Sutherland head-mounted display

From XVRwiki
Jump to navigationJump to search
Sutherland HMD in use
Quintin Foster in the head-mounted 3D display using the ultrasonic tracking system
Graphics seen in the headset
Graphics seen in the headset

The Sutherland head-mounted display is a head-mounted display (HMD) and computer graphics rendering system created by Ivan Sutherland and a team in the 1960s. It was the first head-mounted display with what was being shown on the screen being 3D-rendered by a computer.[1]

It has 6DOF tracking.

It was designed and built under the direction of Dr. Ivan E. Sutherland at Harvard College.

There is a research publication by Ivan Sutherland outlining the details of the system. In the research publication, he claimed that the fundamental idea behind three-dimensional displays is to present the user with a perspective image that changes as he moves, just like the eyes see in real life.[2]

The original system displays the same image to each eye, thereby not enabling stereopsis from binocular disparity. However it was configured to allow virtual eye separation, thus enabling that stereopsis. Ivan Sutherland said, "One of the reasons we didn't do stereo is my left eye doesn't work very well and so I don't have stereo vision at all."[3]

The Sutherland HMD system was made of multiple parts: The headset, a head position sensor, and electronics including a computer connected to custom electronics. The custom electronics consisted of a matrix multiplier, a clipping divider, and a display vector generator.[4]

The system was completed and tested at Harvard in August 1968, only a few days before Dr. Sutherland brought it to the University of Utah. Some portions of the hardware were immediately put into use at Utah, but the display itself was not again used for viewing synthetic computer-generated objects until January 1, 1970.[4]

The headset was later re-used for the Sorcerer's Apprentice system.

The mechanical system is nicknamed the Sword of Damocles. It does not support the headset's mass. The headset is lightweight and rests its weight on the user's head.[5]

The head mounted display has IPD adjustment capability.[2]

Development[edit]

Ivan Sutherland had visited the Bell Helicopter company in Fort Worth, Texas, who had made a head-mounted display system. The engineers at Bell were trying to help their pilots land in narrow clearings at night, so they developed a night-vision head-mounted display system. They mounted infrared cameras underneath their helicopters and attached them to their pilots’ head equipment which had optical combiners on it. When the pilot turned his head, the camera underneath the helicopter moved in sync with the head movements. Ivan Sutherland thought that it was a clever system, and said, "My big contribution to the head-mounted display project was to recognize that we were just repeating what Bell Helicopter had done, but instead of a camera, we could substitute a computer."[6]

Ivan Sutherland experimented with monocular displays after Sketchpad, but moved to a binocular system.

Headset[edit]

Each temple piece of the headset is a miniature cathode-ray tube (CRT) which is 6 inches long and has a 13/16 inch diameter screen.

It is a rear-display head-mounted display design.

A 2D projection drawn on the CRT face is reflected from an enclosed mirror into a series of lenses and finally to a clear glass eyepiece. Through the center of each eyepiece is a combiner, a partially reflecting silver plane that directs the picture into the observer's eye. The observer thus sees a virtual image of the 2-D projection superimposed on his normal visual field.[4]

The headset has about a 40 degree field of view.[2]

Each cathode ray tube is mounted ina metal can that is grounded to protect the user from high voltage electricity.

Computer system[edit]

The Sutherland HMD system is based on a computer with custom hardware connected to it. The custom hardware is a few parts: 1. a matrix multiplier for for doing a transform from room coordinates to viewport coordinates, and 2. a clipping divider.

The mainframe computer is used to process the head-position sensor information and to hold and manipulate the three-dimensional data in room coordinates.

The custom hardware components of the system are all asynchronous to each other: the matrix multiplier, the clipping divider, and the line drawing hardware.

The system has no hidden line removal.

The mainframe computer stores the room-coordinate 3D scene. The real-time head viewport matrix multiplication happens in the matrix multiplier that is connected to the computer.[7]

As a part of the display process, the endpoints of lines in the room coordinate system are fetched from memory and are each transformed to the eye coordinate system by the matrix multiplier.

The matrix multiplier sends the data to the clipping divider, which eliminates information outside of the user's field of view. The clipping divider is asynchronous to the matrix multiplier.[7]

The outputs of the clipping divider are endpoints of 2D lines specified in the CRT's coordinates.

An analog line generator provides deflection signals to the cathode ray tubes through cathode ray tube deflection amplifiers made using transistors.

The system could do a display of 3000 lines at 30 frames per second.[7]

Each of the three units examines an input flag that signals the arrival of data for it. The data are held until the unit is ready to accept them. Once each unit accepts its data, it clears the input flag. In some cases, each unit will commence processing the next frame before its output datum has been taken. If so, it will pause at the point where it would have to destroy its output datum in order to proceed.

Construction[edit]

The acoustic head position and rotation tracker was designed and assembled at the MIT Lincoln Laboratory by Chuck Seitz and Stylianos Pezaris.[7] Seitz later designed the matrix multiplier.[7] Robert (Bob) Sproull simulated, designed most of, built parts of, and debugged the clipping divider.[7] Two others that were part of the project were Ted Lee and Dan Cohen.[7] Ted Lee developed programs to display curved surfaces using stereopsis from binocular disparity.[7] Dan Cohen developed programs that used the entire system, which formed the basis of the example programs shown.[7] Quintin Foster supervised construction of the equipment.[7] Stewart Ogden helped with the project.[7]

Bob Sproull and Harry Lewis were students who worked in Sutherland's lab while at Harvard.[1]

Before the debut of the system, Ivan Sutherland had done some preliminary 3D-tracked display experiments during late 1966 and early 1967 at the MIT Lincoln Laboratory.[7] These experiments involved a monocular head-mounted display system, a system that presented a visual to only one of the observer's eyes.[7]

The acoustic head position sensor, which operated in continuous wave mode, could operate for a few minutes at a time before piling up errors.[7]

The coordinate transformations and perspective computations were done by software in the MIT TX-2 computer.[7] This system was used to view a 3D line drawing of cyclohexane in augmented reality.[7]

Use[edit]

The Sutherland head-mounted display was used for the Sorcerer's apprentice system.

The HMD was later used by Jim Clark for his 1976 paper titled Designing Surfaces in 3-D.[8]

Computer history museum[edit]

In 2022, the headset was photographed in the Computer History Museum in Mountain View, California as part of a documentation process by students of Scott Fisher.[9] The system was photographed by Nathan Fairchild, a USC graduate student.[10]

3D tracking[edit]

Two separate positioning systems were explored: A mechanical system and an ultrasonic tracking system. The ultrasonic system has three transmitters The first transmitter transmits at 37khz, the second at 38.6khz, and the third at 40.2khz.[2]

See also[edit]

References[edit]