The Virtual Viewer

Problems with the new Virtual Viewer

I love the look and feel of the new viewer, and it's great that they are giving us a bigger image, but as anyone who has used it knows, it's hard on your eyes and a lot of images appear blurry. With this page I hope to explain why, and what can be done to rectify this. I believe this to be a major problem because I have actually seen it turn away customers. I witnessed a mother in Target choose not to buy a viewer for her child because it hurt her eyes when she looked through it and she was afraid it would do harm to her son. THIS IS BAD!

Design Problems

Many people can not fuse the images, especially when the picture includes infinity (as in scenic reels).

So what causes this to happen?:

This problem is caused by lens spacing (interocular). Basically the lenses are too close together and force your eyes to diverge when looking at far objects. Our eyes are not designed to diverge so many people can't make their eyes do this. Those that can, feel discomfort when viewing. The diagram to the left illustrates what your eyes have to do to fuse both images. (ouch). When we look at far objects in nature our eyes are parallel at most, and converge on close objects.

The correct way to space the lenses for a stereo viewer is to make them at least the same distance apart as the infinity points on the film chips. This is slightly wider than spacing them the same distance as the center of the film chips. This is true regardless of the separation of the viewers eyes. It's a common misunderstanding that the lenses should be closer together for close set (like for a child) eyes.

The example to the left is the correct way to space the lenses. This is even shown with a with close set eyes. The lenses are the same distance apart as the infinity point on the film chips (around 66mm), so even if the someone is looking through the inner part of the lens (as with close set eyes in the example) their eyes do not have to diverge.

The current design has an interocular of 64.8mm. This is even smaller than the distance of the film chips themselves (65mm). This means you have to diverge your eyes even for most close objects!

Other lens separation for comparison are The Model L at 65mm (low magnification makes it not a problem) and The Model D at 66mm (approximately the same magnification as the new viewer). These differences may sound small, but they make a big difference when magnified by the lenses of the viewer.

I have modified several Virtual Viewers to correct the lens spacing. It takes a lot of time, but the results improve the viewer tremendously. I hope to add instructions on doing this to my site soon.

Anyone who would like to read more about how stereo optics work please see chapter 5 of "The World of 3-D" by Jac. G. Ferwerda.

Manufacturing Problems

Many Virtual Viewers are shipped with warped lenses. This is easy seen by looking through the lens and moving the viewer side to side. With warped lenses you will see parts of the image change magnification as the move across the lens. As I understand this is caused by removing the lens from the molds before they cool completely. I have heard that this problem has been communicated to the workers, and indeed the lenses did improve about the time the first translucent viewers hit the stores, but since I have seen the problem creeping back.

A great number of viewers have badly installed lenses. Even viewers with the better lenses often do not have them installed all the way down on the lens posts and are sometimes even cockeyed. You can easily see this with the new translucent viewers without even opening it up.


Correctly Installed	Not seated all the way on posts

Design or Production Problem

I’m not sure if this is a design or production problem because it wasn’t in the early versions of the viewer, but it doesn’t appear to be limited to just one or two runs . The Problem is that almost all the current viewers tend to show a sprocket hole in the left image window.

I now think I know why. The reel rides on two curved plastic parts at the bottom (you can see this with a translucent viewer). The one on the left (Object A) is too thick or too high and causes not only the sprocket holes to show, but prevents the registration pins from both dropping into the sprockets (Object B). If you take apart the viewer you will find that only the right one is doing anything because the reel can't travel far enough down in the viewer to lock both in.

To confirm this I took a razor knife and trimmed the left one slightly (it just takes a very small amount) and bingo, this solved the problem. This also made the images easier to view because not only was the sprocket hole showing originally, but the left image was actually higher than the right image because of this.

Another reason I have seen for this happening is that the advance lever travels too far up when it is in it's resting position. If this is the case you may be able to make some kind of spacer to limit it's travel.