Mailinglist Archives:
Infrared
Panorama
Photo-3D
Tech-3D
Sell-3D
MF3D

Notice
This mailinglist archive is frozen since May 2001, i.e. it will stay online but will not be updated.
<-- Date Index --> <-- Thread Index --> [Author Index]

Whatsplitter?



Bob Howard writes:

>I think I am in agreement with Marvin Jones that the mirror device does divide the lens cone into two beams to make stereo pair. 

Then maybe it should be called a lensconesplitter. :-/ 

>(e.g. covering
>half a lens with not divide the image to a half image, just stop down the lens..unless you are too far away from the front of the lens when it becomes a mask).

In my experience, putting a D-cap right in front of the lens _does_ produce just half an image at the film plane. This is the method used in making trick double exposures, putting the same person on both sides of the picture, and that sort of thing. This depends on the lens, I suppose. I'm sure you would be right if it were possible to get the mask close enough to the first nodal point of the lens, but on the normal lens with which I've done this, and I assume with most other normal lenses, covering half the lens produces a half image (or roughly half, since there is some fading off at the mid-frame border).

>So since any part of the lens forms a whole image. The 

If that were so, the mirror stereo attachment could never work. It only works because the left half of the frame "sees" what the lens sees through the right-hand mirrors, and vice versa. If it were true that "any part of the lens forms a whole image," then you would have two overlapping whole images, a photo mishmosh.

>device does split it into two images. BobH 

But if you take "a whole image" and "split it into two images," you will have two halves of the original image, ain't it? That's not what the attachment does. It produces two _separate_ images of the same subject through one lens, using different light paths for each image and (dare I say it?) converging fields of view so that they are images of the same subject. 

Without this convergence (i.e., if the mirrors were all at 45 degrees) you _would_ have "a whole image . . . split . . . into two images," one image with a little separation in the middle. Not a stereo pair at all. 







------------------------------