Particle motion simulation - a parallel distributed workstation application for real-time

Particle motion simulation subdivides graphical objects into a set of small particles behaving according to specific rules. Real-time particle motion simulation makes it possible to simulate natural phenomena like smoke or Brownian motion for demonstration and verification purposes. The paper describes how the enormous computational requirements can be solved using distributed parallel computation on a set of workstations. 1 . Particle Motion Simulation Particle motion simulation is a concept applied in the areas of scientific visualization and computer graphics. It is an approach to model and render objects that cannot be easily described in terms of surfaces, volume primitives and motion paths. That is because the frontier between the objects and the environment is varying over time or can only be defined statistically, like fire, fog or smoke [1]. Particle motion simulation can also be applied to models in various sciences like the Brownian motion, economics and social behaviour. Particle motion simulation subdivides objects into a set of small particles which behave according to specific rules. These rules can be e.g. Newtonian mechanics, fluid dynamics and others. Particles can be created, can themselves create other particles, can change their behaviour during their lifetime and can die. There can be particle sources, particle drains and an environment modeled by conventional means. Particles can interact with each other and with the environment. The paths of particles can be used to construct higher level objects like trees or grass [2] for example. Real-time particle motion 1 This paper has been accepted for publication as: Sinkwitz R.C., Particle Motion Simulation A Parallel Distributed Workstation Application for Real-Time, Proceedings of the Workshop on Parallel and Distributed Workstation Systems, Elsevier Science Publishers B.V. Amsterdam, 1992 simulation represents a new method for studying and verifying models of different nature. However the computational requirements are generally enormous. An example of real-time computation on a "Connection Machine" is described in [3]. This paper describes an approach to distribute the computing process to a heterogeneous set of workstations.