Home - nicobabot/Multi-cameraculling GitHub Wiki

Multi-camera culling Research

Introduction to camera culling:

• The camera culling is a code optimization, that in the moment we render on the screen, all the elements that are out of the camera aren't printed.
• In our case, multi-camera culling, our objective will be achieving that optimization for when we have more than one camera.

Things to take in account before going into coding:

• Cámara: (iPoint/SDL_Rect) it tell us the point/proportion of the map that we want to show (coordinates are respect the MAP).

• Viewport: (SDL_Rect) it tell us wich portion of the screen we want to show when we render. (coordinates are respect the SCREEN).

• The viewport doesn't prevent the things to be out if his rect to be rendered, only doesn't show it to you (Bliting still consuming resources).

• In our code the axis of positioning is:

-In the case of the cameras is:

Steps to follow for testing the camera culling.

• Create 2 or more cameras.
• Create a viewport for each one.
• Adapt your Blit function a way that discards all the elements that are out of the camera when printing.
• In the case you are using a map created with tiles, adapt aswell the function that draws all the tiles (exclude all the calculations for the tiles outside the camera).
• With a timer or simply observing the dt, compare the final result with the initial one.