Magazine article American Cinematographer

A 3-D Primer

Magazine article American Cinematographer

A 3-D Primer

Article excerpt

The classical definition of 3-D goes something like this: "When we look at the world around us, we see it with two viewpoints at the same time, our two eyes. Because of this, we are able to see not only the height and width of things but also their depth and distances from us and each other. We are able to see all three dimensions; hence the term '3-D.' Try this simple experiment: cover one eye with your hand and look around you. Notice how flat everything looks? Now uncover your eye. Notice how everything jumps back into proper perspective? Each of our two eyes picks up a slightly different view of the world. To prove this, close one eye and hold a finger up about a foot and half from you. Look at some distant object and place your finger in the way. Keep looking at the far object and open the other eye as you close the first one. see how your finger apparently jumps aside? This apparent displacement is called parallax, and it is the proof that we receive two distinct views of whatever we look at. The brain combines these two essentially flat views into one stereoscopic view.' '

For professional consideration, the classical definition falls down in these areas:

1. The results of the experiments we are asked to perform beg that we infer that all that is needed for lifelike 3-D perception is two views of the subject. This can easily be seen to be false by anyone who has seen a narrow-screen 3-D film. There is stereo relief, but without the peripheral experience wide-screen systems give, realism is found to be lacking. This peripheral image is so important that many people considered Cinemascope to be more realistic than the Academy-aperture 3-D movies of the early fifties (Fox advertised Cinemascope as "the miracle you can see without glasses").

2. The results of the experiments beg that we infer that one-eyed people cannot judge depth and distance. In actuality this is not the case at all, and here is an "anti-experiment" to prove it: cover one eye and get up and walk around the room. Walk into another room. Do you have any trouble judging distances? Sit down. Did you miss the chair because you were unable to judge its distance or position? No, of course not. Much depth judging is done by shading, relative size, and other non-stereo visual cues. And not incidentally, Andre de Toth, director of HOUSE OF WAX, one of the finest 3-D motion pictures ever made, both technically and artistically, has only one good eye.

3. The definition begs us to infer that one cannot judge relative depth and distance in a flat-screen motion picture-a particularly unfriendly inference for directors and cinematographers, since they strive to put depth into a flat picture by lighting, lenses, selective focus, composition, etc. The audience can tell who's standing in front of whom and how far away from the wagon train the Indians are.

In view of these failings, a new definition for 3-D is proposed: "The perception of depth and distance based on binocular vision (as given in the classical definition) combined with the visual cues of shading, clarity, size and speed (relative and apparent versus known) and probably a host of other factors." There is a reason for this ambiguity at the end. The physical mechanics of 3-D perception are pretty well down pat now, and we even have most of the psychological reasons classified and labeled, but there is evidence that somewhere between the two in the never-never land of the psycho-physiological, there may be still other factors which affect depth perception.

It may be that we cannot, at present, give a complete definition of what 3-D is and how we see it. Fortunately, for the practical purposes of this article and the motion picture industry, as well as my overworked brain, the classical definition will do, so long as we recognize its limitations.


Before we can see how the various types of 3-D imaging systems (both inside and outside the motion picture industry) work, we need to recognize the things that all 3-D systems have in common. …

Author Advanced search


An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.