Original Post
As in many texts, transparent objects should always be rendered back to front relative to the camera. Consider the following case, however. Let's say we have a large sphere s1 with radius 10 and position (0,0, -10) and a smaller sphere with radius 4 and position (0, 0, -8). Thus, the smaller sphere is inside the larger one. These two spheres have a certain level of transparency in that they are not fully opaque or transparent. Let's mount the camera at (0, 0, 5) looking toward (0, 0, -1). To the camera, the larger sphere's center is further away, so based on back to front method the larger sphere would be rendered first and then the smaller one. Would this create artifacts because the face of the larger sphere closest to the camera is actually closer to the camera than the face of the smaller sphere closest to the camera while the furthest face is further than? Thanks.