The Origin of Computer Graphics within General Motors
Original Publication Date: 1994-Sep-30
Included in the Prior Art Database: 2005-Oct-07
Software Patent Institute
FRED N. KRULL: AUTHOR [+2]
In 1955, the computer scientists at the General Motors Research Laboratories became recognized as a separate group within the Special Problems Department. Donald Hart was appointed assistant department head with George Ryckman and Edwin Jacks as supervisors. The group was named Data Processing since computer science was not yet recognized as a separate discipline by most universities. The group was later made into a separate department in 1961. Under the leadership of Donald Hart, two major departmental activities were undertaken. George Ryckman's group was responsible for operation of a computing center that, at that time, included an IBM 704 computer, keypunch stations, and off-line card-to-tape and tape-to- printer equipment, together with the system programs to support their operation. The remainder of the department, under the supervision of Edwin lacks, consisted of a programming staff devoted to the development of software applications. Both systems and applications were very fertile areas for work at that time. The systems group was just beginning to gain recognition in the IBM user community as a leader in the development of computer batch operating systems. Early experiments among GMR, North American Aviation, and IBM led to the development of a batch monitor* [Footnote] *At the time of the events described in this article, the term ";monitor"; was used for what would now be termed an ";operating system."; program for an IBM digital computer.1 Most of the programming was still being performed in IBM assembly language, but by 1958 a new Fortran compiler from IBM was being tested at selected customer sites, including GMR. During this same period IBM marketed a film recorder for the IBM 704 computer that could be used to record ";point plots"; on 8-mm film. This facility provided engineers with their first opportunity to view computer-generated graphs and computer-animated movies. Computer- generated traffic simulations were recorded on film using this equipment. For demonstration purposes, IBM also provided a display unit that operated as a slave to the film recorder so that the plotting could be seen by the machine operator. The film recorder and display unit (Figure 1) became the basis for the initial GMR experiments in interactive computer graphics.
THIS DOCUMENT IS AN APPROXIMATE REPRESENTATION OF THE ORIGINAL.
Copyright ©; 1994 by the Institute of Electrical and Electronics Engineers, Inc. All rights reserved. Used with permission.
The Origin of Computer Graphics within General Motors
FRED N. KRULL
This article traces the history of the development of computer graphics technology at me General Motors Research during the period from 1958 to 1967. A concept demonstration was formulated in the late 1950s to show the feasibility of applying computer technology to the problem of vehicle body design. The narration then traces the history of a joint project between GM and IBM for development of new and unique computer graphics hardware. The salient features of the Design Augmented by Computer (DAM 1J system are summarized in terms of nine separate technologies that were brought together for the first time to form a complete computer-based design environment.
It is always instructive to rediscover an idea that still has some merit today and is relevant to current technology and problems. In writing this article, I have, across the distance of time, gained a new perspective of some things that were done right and some that were done wrong during the course of a rather large and involved technical project.
As early as 1952, the General Motors Research Laboratories (GMR) were using a card- programmed digital computer for engineering and scientific analyses. However, notably absent from the applications were problems related to graphical design. To gain an insight into the automotive design process, research personnel began discussions with General Motors (GM) engineers and designers. It soon became obvious to the researchers that drawings, pictures, and models were the principal media for communication and documentation of design ideas.
As a result of these discussions, four distinct types of man-machine communication were identified:
1. Existing engineering drawings. The research project was realistic enough to realize that computers could not replace all the drawings used in the design process. Such a claim would have been akin to the office automation claims of the paperless office, which still shows no prospect of being achieved. Therefore, it was concluded that a computer system must provide means for reading existing engineering drawings and for creating these drawings. Fortunately, because of the nature of automotive design, body drawings are primarily drawn to scale on a background of grid lines with no dimensions. Digitizing this line information seemed to be a feasible task.
2. Interactive manipulation of graphic information. The design process frequently involved one person indicating a problem on a drawing to another person and then their joint exploration of potential changes to the drawing. It had to be possible to "point to" or indicate the location of the problem, and to make an immediate change so that the implications of the change could be evaluated.