Parallel processing between GPU and CPU: Concepts in a game architecture

Abstract

This paper presents a new game loop architecture concept that employs graphics processors (GPUs) for general- purpose computation (GPGPU). A critical issue in this field is the concept of process distribution between CPU and GPU. The presented architecture consists in a very efficient model for distribution, and it was mainly designed to support math and physics on the GPU, but any kind of generic computation can be easily adapted.