ABSTRACT
High quality computer graphics imagery is used in a wide variety of fields in society today. Most people are familiar with the entertainment uses of computer graphics which span the artistic realm and include two-dimensional imagery using paintbox systems, three-dimensional surreal scenes for aesthetic prints, and 2D and 3D animation sequences for use in the video and film industry. There are many major motion pictures which rely on computer graphics rendering to achieve cost effective special effects. The quality of this imagery has risen to such a high level that the public is accustomed to seeing on a regular basis computer generated commercials of photorealistic caliber. In addition, applications such as CAD/CAM, finite element modeling, flight simulation, and molecular modeling use computer graphics to aid in the visualization of scientific and industrial data. The demand for higher quality images from these applications has grown as computer time has become less expensive. Even though faster computers are now available in reference to the past, the time to generate a typical image has not really decreased due to the more elaborate imagery required. Deering [Deer88] noted that "an increase in graphics performance is more likely to cause users to display more complex objects, rather than the same objects faster." A computer graphics display algorithm must be able to
handle this highly complex imagery in an efficient manner. One solution to this problem involves utilizing parallel computer architectures to render the graphics image. If an efficient software algorithm is employed on this type of machine, performance will increase with the number of processors added to the system.
