be pretty bare-bones]. There are tons of things to do if you like graphics programming though. All the above fits well within that category.

I fully agree. Depending on who is going to be grading your effort, you may want to stay away from doing an actual 3D game. They have seen many a student make games for CS degree finals, and it is likely to not interest them. They are likely to give higher scores for applications that do some kind of work, or prove a point. Remember, this is academia land we are talking about.

EDIT: If you have your heart set on making a game, definitely talk to your instructor first. There is a chance they may nix it from the start.

Quote:
EDIT: If you have your heart set on making a game, definitely talk to your instructor first. There is a chance they may nix it from the start.

I agree; the given time frame will be extremely tight for a game; tighter than you think. Especially if you want to do 3D and have never done so before.

Thanks a lot for your replies. As several of you have advised me, I don't think trying to make a 3D game would work out too well, I could imagine spending weeks tweaking game mechanics and gathering/tweaking content, which wouldn't really be relevant to CS...

Drigovas' suggestion of "Procedurally generating plants/terrain on a GPU" interests me, but I'd be worried that I wouldn't have any new ideas to try out, and just try to implement existing techniques instead.

I wouldn't really know where to start with the other suggestions as I'm not familiar with ray-tracing, but I'll read more about them.

I really appreciate your helpful responses, thanks!