Leitner's Cinematography Corner, No. 6

Check out Leitner's Cinematography Corner in its new home

For projects requiring high shooting ratios in the early 1980s, you could shoot 16mm or try on for size one of those newfangled "camcorders" from Sony, Panasonic, or Bosch: Betacam, Recam, or Quartercam. (Mid-word capitalization arrived with the dot.com era a decade later.) The first two featured 1/2in. videotape cassettes, the last, 1/4in. (Ampex in the United States, original inventor of video recording, also proposed 1/4in. helical recording, but never became a player.)

Siemens star of Sharp Max seen through a Carl Zeiss 28mm DigiPrime in viewfinder of a Sony PMW-350.
Photo by D. W. Leitner

Success of 1/4in. videotape, an idea ahead of its time, would await introduction of MiniDV in the late '90s, but the 1/2in. videotape camcorder took off from the starting gate. (Would you believe "camcorder" had to be coined by a reviewer? David Lachenbruch, longtime editorial director of the newsletter Television Digest, also coined "consumer electronics." Anyone know who came up with "prosumer"?)

1/2in. videotape camcorders, epitomized by Betacam, are the reason many of us first encountered the eccentricities and shortcomings of video zooms designed for electronic newsgathering. (Who came up with ENG? Or EFP, electronic field production, for that matter?)

Sharp Max attached to a Zeiss 7mm DigiPrime on a Sony PMW-350.
Photo by D. W. Leitner

In those days we sweated bullets over the precise mounting of our 16mm zooms from Angénieux or Carl Zeiss. 16mm was an exacting high-resolution format—see early '80s indie features blown up from Super 16 to 35mm such as Robert Young's The Ballad of Gregorio Cortez or Victor Nunez's A Flash of Green for confirmation of this—and precise lens mounting required a special device called a collimator, a technician who understood the principles of collimation (not to be taken for granted as it turned out), and deft use of metal spacers called shims (like washers used with nuts and bolts) to adjust the distance between the lens and film plane to within plus/minus 10 microns.

Collimators are like reverse telescopes. Bear with me and I''ll try to explain.

Business end of Sharp Max, with power button, brightness control, and compartment for 9V battery.
Photo by D. W. Leitner

When we see stars at night, they twinkle from such unimaginable distances that their light rays reach us as parallel rays. If you were to take a simple 100mm lens, aim it at the night sky, and adjust it to infinity focus, an image of the stars would come into crisp focus exactly 100mm from the optical center of the lens. That's the definition of focal length.

In the case of a 16mm camera, the idea is to mechanically mount the 100mm lens in our example in such a way that at infinity focus, the stars are sharp. If at infinity focus the stars are not sharp, then we must disassemble the mount at the rear of the lens and insert or withdraw a few thin shims until the precise required mechanical distance between lens and film is achieved. Then, if footage markings on the focus scale of our 100mm lens have been properly inscribed, we can set focus by tape measure and rest assured that once film is developed and dailies printed and projected, the image will appear in focus.

Anyway, that's the way it worked in the heyday of 16mm. (35mm works the same way, but 16mm tolerances are far tighter and more demanding, and the toll of bad focus is steeper, since grain, more pronounced in 16mm than 35mm, is always sharp, even when the image isn''t.)

Infinity mark on focus ring of Fujinon 7.6-130mm HD ENG zoom (ZA17x7.6BERM-M58H) travels past witness mark. Why?
Photo by D. W. Leitner

It's rarely convenient to wait until dark to train a lens on a star field to verify proper infinity focus, so collimators—reverse telescopes—were invented. A collimator is a tube with a miniature, rear-illuminated focus chart or starburst-like Siemens star at one end and a simple lens at the other, set to perfect infinity focus relative to the plane of the focus pattern. Instead of being brought to focus, the image of the focus pattern is projected in reverse out the front of the lens, its rays emerging in parallel.

If a camera lens, in turn, is set to infinity focus and pointed directly at the output of the collimator, it will focus the collimator's images perfectly—that is, if properly mounted to the camera.

Infinity mark on focus ring of Zeiss 7mm DigiPrime stops precisely at witness mark, as it does on all motion-picture lenses. Zeiss, world's oldest lens manufacturer, continues to set benchmarks.
Photo by D. W. Leitner

With 16mm, it's necessary to use an autocollimator to check focus. An autocollimator projects an infinity image into the lens, then by use of a partially silvered mirror and viewfinder, provides a direct look at the miniature image actually projected on film running through the camera. (Since each successive frame of film is unique, poor registration or field flatness can skew the results of examination by autocollimation, which is why I invented a stroboscopic autocollimator in the early '80s. But that's a tale for another day.)

Related Links

Leitner's Cinematography Corner, No. 3 Exhibit B: the new PMW-350, a groundbreaking 2/3in. shoulder-mount EX-series camcorder that records to solid-state SxS cards... Leitner's Cinematography Corner, No. 4 Cinema is scale. Last spring, I saw again Hitchcock''s gothic thriller Rebecca on the towering 40ft.-tall screen of a classic movie palace, Loew''s Jersey Theatre in Jersey City, N.J.... Leitner's Cinematography Corner, No. 5 Stereoscopy, or 3D imaging, has been around as long as photography, at least since1840, when the English inventor Sir Charles Wheatstone, who first explained binocular vision in 1838, fashioned his first stereoscope for displaying photos in stereo pairs...

Video camcorders, by comparison, can output the image that falls on the focal plane of their sensor(s) directly, so complicated autocollimation is avoided. But even so, collimation of videocameras and camcorders never caught on. In the early '80s, before CCDs, professional color videocameras and camcorders utilized three pick-up tubes for RGB, each of which was shock-mounted in rubber, grew hot in use, and routinely had to be mechanically aligned relative to the shared focal plane and each other. Electronically aligned, too, to match linearity, geometry, shading, etc. What a pain!

Unlike 16mm, NTSC video was not a high-resolution format—and this fact forgave many optical sins, not only poor corner resolution but degradations inflicted by 2x extenders (which you'll never find on a film zoom). All this changed with the advent of digital high definition. Suddenly video zooms had to match a higher standard of optical performance, one closer to 16mm.

This led to breakthroughs like the introduction by Zeiss of its DigiPrime series seven years ago, B4-mount video lenses that can arguably outperform any film lens. Which, in turn, upped the ante with regard to both precision mounting and backfocus adjustment of HD lenses—why Zeiss felt it necessary to introduce its own superb video collimation device, Sharp Max.

With Sharp Max, perfect backfocus at infinity is achieved readily, with confidence. The back-illuminated Siemens star seen in the camera's viewfinder leaves no doubt when best results are achieved. Since Sharp Max was designed as a complement to DigiPrimes, it attaches directly to them by means of a clamping ring. But I've tried holding it in front of a variety of other lenses, such as a Fujinon 7.6-130mm HD ENG zoom, which works remarkably well.

In a perfect world, video collimators akin to Sharp Max would be familiar and common location tools.

Don't mean to pull an Andy Rooney here, but have you ever noticed how video zooms focus past infinity? What's up with that?

And why are focus marks measured from the focal plane for all photographic systems, for more than 160 years—except video? Video M.O.D. (minimum object distance), for instance, is always indicated from the front of the lens.

Grist for future columns...