The visible area determination in Nox looks really great. If you haven''t seen it, get the demo. How do you suppose they have managed it? It seems to be very precise, in the sense that even though I''d venture to guess some columns are only a fraction of a tile wide, and still the shadow edge follows the column''s almost pixel perfectly. So it would have to use some sub-tile level system. And how did they manage to make it so fast? One solution would be to have one cone shaped shadow bitmap for every tile position away from the center of the screen, where the character always is. I''d like to create some discussion about visibility algorithms in tile based games, and this is a good place to start .