By David Starkman
With the increasing popularity of twin 35mm camera rigs and even custom built twin 35mm SLR stereo cameras, mounting in separate 2"x2" (50mm x 50mm) slide mounts has also become common. A wide variety of professional quality mounts are available, with an even wider variety of aperture sizes. Such mounts provide the opportunity for superior stereo slide mounting, and this also allows for the possibility of automated 3-D slide projection with Kodak Ektagraphic Carousel-type projectors.
Precision mounting is the key to comfortable 3-D viewing and projecting, and the techniques involved are no more difficult than those used for standard 1 5/8" x 4" (41mm x 101mm) stereo slide mounting. In fact, separate slides allow the use of some techniques, which can make mounting even easier than with one-piece stereo slide mounts.
Before I get into mounting techniques I first want to mention what types of mounts to use. As already mentioned, the mounts should be precision made, allow for adjustable film positioning, have a variety of apertures, and include glass (although the glass can be optional if you are using autofocus projectors) . Two brands which fit the description quite well are Wess and GEPE brands. In Europe Bonum (glassless) and Jedam are also available, and are ideal for stereo use. While each brand of mount has its following, it is not my intent in this article to discuss the pros and cons of each one. They are all capable of providing good mounting results.
The mounts that I personally use are the GEPE brand. Unlike the others these seem to be the most universally available both in the USA and Europe. I will use them as an example, but remind you that any precision mount is acceptable. One caution, however. Since each brand may vary just slightly in thickness or outer dimensions, it is best to choose one brand and stick with it. It is definitely a good idea not to mix brands within a single slide program.
The GEPE mounts consist of front (white) and back (gray) plastic frames, each with an ultra thin piece of glass held in place by a precision die-stamped stainless steel mask, which also forms the film aperture. The mask also has top and bottom cut-out channels to hold the film in place. Many apertures are available, with some of the most useful for stereo being (all in mm) 24x36, 24x32, 24x28, 21x28, 24x24, and 18x24.
The separate front and back halves snap together, and are completely symmetrical, allowing new aperture sizes to be created by crossing the fronts and backs of two different sizes. For example, crossing a 21x28 with a 24x24 will create a 21x24mm mask (Stereo Realist Format).
To facilitate easy and precision mounting the minimum tools required are a tweezers to handle the film chips, a light box, a stereo viewer for viewing mounted slides while on the light box, and tape or tabs to hold the film in place for final mounting (not all mounts require the use of tape). Once you have done 2x2 stereo mounting, and have decided that you might want to do this on a frequent basis, you might want to consider constructing a mounting jig. This jig should incorporate secure holders for the slide mounts so that the film chips can be worked on while stereo viewing them, facilitated by built-in illumination and a built-in viewer. A simple design for a build-it-yourself jig will be shown later in this article.
Good mounting requires several key elements. Leaving slides in their mounts just as they come from the processor may be tolerable to many for hand viewing or previewing, but it is definitely NOT good enough for stereo projection. For comfortable projection the film chips should be:
1) Vertically matched: that is, one film chip should not be higher or lower than the other in the film aperture. The elements that you see across the top and bottom edges of the aperture should be exactly the same in both the right and left slides.
2) Without rotation errors: Matching elements in the right and left film chips should not be higher or lower at one end, indicating that one film chip is rotated in relation to the other.
3) Horizontally adjusted for the best "stereo window". This is by far the least obvious error in mounting, and the hardest one to explain, especially in writing, without the benefit of a "hands on" demonstration.
I think that 1) and 2) are basically self-explanatory, so I will concentrate now on 3):
Proper 3-D slide mounting and the stereo window.
When a 3-D slide is viewed, whether in a hand viewer or by projection, the dark edges that frame the image become, even more so than with flat photography, a window. This creates a spatial frame and reference by which the depth of the scene is both framed and measured.
In every sense, the frame of the 3-D slide is analogous to a real window. All of the visual cues which tell us that we are looking at a three dimensional scene outside of a real window need to be correctly recreated in the equivalent three dimensional picture.
Since stereo cameras do not have the ability to continuously adjust and converge on the changing world of 3-D subject material, the lenses of 3-D cameras, and twin camera rigs, are generally fixed parallel, or at very close to parallel configurations. Depending on the focal length of the taking lenses, a varying amount of the same scene, captured from two slightly different angles, is seen by each lens. The part of each scene that is common to both images will be seen in 3-D. The longer the focal length of the lenses, and/or the closer the camera is to the subject, there occurs an area at the outer edge of the right and left images that is totally different in each. With separately mounted slides one could simply superimpose the two film chips and match the main subjects. Then one could quickly see the amount of "extra" film sticking out past the right and left edges of the roughly superimposed scene to see how much area is not common to both images.
For comfortable and aesthetically pleasing projection this not common area should be masked away - usually by mounting in a slightly narrower aperture mask, e.g.. going from 24x36mm to 24x32mm masks.
In addition, one has a choice to make in the horizontal adjustment of the film chips in the two masks. When the film chips are moved towards each other it brings the position of the scene closer to the stereo window. When the film chips are moved away from each other the scene is pushed back farther from the window. The key to good mounting is finding the best and most consistent method of aligning the film in this manner.
As with many aspects of 3-D photography, there are differing opinions as to the best ways to do anything. In this article I am advocating mounting to a consistent "near point" image separation. In 15 years of mounting I have found that this method makes for comfortable 3-D slide projection, with a minimal amount of projection ghosting.
