American Cinematographer (1983) - 3-D: The Slowest Revolution
- magazine article: 3-D: The Slowest Revolution
- author(s): Daniel L. Symmes
- journal: American Cinematographer (01/Jul/1983)
- issue: volume 64, issue 7, page 72
- journal ISSN: 0002-7928
- publisher: American Society of Cinematographers
- keywords: Alfred Hitchcock, Chicago, Illinois, Dial M for Murder (1954), New York City, New York, Nigel Bruce, Paramount Pictures, San Francisco, California, Universal Studios, Warner Bros.
Writing a complete and accurate article on three‑dimensional (3‑D) motion pictures is little easier than writing about the invention of the wheel; who did what first and when? This is compounded by the fact that, during a recent patent search on 3‑D motion picture systems, I found about 20 more or less workable designs of which only a few were ever exploited. Finally, I have found that few of the 3‑D pioneers still living have any accurate, let alone unbiased, information available.
With all of this against me, I obviously cannot guarantee the absolute accuracy of my information. What I have done is indicate references where needed and when available.
Stereoscopy has as long a history as most any branch of "modern" science and, for lack of space in this article, I suggest those unfamiliar with the principles of depth perception make a point of reading "A 3‑D Primer" elsewhere in this issue.
As of the writing of this article, one basic law still applies to 3‑D: objects in a given scene must exhibit a spatial differential to one another with regard to a given viewpoint. If said objects do not, they are 2‑Dimensional or flat. The 3‑D effect attributed to bigscreen systems (i.e., Cinerama) is actually a panoramic or peripheral illusion. Literally hundreds of so‑called "3‑D" systems are nothing more than big‑screen displays. We are presently concerned, therefore, with systems that display spatial differentials.
Experimentation to add depth to pictures actually predates cinema. According to a very well researched thesis by H. Mark Gosser (Selected Attempts at Stereoscopic Motion Pictures and Their Relationship to the Development of Motion Picture Technology, 1852‑1903 /1977, Arno Press), there is more than circumstantial evidence that motion pictures may be the result of 3‑D research! Stereo photography (still) was very common prior to cinema. When stereo photographers looked at the old stereograms, two things came to mind: color and movement. It's easy to see how 3‑D could "require" motion before flat photography would. It's an interesting observation.
Methods of achieving depth can presently be slotted into six basic categories: 1) Stereoscope, 2) Anaglyph, 3) Polarized, 4) Raster (grid or lenticular), 5) Alternating View, and 6) Holography.
The simplest and most (ikely first 3‑D system projected two pictures of the same scene (photographed 2 1⁄2" apart‑the approximate distance between your eyes) side‑by‑side on a screen. Spectators had to hold up and look through stereoscopes, much like the ones found in most homes throughout the 1800's. The 'scope would allow the spectator's left eye to see only the left picture and vice versa. Around 1890, such a system was proposed by William Friese‑Greene who, with several experimental screenings, found it to be totally impractical for commercial applications.
Experimenters studied the problem of selective viewing of a "stereo pair" (right and left eye views) and came up with some fairly interesting answers.
Using principles set down by the scientists Rollman and J. D. Almeida, the stereo pairs were tinted in complementary colors and superimposed. More specifically, an anaglyph, as this method is called, is made by tinting the left image, for example, red‑and tinting the right image a complementary color, with blue‑green being the most popular. The two images are projected in superimposition and viewed with glasses made with filters of corresponding colors. The eye with red filter (left) would see only the left picture, which is red. The green filter (right eye) would see only the right (green) picture.
As a practical color film stock had yet to be devised, early pioneers had to either project the two views through interlocked projectors in front of which were the appropriate color filters or they were printed back‑to‑back on two‑sided print stock (emulsion on both sides), each side then being tinted its respective color. The former method was quite acceptable and remained the most popular until the introduction of bi‑color Technicolor. The latter method had, I suspect, problems with focusing as the film base was located between the views. Chances are, one of the views was always slightly out of focus.
When Technicolor introduced its bicolor process in 1921 (the two colors coincidentally being red and green), very high quality anaglyphs were made possible. More experiments ensued.
One scientist (J. Anderson) suggested the idea of using polarized light for stereo projection. A polarizing filter is like a comb that combs light so that it travels in one direction. If the two combs are oriented in the same way, they let light through. If they are oriented 90° to each other, they cross‑polarize and block light, thus becoming opaque (see "A 3‑D Primer").
Polarizing filters, substituted for the colored ones used in the anaglyph method, produce the same "selective" effect without coloring the images. With polarization it was possible to present full‑color 3‑D films.