Online-Bibliothek

This book is the result of the research in the implementation of polygon-based graphics operations on certain general purpose parallel processors; the aim is to provide a speed-up over sequential implementations of the graphics operations concerned, and the resulting software can be viewed as a subset of the application suites of the relevant parallel machines. A literature review and a brief description of the architectures considered give an introduction into the field. Most algorithms are consistently presented in an informally defined extension of the Occam language which includes Single Instruction Multiple Data stream (SIMD) data types and operations on them. Original methods for polygon rendering - including the operations of filling, hidden surface elimination and smooth shading - are presented for SIMD architectures like the DAP and for a dual-paradigm (SIMD-MIMD) machine constructed out of a DAP-like processor array and a transputer network. Polygon clipping algorithms for both transputer and the DAP are described and contrasted. Apart from the information presented in the book and the useful literature survey, the reader can also expect to gain an insight into the programming of the relevant parallel machines.