In verbal terms, this means mounting so that the nearest object appears to be at the plane of the window created by the mask apertures, with the rest of the scene falling behind this point. (Of course, there are always exceptions, such as when you have a subject that does not touch the edge of the window that you WANT to appear in front of the window frame.)
In measurable terms this is achieved when the measured distance from one edge of the mask to a near point in the left image is the same as the distance measured from the same edge to the same point in the right image. This is more easily seen in figure X, where the distance to the near point is measured to a distance "B" in both images. This would mean that if one were projecting with the two apertures superimposed on the screen, then this near point would also be the one part of the entire 3-D scene which would also be superimposed. In terms of the visual 3-D effect, this would make the near point appear to be at exactly the same plane as the window frame.
Of course, in mounting, it may occur that meeting this requirement will mean that one or both of the images leaves a gap at the edge of the film aperture. This is not acceptable, and requires the use of narrower masks until both properly adjusted film chips will fit into the apertures without leaving any gap.
Two methods to achieve matching near point distances have already been mentioned. One would be by separately measuring the distances with a ruler or scale.
Superimposition is a quicker and easier method, which is greatly facilitated by making a mounting "nest" that snugly holds the 2x2 slide mount and allows two mounts to be stacked in the nest, one on top of the other. Add a bright light source under the nest and a magnifier on top of it, and mounting by superimposition becomes much quicker and easier.
Although I personally use superimposition regularly, especially as a final check for rotational errors, my favorite method is to use a mounting jig, like the one illustrated in Figure 1. (This was originally designed by Arthur Girling and David Burder, and appeared in No. 112 issue of the Stereoscopic Society Bulletin.) This jig allows one to actually see the magnified image in 3-D while adjustments are being made. The edges of the apertures themselves become the reference gauges for both vertical and horizontal adjustments.
The jig illustrated is intended to be made almost entirely of plywood (or any material you choose of a suitable thickness). The lens panel is also of the same wood, with holes cut out to hold whatever diameter suitable lenses that you may find. To allow for enough room to work under the lenses approximately 3" (75mm) focal length would be suggested. The slot in the lens back support panel allows for individual focusing adjustment. A single light bulb could be used as the light source, although the design could be adapted to use compact fluorescent "bulbs". A white plastic panel is inset in the area behind the slide holder, and the holding frame (strips A, B, C, and D) could be made either of wood or plastic strips. The main requirement is that the two slide mounts be held snugly in place. Making strips C and D adjustable by the use of slotted holes, and held in place with wing nuts, is quite desirable.
Although the film aperture edges may be used for mounting reference, it may also be useful to add a grid of lines over the white plastic area to provide more mounting references. Due to their ready availability many people have used the Reel 3-D Enterprises "3-D Slide Alignment Gauges" (Stock No.8001), cut up, for this purpose.
Unfortunately, there is no manufacturer of such a jig at the present time, but, if you are serious about 2x2 slide mounting, such a jig is thoroughly invaluable. It will increase your speed and accuracy in mounting, which, in turn, makes for accurate and enjoyable projection, without the need for projection adjustments after the initial projector alignment. Study the diagram and build your own as soon as possible!
Once you have mounted your slide pair by the near point method, have checked for both vertical and rotational errors, there is one last thing that should be checked – the infinity point separation. Referring to figure 2, this distance of the infinity point "a" should be no more than 1.5mm (the figure shows 1.43mm, but this is a bit impossible to measure) farther from the edge of the right image than it is in the left image. The theory on this is that with a superimposed near point during projection, this will mean that the infinity points, if actually measured on a typical 50x50" screen, would be about 2.5" apart, the same as the average human eye distance. If the infinity points were to get much farther apart the eyes would have to "toe-out" to view the image, and this is not easily possible or comfortable, and can be another source of eye strain.
At this point, whether the infinity points are too far apart has mainly to do with how the image was taken, and cannot really be corrected in mounting. In reality, the eyes do seem to be able to toe-out a little bit in viewing 3-D slide projection, but how much and to what degree has not been easily quantified. You will have to be the final judge, and the best method is by a final projection test for yourself, and perhaps for some family or friends who are willing to give you some honest feedback. See for yourself if there is any eyestrain, and ask your test audience for their opinion, also. If a slide ends up having too much parallax for comfortable projection then it really should not be part of a slide program, and should be rejected. In some cases a slide can be salvaged by creative masking, but if this is not possible save the slide for hand viewing only.
Projecting 2x2x2 slides with two projectors
Although this is not the main topic of this article, once you have mounted separate 2x2 slides you will want to project them. It is actually quite easy. Here are a few basics:
Get a matched pair of projectors, such as Kodak Carousel® or Ektagraphic® projectors. The Ektagraphic series is preferable, as it has a positive slide gate registration that means that the slide alignment is consistent from slide to slide. The projectors can be placed side-by-side or stacked one above the other. There is a bit less keystone distortion if they are stacked. Get or make an identical pair of "target" slides to use for superimposing the images from both projectors. Zoom lenses may be used, and target slides are essential for matching the image sizes. The film apertures from both images should superimpose, and focus should match. Polarizers should be mounted in front of the projector lenses. Taping them to the lenses is the easiest method. While projecting onto a silver screen (white will not work!!!) and wearing your polarized 3-D glasses, separately project with the right projector so that you can see the image on screen with your right eye and it is dark to the left eye. Then reverse the process and do the same for the left projector and left eye. Now you are ready to project your properly mounted 2x2 stereo pairs!