DETAILS


COLUMNS


CONTRIBUTIONS

a a a

CG PIONEERS

Vol.33 No.2 May 1999
ACM SIGGRAPH

What Does it Mean to be a CG Pioneer?



Carl Machover
Machover Associates Corp.


May 99 Columns
Student Gallery Public Policy


CG Pioneers
Previous CG Pioneers Column


Pioneering Companies

Adage1965
Applicon1969
Calma1968
Computek1968
Computervision1969
Evans & Sutherland1968
Houston Instrument1968
Imlac1968
Information Displays1961
Lundy1970
Megatek1972
Princeton Electronics Products1969
Ramtek1971
Science Accessories1969
Talos Systems1974
Summagraphics1972
Vector General1969

Table 1: In the late ‘60s and early ‘70s, a number of new computer graphics companies were organized. Listed above are just a few of these companies and the dates on which they were founded or on which they became active in computer graphics.

When the Computer Graphics Pioneers were founded about 10 years ago, the founders decided you weren’t really a pioneer unless you worked in computer graphics before the introduction of the storage tube. Before the storage tube, CG was a cure for no known disease - an expensive one, at that.  After the storage tube, CG became a cure for every known disease and the prices have gone down considerably. So we all felt that a true pioneer was one who worked in the tough CG world prior to the storage tube. However, we realized that if we kept that criteria we’d essentially be a celibate organization and would die out, just like the Shakers. So we reestablished the criteria and said you were a CG Pioneer if you had been in graphics 20 years. 

Well, that gets us back to 1979, but I think for a lot of us by the time you got to ‘79, you were really no longer in the pioneering years. Computer graphics started in the era of the Ferranti Mark 1 and Whirlwind computers in the early ‘50s and had about 25 years of development before ‘79. The marvelous Dilbert cartoon that expresses the feeling of the “real pioneers” is reproduced below.

I’m going to look at being a CG Pioneer in the sense of much earlier times. I graduated from Rensselaer in 1951. While I was in college, the transistor was developed, and my design experience was with vacuum tubes and magnetic amplifiers. By the time I got involved with graphics, which was at the end of the ‘50s, the term “computer graphics” had yet to enter the popular vocabulary.  At first we called devices computer-controlled displays …then “information displays”…which evolved into the term “computer graphics.” I’m sure someone can take credit for the term, although I haven’t the foggiest idea who it might be – it became generally used from the ‘60s onwards. 

I remember doing a lecture at Rensselaer Polytechnic Institute a few years ago and one of the students in the class asked me how long I’d been involved with graphics. I realized as I was driving home that night that the answer to that question was that I had been involved in graphics longer than many of them had been alive! That allows me to offer the insight below!

Being a pioneer means…

…working with spectacularly unreliable equipment when time between failures was measured in minutes, not thousands of hours. 

…explaining that the difference between batch and interactive was that in interactive computer graphics you got an answer back before you forgot what the question was. 

…being active when some of these very early companies were founded.  Adage had been in business for a long time (although it didn’t get involved with graphics until 1965), but companies like Vector General, Evans and Sutherland, Megatek were all founded during the pioneering periods (See table 1). It was one of the empire-building times, and software and hardware companies such as these were going through an enormous number of changes and metamorphoses. 

…knowing the number of the walking algorithms before your hair either turned gray or you lost it.

…being in the field at a time when cost justification was virtually impossible. There wasn’t enough application experience using the systems, the costs were extraordinarily high and there was essentially no software to run a graphics terminal. Someone who decided to make that $100,000 investment to buy a graphic system would find it delivered into his/her lab, and the only display they could get was the red pilot light when the system was turned on. 

…remembering when graphic terminals were categorized as dumb, smart or intelligent.

…being involved in the early stages of setting standards. We debated what a “transportable program” was and generally agreed that if we didn’t need to change more than 20 percent of the code to move from one platform to another, then the program was transportable. The efforts of SIGGRAPH and others to create standards that made it easy to transfer programs and images from one system to the other were extraordinarily important. 

…you remember that the National Computer Graphics Association (NCGA) grew out of the SIGGRAPH effort to create application standards.

