Path: utzoo!utgpu!news-server.csri.toronto.edu!rutgers!usc!rpi!uupsi!eye!erich
From: erich@eye.com (Eric Haines)
Newsgroups: comp.graphics
Subject: Re: Distributed Ray Tracing
Summary: Here are a bunch of references
Keywords: ray tracing, distributed
Message-ID: <1991Mar11.152453.1194@eye.com>
Date: 11 Mar 91 15:24:53 GMT
References: <6678@celery34.UUCP>
Sender: Eric Haines
Reply-To: erich@eye.com (Eric Haines)
Organization: 3D/Eye Inc., Ithaca, NY
Lines: 111

In article <6678@celery34.UUCP> stephens@motcid.UUCP (Kurt Stephens) writes:
>
>	Most implementations (I've seen or heard or) of distributed ray
>tracing, distribute the object database between multiple processors and
>each processor computes a subset of the image.
>	Has any research been done on distributed ray tracers
>which each processor handles the ray-object intersections for
>subset of the database, while a master processor (or group of
>master processors) computes the image, by sending messages to the
>appropriate object processor, e.g. "intersect ray X with object Y"?

Check out:

%A John Salmon
%A Jeffrey Goldsmith
%T A Hypercube Ray-Tracer
%J Proceedings of the Third Conference on Hypercube Computers and Applications
%D 1988

They have an interesting method of taking a database which has been put in a
hierarchical bounding box efficiency scheme and splitting up the hierarchy
for efficient ray tracing.  They also talk about image decomposition.

%A Mark E. Dippe
%A John Swensen
%T An Adaptive Subdivision Algorithm and Parallel Architecture
for Realistic Image Synthesis
%J Computer Graphics
(SIGGRAPH '84 Proceedings)
%V 18
%N 3
%D July 1984
%P 149-158
%O also in Tutorial: Computer Graphics Hardware: Image Generation and Display,
Computer Society Press, Washington, 1988, p. 285-290.

They propose an algorithm for adaptive load distribution by splitting up the
database into a grid of hexahedra and adjusting the boundaries of these
hexahedra when bottlenecks in ray passing are detected.

%A Keiji Nemoto
%A Takao Omachi
%T An Adaptive Subdivision by Sliding Boundary Surfaces for Fast Ray Tracing
%J Proceedings of Graphics Interface '86
%I Canadian Information Processing Society
%C Toronto, Ontario
%D May 1986
%P 43-48
%Z adaptive subdivision algorithm on a parallel architecture

These take Dippe & Swensen's ideas, but use regular boxes instead of hexahedra
for efficiency.

Hmmm, there are actually quite a few papers on this area of research.  Give
Tom Wilson's ray tracing abstracts a look.

I just remembered, there's a nice overview of this topic.  The following
related references are from an article in the hardcopy Ray Tracing News (which
I wish was reprinted somewhere else) by David Jevans:

%A Hiroaki Kobayashi
%A Tadao Nakamura
%A Yoshiharu Shigei
%T Parallel Processing of an Object Space for Image Synthesis Using Ray Tracing
%J The Visual Computer
%V 3
%N 1
%D February 1987
%P 13-22

%A Isaac D. Scherson
%A Elisha Caspary
%T Multiprocessing for Ray Tracing: a Hierarchical Self-balancing Approach
%J The Visual Computer
%V 4
%N 4
%D October 1988
%P 188-196

%A Andrew Pearce
%T An Implementation of Ray Tracing Using Multiprocessor and Spatial
Subdivision
%R Master's Thesis
%I University of Calgary, Dept. of Computer Science
%D 1987

%A Hiroaki Kobayashi
%A Satoshi Nishimura
%A Hideyuki Kubota
%A Tadao Nakamura
%A Yoshiharu Shigei
%T Load Balancing Strategies for a Parallel Ray-Tracing System Based on
Constant Subdivision
%J The Visual Computer
%V 4
%N 4
%D October 1988
%P 197-209

%A Hiroaki Kobayashi
%A Satoshi Nishimura
%T Parallel Architecture for Fast Image Synthesis Under Dynamic Environments
%B Proceedings of Computer Graphics International '89
%I Springer-Verlag
%D 1989

[does anyone have the page numbers and editors of this last reference?]

There are probably more I've missed.

Eric Haines, erich@eye.com