VR Hardware


Sensory Displays

One of the basic goals of a virtual reality system is to supply your senses with information from the computer-generated reality in much the same way as you experience the real world. Since most people have two eyes, a natural way to see the world requires not one computer display, but two. A common way to produce a realistic 3D view of a virtual world is to place a small computer monitor in front of each eye. Each monitor displays the perspective that the corresponding eye would see in an actual environment. Such a system is called a binocular head-mounted display (HMD).

Most people also have two ears. This is the main reason for the appeal of stereophonic sound. Just as two visual perspectives make a 3D view, two audio perspectives can make a 3D soundscape. However, with free-standing stereo speakers the left and right sounds are mixed: both ears hear sound from both speakers. By using headphones and presenting the correct acoustical perspectives to each ear, many of the spatial aspects of sounds can be preserved. HMDs often have headphones built into them.

Additional displays can be used to engage other senses in VR. Since there is not much of a demand for such things as smell or taste generators, you generally have be creative and figure out your own way of catering to more senses than just vision and hearing.


A standard personal computer can be equipped to run simple virtual environments. The processing speed of the computer will determine the maximum complexity of the virtual environment you can build. In order to maintain a decent illusion of reality, the spatial aspects of the sensory displays must be recalculated and updated more than 20 times each second. In order to figure out how much complexity can be included in your virtual environment, you need to be able to describe the speed of your computer in VR terms.

Most of 3D graphics is based on building objects out of triangles or other simple polygons. For computing visuals, a convenient metric is the number of polygons your computer system can draw in one second. Since the computer must draw separate views for 2 eyes at least 20 times each second, you must divide the number of polygons per second by 40 to determine the maximum number of polygons that may be simultaneously visible in your virtual world. Thus, a computer that can draw 50,000 polygons per second will be able to support a virtual environment containing a maximum of:

50,000 polygons per second
------------------------------ = 1,250 polygons
2 eyes * 20 views per second

Since 1,250 polygons is not very many from which to build a whole world, and since the polygon drawing speed quoted by hardware and software manufacturers is often optimistic, you generally need to either design very simple worlds, or get extra graphics hardware to help out your computer.

Process Acceleration Cards

There has been an explosion of manufacturers producing 3D graphics accelerator cards for personal computers. These cards (as of September 1996) perform in the 500,000-2 Million polygon per second range, and vary greatly in cost from $300 to $20,000 (and the cost does not necessarily correspond to performance!).

There are also several manufacturers producing 3D sound cards. These sound cards allow you to give a moderately good sense of position to a small number (1-4) of independent sound sources.

Tracking System

The tracking system measures position and orientation. From the position and orientation of your head, the computer can determine how to display the virtual world so that it seems your are in it as opposed to watching it on television. When you turn your head the head tracker senses the change in position, and adjusts the displays accordingly.

The head tracker needs to be capable of taking a measurement of position and orientation at least 20 times every second. There also must be no more than a 1/20th of a second delay between when the measurements are taken, and when the visual display is updated. Any slower than this, and the eyes and inner-ear give your brain conflicting information about which direction your head is pointing. This is similar to what happens on a small boat in rough waters: It can make you seasick, or in VR terms, simulator-sick.

Input Devices

You use input devices to communicate your intentions and actions to the virtual world. Since it is often difficult to touch-type or use a mouse while standing up and wearing a head-mounted display, other types of input systems are used instead of or in addition to keyboards and mice.

A wand is basically a hand-held joystick with a number of buttons on it. Wands often include a tracker which then allows you to pick up and rotate objects in the virtual world. You can use a wand in VR in much the same way you use a mouse on a computer desktop. Moving the wand in space moves a 3D pointer in the virtual environment. You can also click and drag virtual objects, but instead of just moving them vertically and horizontally, you can also move them in depth and rotate them about all three axes. This is why wands which can move objects in the X, Y, and Z directions and also rotate them about the X, Y, and Z axes are sometimes called six-dimensional or 6D controllers.

Since distances can be arbitrarily large in a virtual environment (and trackers have a limited range) it is not usually practical to travel through VR on foot. Some of the wand buttons are often used for "flying": you pointing your wand or your head in the direction you wish to travel, and then press the "fly" button. In VR, there is no speed limit.

VR gloves are like wands, only more complex. They consist of a tracker to sense the position and orientation of your hand, and some kind of flex sensors to measure the bend of your fingers. Gloves tend to be expensive and tricky to use partially because the computer must be able to recognize intricate hand signals instead of simple button presses.

Next Topic | Home