Re: The problem with Pulfrich

[P3D Larry Berlin <lberlin@xxxxxxxxx> To: photo-3d@xxxxxxxxxxxxxxx]

>John Bercovitz Comments:
>Actually, as long as we're being precise about words, panning is when you rotate the camera about a vertical axis, preferably approximately through the lens. This won't give you the Pulfrich effect because you won't get a different perspectives. There is a word for what you're talking about, but I can't remember it. At any rate, the camera has to translate relative to the objects (or vice versa) to get the effect. 

Well, I was being precise in the use of words. The camera was panning, rotating about a vertical axis, to follow an actress who was walking through a park area. The result was that she stayed relatively unmoving in the center of the picture while the background composed of trees, grass and other people, moved steadily past. That does in fact result in a definite Pulfrich effect if you are wearing the sunglass lens over the appropriate eye. If I had been watching it on a high resolution movie screen, she would have appeared to be flat like a cardboard cutout and still would have appeared to be walking *in front of* the other objects in the park. The reason for the flatness was that she was not walking in a straight line but in an arc which kept the same side of her in the same perspective to the camera.

The eyes don't really care whether the camera is rotating or translating or trucking or being carried in an airplane. As long as horizontal motion is present in differring amounts within the finished scene, the eye with the sunglass lens will perceive a delayed signal in comparison to the other eye and the effect is present. Camera motion is important but only relative to the subject motion. The whole thing is relative really, since it's only relative motion *on-screen* that will produce the effect. Whether it's the camera, the subject, the surroundings or some combination of these factors, it still boils down to relative horizontal motion. I think Einstein would have loved this one! ; -)

If everything is moving at exactly the same speed (no depth, or all-one-depth, with Pulfrich effect) and then they display a still title over the motion, the title appears to be either in front of the scene or in back of, depending on which eye is covered. 

>I think the cardboard cutout effect you're seeing is probably due to the usual cause which is poor resolution.

Yes, you are right. I hadn't thought that trying to see a pulfrich effect in the real 3D world around me was something I might want to try. I have yet to take my sunglasses to a movie theater. The most accessible form of motion imagery happens to be the TV set, which isn't a wide-screen model. The cardboard cutout effect would still apply most often unless a precisely accurate path of both camera and subject happens to take place, allowing for an accurate roundness to the subject. The scene I described with the airplane flying over some mountains had the necessary relationships and made fairly accurate roundness even on the TV. Part of my point is that current TV and Movie productions don't plan for inclusion of accurate Pulfrich renditions any more than they provide for stereoscopic 3D by other methods. 

In today's productions the camera moves all over the place and so do the actors and everything else. They are just as likely to show a scene which exhibits a positive pulfrich effect for the action but a negative pulfrich effect for a particular actor or object. Or objects moving in opposite directions but not really occupying different depths in the scene which places them at strange places when viewed with Pulfrich glasses. 

Larry Berlin

Email: lberlin@xxxxxxxxx
http://www.sonic.net/~lberlin/


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