A while back I read some interesting posts from David Mullen, Jeff Kreines
and others about using an anamorphic lens in order to shoot 2.40:1 on
Super 16mm. Reading this past month's AC (Sept. 2003) set my mind wandering
back to the subject.
"I still have another problem
with [using an isco 1.5x projection adapter to shoot anamorphic
super16]. The 1.33x would be the ideal magnification, yielding
almost exactly a 2.39 frame from a 16:9 format. This is a
1.5x squeeze. One ends up wasting almost exactly the same
amount of image area on the sides as one would by shooting
straight 16:9 and cropping top & bottom for 2.39. That
plus the use of an additional distorting lens element. I see
no advantage to Aaton's experiment. I'd still like to see
a professional-grade large size 1.33x anamorphic element."
Does anyone have on hand one of the old Panavision variable anamorphic
I believe that they can go from 1x to 2x+, and can be varied to any degree
within that range. I'm not sure how horrible the image would look after
coming through one of these old adapters, but it seems like they could
be used to test a 1.33x squeeze for super 16 anamorphic. Also, it seems
that if the aperture size for S16 is .486 x .295, then that would make
an aspect ratio of 1.65:1, which would require an anamorphic squeeze of
1.45x to get to 2.40:1(even if the ratio is 1.66:1 this still holds true).
Why not use more negative space instead of using 16:9? If someone could
clear this up for me I would be grateful. I personally find the concept
of shooting anamorphic S16 alluring.
-Also, this is my second or third post to the forum, and I'd like to take
the chance to thank Geoff Boyle for taking his time to create something
that has been so helpful to budding DP's such as myself.
Years ago I worked on a commercial where the director wanted a mixture
of formats within 1 spot. The commercial was predominately spherical 4x3,
but with anamorphic sections dropped in. We were using Arri III's with
Zeiss standards. 435's weren't built yet! The rental house had a set of
anamorphics by Ultrascope. It was a long time ago, but I think the focal
lengths were: 28, 50, 75 & 120(?), I'm not sure about the 120mm. There
were 4 lenses & there was one longer than the 75mm! I do remember
them having a good quality about them, less prone to flare than some of
the C series anamorphics. A few things did really piss me off about them
though, none of the fronts were the same diameter, so filtration was cumbersome,
none had focus gears & the minimum focus on all of them was a disaster
where any kind of close up required diopters.
The lenses were all in Arri standard mount, so a PL adaptor can be used
for the older 35mm Arri cameras, but may be a problem with 435's. They
can be used on a 16mm camera because of the Arri standard mount.
The Super-16 Full Aperture is .493" x .292" (1.69 : 1) so it
takes about a 1.41X squeeze to get a 2.39 : 1 image.
There were the old non-2X anamorphic formats of the 1950's – Ultra
Panavision, which had a 1.25X squeeze, and Technirama, which had a 1.5X
squeeze, but most of those lenses are now collector's items and museum
A 1.33X anamorphic lens series would make a lot of sense all around, from
squeezing 1.78 onto 1.33 to squeezing scope onto Super-16, etc. However,
2X anamorphics are already a small subset of spherical lens manufacturing...
Cinematographer / L.A.
The biggest hurdle with using an anamorphot of 1.33x or 1.5x or any ratio
that isn't 2x is what to do with the footage in post. If you are simply
going to finish to video then various devices can be used to unsqueeze
the image, but then again it could be much simpler to crop the spherical
frame in this case. But if you're going to a film print, then you need
to get your Super-16 1.33x (or whatever) image to a 35mm 2x image. That
would require a very odd sized anamorphot indeed, one that is most certainly
not in existence. Or you'd have to add steps with unsqueezing one image
to a spherical Super-35 print and then resqueezing to a 35mm 2x print.
That means an anamorphot at three stages: on the camera, for the IP and
then for the IN. That's just begging for a lot of distortions and loss
of resolution, not to mention what's going to happen to the grain structure
of all those generations with grain that's been stretched and squeezed
repeatedly. Could be a funky look, but I don't know how desirable.
In the end I still say that the simplest, most straightforward and probably
best-looking way would be to just shoot spherical Super-16 and crop the
image in post.