High performance parallel rendering on the PC platform

• Main Author: Wong, Chee Kien.
• Other Authors: Wang, Jianliang
• Format: Theses and Dissertations
• Published: 2008
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
• Online Access: http://hdl.handle.net/10356/3740
• Institution: Nanyang Technological University

The recent availability of high-performance, low-cost commodity graphics hardware has brought about a shift in parallel rendering paradigms. Coupled with advancement in semiconductor fabrication technology, parallel rendering is not longer implemented only in expensive and specialized multiprocessor systems but on the desktop computer platform as well. To date, a $500 piece of PC graphics hardware (GeForce4Ti) [7] is already capable of achieving a peak performance of processing 146 million polygons per second. Our research addresses the suitability of the PC as a platform for parallel rendering and describes the actual implementation on the PC using off-the-shelf graphics components. The design of the corresponding load balancing algorithm and real-time rendering framework are also provided in detail in this thesis.