…working in a time when a color was extraordinarily difficult to obtain. The earlier systems were all vector writers, because the cost of memory was high (a “buck a bit”) and you could store display lists in about one tenth the space it took to store a bit map. Early color systems used all manner of strange technology. There were industrial control systems that used the old CBS color wheel (Dumont built one, the color was gorgeous, and early color TV images from the moon were transmitted by a camera using a color wheel) … or used technologies called the penetration tube, which allowed you to take a CRT and put it in a conventional socket. By changing the anode potential or changing the intensity you could get a fairly recognizable range of colors from red through orange to yellow. 

…I know that the early issues were totally different from the issues that people confront today. There were heated battles about the best way to make characters using strokes, raster or dots. There were heated battles about how to draw lines (vectors) on a screen to get the kinds of speed you need, because in the non-storage tube refresh displays, the task was to make the hardware function generators fast enough so that you’d get a reasonable amount of picture without the picture flickering (and even flicker was a matter of definition). 

…I know there was a constant hardware vs. software battle. In the early days, when a character generator would sell for $10,000, it made sense to develop characters as a software function.  As the cost of specialized chips went down, and down, and down, and you could build a character generator that could fit on a chip the size of your thumbnail and cost fractions of a penny, that issue gradually died away. We continually go back and forth on the question of hardware vs. software, especially as the capability of the processor increases.

…I know that on the other hand, the relative availability of ASICs (application specific integrated circuits) brought about by one of the early computer graphics applications … CAD/CAM … very often settles the argument on the side of hardware (You’ll guess “correctly” from my comments that I’m a hardware person, not a programmer!).  Another early issue of hardware vs. software was how to get picture motions. Early systems, like the Adage, Vector General and Information Displays, Inc. (IDI), all featured three-dimensional rotations.  Adage and Vector General did it through various kinds of hardware matrix multipliers and IDI did it through software. We had heated religious arguments about which was the better way to do it. 

…remembering that very early character generators used a wide variety of technologies including curve following. 

…you are cognizant of the early quality issues. There was just so much you could do, although it was pretty spectacular, with a stroke-writing display. Some of the early character generators had readable but not very high graphic quality characters. One of the first higher quality character sets was the Hershey fonts, a set of stroke descriptions of character fonts developed by Hershey, who worked for the Navy. One of the early software companies that was very much involved with visual quality was ISSCO. The lore is that they offered a prize … in the order of $2,000 … to the first person who could get a computer-generated chart in Scientific American (early chart qualities were generally so poor that these established magazines simply wouldn’t accept computer-generated pictures). Well, we got to that point and today computerized typesetting has simply taken over from many manual methods. 

…that you went to most of your computer graphics conferences as part of the semi-annual Fall and Spring Joint computer conferences sponsored by the American Federation of Information Processing Societies (AFIPS).  AFIPS died away about a decade ago because they were essentially a mainframe organization and did not get into the workstation and PC era.

…that you explained to people how a mouse worked by telling them that a mouse was a trackball turned upside down. 

…constantly debating the merits of a light pen vs.  A trackball … a trackball vs.  A graphic tablet and, if you were a light pen aficionado, constantly combating the argument of orangutan arm (your arm would swell up because you had to keep holding the light pen to the screen). 

…being gratified when IBM finally came out with their commercial graphics system, the 2250 in conjunction with the IBM 360. Prior to that, the constant rejoinder to “Will you buy computer graphics equipment?” was, “If it was so good, why doesn’t IBM have one?”

…remembering when Ed Fredkin was still with Computer Control Corporation - long before he established Information International and programmed some of the early systems.

…taking part almost every year in a conference entitled, “Why is computer graphics always a year away?”

…engaging Malcolm McCauley of Data Displays, Inc. in a conversation in which he defended the use of vacuum tube transmitting tubes as the drivers for his display systems, when everybody else was going to solid state. His argument was, “Why not use tubes … the CRT is a tube also?”

…remembering that before Newman & Sproull wrote their classic CG book in 1973, there were a generation of display graphics books that preceded. They carried titles like, William A. Fetter’s Computer Graphics in Communication (1964), Harry Poole’s Fundamentals of Display Systems (1966), Fred Gruenberger’s Computer Graphics Utility/Production/Art (1967), Murray Milne’s Computer Graphics in Architecture and Design (1969), Parslow, Prowse, and Green’s Computer Graphics Techniques and Applications (1969), Horton’s Data Display Systems (1970), Sol Sherr’s Fundamentals of Display System Design (1970), David Prince’s Interactive Computer Graphics for Computer-Aided Design (1971) and Jasia Reichardt’s Cybernetics, Art and Ideas (1971).

