Can someone explain the process in lamens terms. I''m no mathimatician. I do know what normals are and their purpose. I understand what the cross product and dot product''s are used for. Yet I''m confused on how to tie it all together to get the results I need. I have a game. I load a mesh consiting of quads. Its translated to a location. I know I need to go through the the individual faces of the mesh and grab at most 3 vertices to create a normal. Then I take the dot product of the normal and the camera''s locaiton. I think. Actaully I''m not sure. Questions: 1.Do I need to add the actually location of the mesh to the normal? 2. Do i need a normal for each face? The mesh is at x:0.0 y:1.5 z:-18.0 so it''s approximately 19 or so units away. I just need something simple to know I''m close to it. Eventually I''ll add walls and creat a maze. I''ll need to know when I''m hitting a wall. What I''m essentially asking is what are the steps to collision detection?

http://www.gamasutra.com/features/20020118/vandenhuevel_01.htm

very good article, in a matter of hrs I have manage to get some decent detection going on. When I add some walls this is going to get complicated.