Game Development Reference
noted, however, that if the focal length and aspect ratio are known for a particular
view frustum, then the formulas in Table 5.1 provide a significantly more effi-
cient way of calculating normalized frustum planes.
5.6 Reflections and Oblique Clipping
Many scenes contain a reflective surface such as a mirror or a body of water for
which a reflection image needs to be rendered. The typical way in which reflec-
tions are shown in a scene is to establish a separate image buffer called the re-
flection buffer to hold the result of rendering the objects in the scene that are vis-
ible in the reflection. The reflected scene is first rendered into the reflection buff-
er, and then the main scene is rendered into the main image buffer. When the ge-
ometry representing the reflective surface is rendered, colors from the corre-
sponding pixels in the reflection buffer are read and used to contribute to the final
The reflected scene is rendered through a virtual camera that is the reflection
of the main camera through the plane of the reflection, as shown in Figure 5.19.
Figure 5.19. The upper view frustum represents the actual camera rendering a scene that
contains a reflection plane. The virtual camera used to render the reflection is represented
by the lower view frustum, and it is itself a reflection of the upper view frustum. The x
axis points out of the page for both view frustums, and consequently, the camera-space
coordinate system for the camera rendering the reflection is left-handed.
Search Nedrilad ::