Array control for high-performance SIMD systems
Journal of Parallel and Distributed Computing
| Hi-index | 0.00 |
High performance graphics subsystems play a critical role in many computer systems, but result in high priced systems that often still fall short of the required graphics performance. The root of the problem lies in the complex graphics algorithms that require large amounts of object data to be manipulated with high throughput, such as what is required for ray tracing. This has resulted in complex, real-time 3D graphics to be limited to high-end systems. In this paper, we present a library of 3D graphics algorithms that have been mapped to the Micro-Grain Array Processor (MGAP), an inexpensive and versatile SIMD processing board capable of fitting in a typical workstation. Our results show that the MGAP can produce comparable data throughput as more costly graphics subsystems, while maintaining the flexibility of being a general purpose parallel machine.