home .. forth .. misc mail list archive ..

Re: Hardware/software:2


Louis Frazier,

I am sorry to be so verbose in this file, but since you ask:
("I am interested in learning from anyone who has ideas about
applying MISC to art.")

Hud Nordin and I began developing an application in 1988, called  
VidSynth.  VidSynth is an application, currently available for the  
Macintosh, which allows artists to create animated video synthesis  
artworks by menu selection, composition of functions (chaining), or  
by describing the formula(e) directly in a C-subset-style notation.

The entire approach we have taken is to keep the descriptions of  
the geometries comprising the animations segmentable to the pixel  
level.  In other words, we can transform the descriptions of the  
artworks into a separate equation for each pixel, or group of  
pixels, which describes the color of each pixel independently at  
each point in time.

We did this, in part, because we forsaw a time in the future when  
there would be parallel machines which would contain large groups of  
processors.  It is our intention to evaluate these functions in  
parallel, without blocking due to data-dependencies stemming from  
neighboring geometries.  It is our hope and belief that this  
approach will allow highly complex animations to be rendered in  
real-time.

We have explored a significant part of the simple mathematics which  
is convenient to this architecture, and recorded the results on  
videotapes, to which we hold copyright.  Creating these artworks is  
not as simple as it might seem; mathematics is a large space, and  
very little of it is interesting to look at.  We currently have a  
library comprising a few thousand of these animated artworks.  Many  
thousands of man-hours were required to create this library.

When someone finally builds a million processor machine, we are  
ready to use it.  I believe that MISC affords the most expedient  
path to the implementation of such a machine.

Thank you for your kind indulgence,

			Sherwin Gooch


Begin forwarded message:

Resent-Date: Wed, 31 Jan 1996 12:23:11 -0500
Date: Wed, 31 Jan 1996 11:13:47 -0500 (EST)
From: Louis Frazier <cogniscu@tmn.com>
To: Wayne Morellini <waynem1@cq-pan.cqu.edu.au>
cc: misc <misc@pisa.rockefeller.edu>, Wayne Morellini  
<waynem1@cq-pan.cqu.edu.au>
Subject: Re: Hardware/software:2
In-Reply-To: <199601310605.QAA26984@cq-pan.cqu.edu.au>
Resent-From: MISC@pisa.rockefeller.edu
X-Mailing-List: <MISC@pisa.rockefeller.edu> archive/latest/326
X-Loop: MISC@pisa.rockefeller.edu
Resent-Sender: MISC-request@pisa.rockefeller.edu

Wayne, I don't know if this kicks off (HS)embedded, but I'll give  
it a try.

I blush to say such obvious things in the presence of Chuck and probably 
Jeff as well as anyone else who knows what went on at Xerox Parc
(Palo Alto Research Center) 25 year ago, but xx21 processors should  
be in
every printer in the world. I found my way to FORTH via Postscript, the 
rpn language which is the defacto standard for page/screen description. 
It is the standard because it is best able to put points on a surface or 
in space. I don't know which came first (I'm sure someone will tell me 
shortly), but Forth and PS are so much alike, either could be
encapsulated in the other -- and probably has been.

This brings me to the mass market. Sun is touting "net-centric" computing
in trade shows this week. They are talking about the successor to the PC
being a good display and keyboard hooked to a fast graphics computer that
gets all it's software and information from the net, at three  
megahertz, as
it needs it. I'm betting that xx21-type hardware (MISC, FORTH) will  
be the
best candidate for these systems which will be more pervasive than laser
printers in a few years.

HTML, the primitive composition program we are using on the net  
today, can
easily be written in Postscript. (Adobe has no doubt done it.) Netscape 
and all the other browsers are going to  be forced to go to Postscript 
eventually to be really competitive. Things are happening awfully fast 
right now, and probably most of the software work is in alpha at least.

People who are interested in applications these days can start with 
silicon, as Chuck has  proven with MuP21. We who are following MISC are 
on the leading edge of something big and it behooves us to learn as much 
as we can from each other, whatever our interests are.

At 70 I consider myself an artist dedicated to expressing the  
beauty and the
power of the technologies we've been working in and applying since  
Edison
invented the triode.

I want my art to demonstrate the leading edge of networking. I want to 
build a piece that is a free-standing entity on the net, able to  
interact
with its viewers, ascertain their interests and go to the net to satisfy 
them - with music, images, sounds, questions, answers -- anything that 
seems to interest them. I expect that many artists will produce  
thousands
of works in the next few years to appeal to a new generation of
technology-aware art lovers, just the way millions of painters and
sculptors have met their aesthetic needs in the past.

To do this, artists will need all the usual art materials plus vast
quantities of cheap chips like xx21's, just they way they've consumed
millions of tubes of paint since someone put paint in a lead tooth paste
tube and started the impressionist movement of outdoor painting in the
19th century. I am interested in learning from anyone who has ideas about
applying MISC to art.