American cinematographer (Jan-Dec 1942)

priorities on metal, wooden construction— if sufficiently strong and accurate — should certainly not be scorned. The base should provide for mounting the cameras so that their lenses are horizontally parallel, and spaced 2% inches apart from center to center. It should also provide the necessary i/i-inch bolts to screw into the cameras' tripod-sockets for attaching them to the base, and a similarly threaded socket (a countersunk nut will do) by which the base may be attached to the tripod. Some of the amateurs who have experimented in third-dimensional cinematography have gotten adequate results mounting their cameras on a base which held them rigidily in fixed position, but I think a much better plan was one used by one experimenter, .J. Kinney Moore (a two-time prize winner, by the way, in The American Cinematographer's International Amateur Movie Contests), who divided his camera-base lengthwise along the center-line, and hinged it at the front end. This construction has two advantages. First, it permits swinging the left-hand camera outward so that the, right-hand one can be more easily reloaded. More important from the standpoint of results, it permits the cameras to be "toed in" so that both are centered on the same object, to minimize objectionable parallax effects. The point to align the cameras on is usually the most distant important object in the scene. Pictures made with a camera set-up that can be lined up this way are usually more easy on the eyes than those made with a fixed-camera set-up. Mechanically interlocking the movements of the two cameras is a problem that differs according to the type of camera used. It is easiest with cameras which, like the Cine-Special, the Bolex, the Victor and some of the Filmo 70's, have or can be fitted with a handcrank drive. In this case, all you have to do is insert a short shaft into the hand-crank shaft opening of the cameras, fit a sprocket to each shaft, and interconnect the cameras by means of chain belts and a common shaft in the base. If you use a pivoted base, of course, this connecting shaft should be fitted with a universal joint so that the cameras will operate together when "toed in." With other types of cameras, this matter of interlocking may not be so easy, as it will probably require a minor operation on the mechanism side of the camera-case to gain access to some shaft or gear in each camera that you can use for establishing the interlock. Another experimenter. Dr. O. E. Christ, of Glendale, California, found it comparatively simple, though, using two Model 20 Eastman "eights." Dr. Christ made use of the large winding-gear on the motors of these cameras. He cut an aperture in the case of each camera (on the motor side, of course) to expose these gears. He had a similarly toothed gear in the solid shaft in the base which interconnected the two cameras. Then it was a Three-dimensional projection setup, using two Ampro 16mm. projectors. Note simple but positive interlocking system from hand-turn shafts, and mounting of pola-screens rotated at right angles to each other. simple matter to mount the two cameras so that the gears protruding from the base meshed with the gears on the winding-shafts of the cameras, forming a simple and positive mechanical interlock. The same basic methods can be used in mounting and "syncing" the two projectors for showing the stereo-films. The projectors should be mounted on a common base, with some sort of provision for "toeing in" one projector, so that regardless of the projection-distance, the two images can be projected in approximate register. Asbolute register isn't possible, of course, since the right-hand and left-hand pictures are taken from slightly different angles; but if they are pietty accurately superimposed, the audience will find the pictures a lot easier to look at. Setting up a mechanical interlock is usually easier with projectors than with cameras. Most 16mm. and 8mm. projectors have a shaft — usually at the front of the projector — which carries a knob by means of which the projector can be turned slowly by hand when threading. With the hand-turning knobs removed, these shafts offer an ideal means of jnterlocking the projectors. If you are going to do all or most of your projection at a fixed distance, you can simply put a sprocket on each shaft, and interconnect them by a chain belt. If you expect to change your projection screen distance frequently, you'll have to allow for changing the "toe-in" of one projector, so your interlocking mechanism will have to be moi-e flexible. One way to do this is to fit a pair of bevel gears to each projector and interconnect them with a horizontal shaft which is fitted, of course, with a universal joint. In front of each projection lens is mounted a pola-screen, either attached directly to the projector (separate from the lens, of course, so that the polarizer won't be revolved when the lens is focused!) or suppoi'ted by an upright from the base on which the projectors are mounted. The pola-screens should of course be rotated so that their planes of polarization are at right angles to each other. Since the pola-screens in polarizing the light, also absorb a lot of it (on a camera they have a factor of 4), you'll do well to use the most powerful projectors and globes you can get. Even then, you'll notice a loss of brilliance compared to normal "one-eyed" projection. Since the screen has to reflect polarized light, an old-fashioned silver-surfaced screen with a very uniform, smooth surface is best. If you make your own sci'een, use a metallic silver paint and paint it — or, better, airbrush it — on with the brush moving only in one direction across the screen rather than with a two-and-fro stroke. The audience must be supplied with some form of polarizing spectacles through which to view the picture; otherwise, all they'll see is an incrediblelooking double image on the screen. The cheapest way to do this is to get your polarizing material in gelatin sheet form, and cut it up as necessary. You don't need a big piece for a viewing-polarizer: a square V2 or % of an inch in size is ample, and you can even cut this down to a rectangle V2-inch long by ^^-inch high and still give plenty of room for viewing a small-screen picture. The gelatins can be cemented to a cardboard disc which in turn is mounted in a pair of dime-store spectacle-frames or, if you want to be especially conservative, mount the polarizers in a small, mask-shaped piece of cardboard which the spectator can hold in front of his eyes. If your stereo program is more than about half a reel in length, though, the spectacle (Continued on Page 366) American Cinematographer August, 1942 363