Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Posting-Version: version B 2.10.2 9/18/84; site lanl.ARPA
Path: utzoo!watmath!clyde!cbosgd!ihnp4!mhuxn!mhuxr!mhuxt!houxm!mtuxo!mtunh!mtung!mtunf!ariel!vax135!timeinc!phri!pesnta!amd!vecpyr!lll-crg!gymble!umcp-cs!seismo!cmcl2!lanl!rmc
From: rmc@lanl.ARPA
Newsgroups: net.graphics
Subject: Re: fractals
Message-ID: <27736@lanl.ARPA>
Date: Sat, 29-Jun-85 12:16:55 EDT
Article-I.D.: lanl.27736
Posted: Sat Jun 29 12:16:55 1985
Date-Received: Tue, 2-Jul-85 05:38:27 EDT
References: <1909@ukma.UUCP> <243@kovacs.UUCP>
Organization: Los Alamos National Laboratory
Lines: 33

> In article <1909@ukma.UUCP> sean@ukma.UUCP (Sean Casey) writes:
> >I'd give my right arm for some good 3-D fractal generators. Like Mountains,
> >or those swirly 3D things I saw in a magazine once.
> >

	Last year, I wrote a fairly complete package
for generating analytic 3-D fractals ("those swirly
things").  The code (a) calculates boundary representations
of fractal surfaces generated by iterating polynomials
over the quaternions and (b) casts the model into a
z-buffer, with interpolated shading.  (Fractal surfaces
are infinitely discontinuous, so where's the normal?)

	I wrote the course as a term project in CS175,
Harvard's lab course in computer graphics.  The code
is thus available to anyone who wants it.  (It runs
on 4.x and it's all written in C; the output files are
fairly standard pixel maps.)  The code was written as
a real fire drill; I won't maintain it now, but it did
work quite well in the not-too-distant past.

	A high-resolution run takes hours on a 780 with
FPA (all the calculations are integral, so there!), but
the resulting pictures can be quite dramatic.

	Please let me know if you're interested.

			R. Martin Chavez (rmc@lanl.ARPA)

			Q-4 Nuclear Safeguards
			Mail Stop E541
			Los Alamos National Laboratory
			Los Alamos, NM 87545