Quote: "If the cubes are in a grid, you can precalculate which cubes will be hidden if a certain cube relative to the camera exists. You'd only do this for cubes close to the camera.
Basically you'd have a 3d grid in memory for each grid cell near the camera. You fill these grids before-hand with either 0 or 1 depending on whether they are hidden if the grid cell is solid. In the actual game, you take the grid for each solid cell and cull all the grid cells set to 1 in the grid."
This is really specific to what I'm doing.
So to bring in some additional information: All cubes are the same size and orientation. They are in a 3D grid as Diggsey describes where some may be filled and some may be unfilled (and thus transparent). This doesn't need to be precise, so it doesn't matter if less than 100% of the cubes that are obscured are culled.
@ Diggsey
So are you saying that we can do most of the work just by looking at the player's position. I'll prototype this at a 2D level first; if we look for blocks that are filled around us then we can instantly cull large sections like this:
Above the green square is the camera, the grey squares are filled spaces (the one non grey square surrounding the camera is unfilled), the blue squares are squares that might be visible to the camera and so should not be culled, the red squares are squares that can be culled.
Quote: "On a side note: I'm surprised you're asking all these questions Michael P - you made DarkNet xD - You seem like you should be able to do all this better than the rest of us"
DarkNet is a different ball game, there's no mathematics or 3D stuff in there. I'm very inexperienced with both these things hence the questions.
