I've been working on a new game for a while now, but it's a bit more experimental than my usual and relies heavily on procedural generation. So much so that I'm not even sure if I can make it work. I've been prototyping purely with text output so far (not roguelike text output, just straight console out) but it's getting to the point where I need to actually put it into a world to get a better feel for how it will eventually play.
Eventually I want to have a proper 3d world (think subversion
), but the world generation isn't something I want to focus on yet. At the moment I seem to have three options:
1. Throw something together with Java2d, with a crude 2d tile world. Advantages: simple, won't get bogged down in graphics programming. Disadvantages: will probably have to rewrite the whole thing when I do the world proper.
2. Start developing the 3d world with a 3d engine (most likely Ardor3d). Advantages: Minimal wasted work, closer to actual final feel. Disadvantages: Slow to develop, learning curve as I figure out Ardor3d. Possibility of being sidetracked by pretty graphics.
3. Do some crude 3d with OpenGL directly. Advantages: Quick to get started. Will only have to throw some code away. Disadvantages: Probably involves a lot of tedious 3d engine ground work, even for a simple world.
For the final game I'll probably be using Ardor3d (or at least some kind of 3d engine) so part of me wants to start with that. But a java2d tile engine would probably be a whole lot quicker and probably keep me focused on the actual gameplay rather than graphics.