3d models and a top down orthographic camera
I want to use 3d models because I think that it would be less work than 2d sprites. My artistic skills are pretty shit, so I’d imagine that it’s easier to make and animate a single 3d model than it is to draw a decent sprite from every point of view. Keeping in mind that the gameplay is going to be completely 2d, how do I make a 3d model seem like it’s being looked at slightly from the side, while maintaining 2d gameplay? The simplest answer seems to be to displace the vertices of the mesh on the Y axis the higher they are off the ground.
This creates another issue where it’s very difficult to tell how close spheres / round objects are to the player. Other shapes aren’t so bad. In my opinion, there needs to be clear indicators of when the player will be hit by something and when he will not.
The current solution that I have in mind
I will keep all bullets and projectiles at a specific height. These sorts of projectiles will almost always be round. They will only be able to collide with the player on a smaller hitbox that is highlighted on screen. (Borrowing from Touhou type games.) I think these will always be sprites.
Other sorts of obstacles, such as spikes on the ground or rolling spiked columns, will collide with the player at feet level. I will try to have these obstacles cast shadows and somehow make it obvious that they are on the ground. Perhaps their lighting will be more standard and blending in with the level. The bullet sprites will probably be unlit and quite bright.