…recalling the early days of graphics memories. Since the cost of “core” memory was so high we turned to magnetic drums and magnetostrictive delay lines. Drums were OK, but not very flexible. The magnetostrictive delay lines were pretty good, unless someone happened to turn the power off and you lost all the information. (This happened to us when we were installing a system in Canada and the cleaning lady came and pulled out the plug for our system!)

…remembering when display instructions were inserted on a bit-by-bit basis through front panel switches. The memory was so expensive that if you were driving a display that used a 24-bit word organization from a 36-bit computer, you’d be sure that you packed one 24-bit instruction plus half (12 bits) of the next instruction to fill out the 36-bit word. It also meant being part of the world when the transition between the cost of memory and the cost of programming occurred - it was much more economical to waste bits than it was to waste the programmer’s time. 

…showing Louis Earl of Mountbatten, at the 1970 IFIPS Conference in Edinburgh, how he could use a light pen on the face of a CRT to simulate the raising and lowering of a gun on a battleship. It meant being very proud when your equipment booth showed up on national TV as part of a David Frost Show illustrating one of the early computer pool games. 

…you remember when an early TV drama arranged to use your system in an air traffic control sequence in which the director said that he wanted to show the tracking of two aircraft on a collision course so that as the two aircraft impacted, “a flower of death” would appear on the screen. 

…one recalls the ‘60s when women were scarce in the CG field and when, if you used one of your women marketers in an exhibit booth, people naturally assumed she was a model and were taken aback with her very knowledgeable responses to what they thought were their technical trick questions!

…remembering that you sold other pioneers, like Dave Evans, Jose Encarnação and Marcelli Wein, their first graphic display. 

…that much of the computer art that you remember looked like geometric string pictures.

…having known most of the mentors that appear in the Webster-Badler conetree that was developed for the SIGGRAPH 25th Conference Celebration.

…being especially saddened by the recent passing of close friends and professional colleagues like Steve Coons, Dave Evans and John Lansdown.

…that what was once clear and precise begins to get hazy as time goes by, and that haziness is reflected in this article. 

Let me hear your recollections about the early days of computer graphics.

A Tribute to Professor John Lansdown


John Lansdown
January 2, 1929 - February 17, 1999

Robert John Lansdown, computer graphics pioneer, Professor Emeritus at Middlesex University’s Centre for Electronic Arts and my dear friend and professional associate died on February 17, 1999, at age 70. 

As early as 1960, John Lansdown, a very successful architect with offices in Russell Square, was a believer in the use of computers for architecture and other creative activities. He pioneered the use of computers as an aid to planning, making perspective drawings on an Elliott 803 computer in 1963, modeling a building’s lifts and services, plotting the annual fall of daylight across its site and authoring his own computer aided design applications.

He joined ACM in 1972 and Eurographics in 1983. From the early 1970s to the 1990s, he took influential roles in several professional bodies, and he chaired several committees, through which he drove the world leading strategy for developing computer aided architectural design in U.K. universities. He had enormous influence as founder, member and secretary of the Computer Arts Society (1968-1991). He was on 10 editorial boards and chaired and organized many international conferences. Event One at the RCA (1969) and Interact at the Edinburgh Festival (1973) were seminal events in establishing the use of computers for the creation of art works. 

He had industrial links with Ove Arup Partners and Crosfield Electronics. From 1977, with his colleague George Mallen and others from the Computer Arts Society, John founded System Simulation, a company he chaired until 1988. Through it, he developed major innovations in computer animation, such as special effects for advertisements and television titles, the feature films Alien, Saturn III and Heavy Metal. John created what was, at the time, the world’s largest computer generated mural.

He became directly involved in education as Senior Research Fellow and Tutor at the Royal College of Art and Senior Visiting Fellow at the Department of Architectural Science, University of Sydney from 1983. He joined the staff of Middlesex Polytechnic (now University) in 1988 as Head of the Centre for Advanced Studies in Computer Aided Art and Design (now Centre for Electronic Arts), also becoming Dean of the Faculty of Art, Design and Performing Arts in 1992 and then Pro-Vice Chancellor of the University in 1993. He relinquished these roles on formal retirement in 1995, but continued to be very active and influential as Emeritus Professor in the Centre for Electronic Arts. He continued, through supervision and example, in the role of research mentor.

