Optics question re: anamorphic

[Chris Keth]

So I watched Transformers this evening and wondered something. Perhaps someone here can explain the optics. Why out of focus points are tall ovals rather than short ovals?

 

Let me pose my guess: An out of focus point takes the shape of the iris. We all know this from observing and reading about the MK2 superspeeds with the wierd triangular irises. I would postulate that out of focus points take the shape of the iris as you would see it if you were looking through the lens from the film plane and since the anamorphic lens squishes everything side-to-side the apparent shape of the iris is squished to the tall oval we all love and know. Logical?

 

I'm really wondering and maybe someone can explain it :)

[Jon Kukla]

With an anamorphic lens, you have - essentially - two lenses: a prime lens and an anamorphic lens, which are used together in conjunction. (In the early days, some of the anamorphic lenses actually were primes with anamorphosers that needed to be added and calibrated by hand everytime a lens was changed.) In short, you have your spherical focus and your anamorphism "focus" for the power of the anamorphic element.

 

The old CinemaScope lenses simply set the anamorphic element's focus to maximum "DoF" (I'm putting quotes around these terms bc the technical optical concepts are slightly different) - which is why they restricted subjects from approaching closer than about six feet.

 

Panavision's unique concept was to use dual rotating anamorphic elements to maintain a constant 2x anamorphic power which could also track with the focal plane. This eliminated the so-called mumps problem of close subject matter, but also made the falloff more pronounced. The vertical oval bokeh is thus the result of the out of focus background being subjected to a higher than 2x anamorphic power, which makes it look over-squeezed.

 

Make sense?

[David Mullen ASC]

The ASC Manual has a good article on the history of anamorphics, I think written by John Hora.

 

Apparently as you focused near minimum on the original CinemaScope lenses, the amount of compression changed, was less than 2X. This caused faces to look fat when uncompressed by the constant 2X by the anamorphic projector lens. Panavision redesigned the lenses so that the point of focus area was always a 2X squeeze, but the result is that the background when out of focus now gets more than a 2X squeeze, thus still looks skinny even when unsqueezed.

 

But the ASC Manual article explains this better.

[Chris Keth]

That makes perfect sense, now that you have me thinking back to reading about the original cinemascope gear. Thanks guys. I'm going to check out that article. I forget that the ASC manual is more than just reference sometimes :)

[Leo Anthony Vale]

Apparently as you focused near minimum on the original CinemaScope lenses, the amount of compression changed, was less than 2X. This caused faces to look fat when uncompressed by the constant 2X by the anamorphic projector lens. Panavision redesigned the lenses so that the point of focus area was always a 2X squeeze, but the result is that the background when out of focus now gets more than a 2X squeeze, thus still looks skinny even when unsqueezed.

 

Cinemascope and other anamorphics, including the Delrama prisms on Technirama, with the same type of focus have the same elliptical out of focus points.

 

The reason is the anamorphic system has two focal lengths, a horizontal and a vertical.

The depth of field is different in the vertical plane and in the horizontal plane.

The plane with the squeeze has the greater depth of field.

[nathan snyder]

Since the subject has been brought up can anyone explain how anamorphic primes with a front anamorphic element overcome the above-described astigmatic problem?

 

When I first got interested in anamorphic lenses I fooled with a couple of different anamorphic lens blocks. One was a dicromatic prism pair, the other was the lomo 35mm lens but just the anamorphic bit without the prime. I would set these anamorphic elements up in front of a prime and shoot test targets. I found that, for the most part, that when using the prism pair and a prime, like a 50mm prime, I would indeed get two distinct focal distances. This was most evident when I would shoot a simple star chart. I would set the prism pair to be as close to a 2X squeeze as I could get and then focus the just the horizontal lines. Well, of course the vertical lines would be out of focus, but with out touching anything on the lens I could move the star chart and find a distance where the vertical lines in focus but then the horizontal lines would be out of focus. I did not have as much troubled with the lomo lens block. I could do the same setup but replace the lomo anamorphic element for the prism pair. But here I would actually use the distance markings on the focal ring on the lomo lens. I did find that the prime lens markings were not 100% accurate for the sharpest focus. I still had slight fuzzyness with either the horizontal or the vertical focus. My solution was to stop down until the each of the focal planes depth of fields became wide enough to overlap. Not a great solution.

 

Later I purchased a complete lomo 80mm anamorphic prime. Both the horizontal and vertical focus appears to be on the same plane. The lens has fairly good focus even nearly wide open. Can anyone explain how the astigmatism is overcome?

[Kenny N Suleimanagich]

I've noticed in Chinatown during some focus racks (once inside Gittes' office once outside Mulwray's house) theres very noticeable compression/decompression. As far as I've noticed. There might be a few I missed but these stood out to me. Anyone know which anamorphics they used?