hello all, this is not about efficient/fast ray tracing algo. but to understand the math & procedure behind it. i've my view position set @ origin ; and generated my scene by hand computing the orientation (ray from view pos. to poly. edges) of the vectors from origin to polygon edges with a known field of view (camera) and placed the object (basically a square) some distance units away from origin; and now i am able to view the scene; how do i compute the orientation from the view position to any pixel on the screen ? Once i get this orientation, I would try to find the intersection between this ray and the plane in which polygon lies. (i could find the math of this but doesn't know how to compute the orientation for a given a pixel value in the screen). .kavitha