John Lansdown’s range of publications began to diversify from the early 1970s. He wrote the classic Teach Yourself Computer Graphics (Hodder and Stoughton, 1987), exhibited algorithmically generated images, animations, compositions, conversations, sword fights and choreography, such as the 18 minute dance piece A/C/S/H/O commissioned by the One Extra Dance Company and performed at the Sydney Opera House in 1990. He contributed as author and/or editor to 34 books and worked on more than 100 conference and journal publications, including a special May 1995 CG&A computer art issue that I edited.

The 51 articles in his regular quarterly series, “Not only computing - also art” in Computer Bulletin (1980-1992) introduced readers to subjects as diverse as Escher, Rubik cubes, word processing, chaos theory and Turing. He has been referred to as “the British Martin Gardner.” A careful theoretician, he believed in a “hands on” approach, developing his own software in up to date languages - his last conquest was JavaScript.  As an academic he embraced multimedia and the World Wide Web, contributing to innovative courses in the Centre for Electronic Arts. His latest enthusiasms were for the lively Sonic Arts group and the brand new course that allies his two major disciplines, the M.A. digital architecture. 

The most genial and witty companion, those who worked closely with him will remember his positive attitudes and infectious enthusiasms, his sheer humanity and subtle sense of humour, expressed in finely crafted language. He encouraged others to extreme physical expression through dance, fashion and experimentation in music - he only attended concerts which included at least one work by a living composer. Having sight only in one eye, he hated being left out in the era of binocular 3D computer effects. He leaves his wife of 47 years Dot (Dorothy), his son Robert and daughter Karen, their spouses and five grandchildren.  His wife can be reached at 50-51 Russell Square, London WC1B 4JP, +44-71-580-2410.

This tribute was adapted from more extensive obituaries prepared by his Middlesex University colleague, Huw Jones (huw1@mdx.ac.uk) and his SSL business partner, George Mallen (george@ssl.co.uk).


Photo credit: Louis Fabian Bachrach

Carl Machover
President
Machover Associates
152A Longview Avenue
White Plains, NY 10605

Tel: +1-914-949-3777
Fax: +1-914-949-3851


The copyright of articles and images printed remains with the author unless otherwise indicated.

Carl Machover is president of Machover Associates Corporation, a consultancy that provides a broad range of management, engineering, marketing and financial services worldwide to computer graphics users, suppliers and investors.  Application areas served include CAD, CAM, CAE, CIM, business, slidemaking, art, animation, graphic arts, multimedia, visualization, VR, process control, technical documentation, engineering and science. He is also an Adjunct Professor of computer graphics at Rensselaer Polytechnic Institute, President of ASCI and Past-president of the Society for Information Display, National Computer Graphics Association and The Computer Graphics Pioneers.

Machover is an internationally recognized expert in computer graphics. In addition to his extensive consulting activities, he is on the editorial advisory boards of many industry publications, including Computers and Graphics and Computer Graphics & Applications. He has published more than 180 articles on computer graphics, has given computer graphics seminars and lectures worldwide, is editor of The CAD/CAM Handbook (1996), co-author of Japanese Computer Graphics Industry and Markets, associate consulting editor of the WSPC Information Display series, the Tab (McGraw Hill) Professional & Reference Book series, Computer Graphics Management and Technology and series advisor on the McGraw Hill Visual Technology series. In 1988, Machover received the North Carolina State University Orthogonal Award, and was inducted into the FAMLI Computer Graphics Hall of Fame. In 1993 he was the first recipient of the NCGA Vanguard Award. He was the 25th Conference Celebration Chair for SIGGRAPH 98 and is co-executive producer of the ACM SIGGRAPH 90 minute HDTV documentary The Story of Computer Graphics. 

He graduated as an electrical engineer from Rensselaer Polytechnic Institute, and has four decades of experience in computer graphics. He is a Fellow of the Society for Information Display and the Eurographics Association and member of the IEEE, SIGGRAPH, ADDA, Computer Graphics Pioneers and other professional societies.