So, I've started (at last) a 3d game with models instead of thousands of sprites. Oh my word it's so much easier! I wish I'd started this a couple of years ago.
But, all the matrices take a bit of getting used to. Camera position, where it's looking, clipping etc can be a bit mind bending, for my little brain anyway.
I'm just wondering how best to approach camera position in a character based game where the character is moving. The best analogy I can think of is a star field like the old Windows screen saver. When animating that with 3d models, would you move the camera through the field creating new stars at an ever increasing distance, or would you keep the camera still and move everything past it?
One approach obviously needs a lot more updating. Would you use one approach for a star field, but another for a scenery based game like a fps?
My testing so far has only been with top down style staging so z is only used for effects while all collision detection takes place on z=0. I'm just trying to understand the best practices before going hell for leather off in the wrong direction!!