Algorithm to only draw what the camera sees?

Question

I'm making a 3D FPS with OpenGL and here is the basics of how it works. The game is a 3D array of cubes. I know the location of the player's current cube, aswell as the camera x,y,z and I know the x, y, z rotation of the camera too. Right now I just make a square around the player and render this and then add distant fog. The problem though, is that I'm still rendering everything that the player is in back of. How could I selectively only render what the player sees, not render everything within an X radius as Iam doing now.

Thanks

Answer 1

Are the cubes rooms of a maze through which the player navigates? If so, and assuming the rooms are static over the course of the game, you could use a BSP tree to traverse the scene in order of depth, stopping when you pass the player.

Answer 2

This sounds like you need to look into culling concepts.

Answer 3

You are talking about frustum culling, if i get you right. I suggest that you take a look at this tutorial. They provide nice demos and explain everything in detail.