December 1, 1990

Taking 16mm into the '90s

If 16mm film production once seemed threatened to near extinction by small format video (and in news gathering, at least, this is indeed a fait accompli), three technical innovations promise to take the small format film robustly into the '90s.

How does this sound for your next documentary: an inexpensive, well implemented film stock suitable for high-definition distribution, with machine readable code for easier posting, and digital stereo optical track?

Like 16mm in the 1990's, perhaps. Not with standing already accomplished advances in 16mm film production—including Kodak 's tabular grain stocks, which with 7245 introduced a 50 ABA daylight film to the growing arsenal of small weaponry—there 's always the future. And for 16mm in particular, the future is not inconsiderable.

The first development, and possibly the most immediately relevant, is Eastman Kodak's introduction of Key­ code, a substantially revised and very promising variation on its failed and misunderstood Datacode. Negative matchers and film labs have long relied upon the edge numbers reproduced on film negative by all manufacturers. Because the edge number is unique to any particular foot of film, it should serve film post production in the manner of SMPTE timecode, the universally utilized 80-bit-binary-coded decimal that made electronic video editing take off. But so far edge numbers haven 't served film in an automated way because i t is not machine readable, which obviates electronic interface. Datacode was an attempt to layer gamma-ferric oxide in a thin magnetic stripe to convey the edge numbers to an intelligent reader. It worked well for 16mm stocks, too, but adding the stripe microscopically increased the negative 's density, decreased the signal passing through the film, and most importantly, gave filmmakers all the excuses they needed to reject it.

Keycode should do for film negative what the barcode did for the checkout counter: the price tag (edge number) will still be there for the analog reading method, but a prestrip­ing of optical code—which repeats the longitudinal address inherent in the edge number, every 64 perforations—will make a more sophisticated read­ out possible without adding density. As gravy, other information such as the stock number being used will also be printed in a separate opcode. This newly available information can greatly enhance post-production processes, particularly negative matching and conforming, certain telecine operations, and color timing. And there's still room for timecode, if the film community shows sufficient interest in electronically editing film. But for now this means documentarians who have been tempted to shoot film, edit video and conform later, may find themselves with the final task almost automatic and trouble free. Kodak expects the 16mm of Keycode in the second quarter of 1990.

The second two developments affect film exhibition, by both tape distribution and theatrical presentation, but are far less imminent or even inevitable. But if a company called Optical Radiation Corp. ( ORC) sounds as if it will take technology into the 90s, that's not a bad guess when it comes to 16mm. This Azusa, California-based company is leading a development effort toward creating optical soundtracks on film with infrared lasers, and is currently co-developing with Kodak a six-channel digital optical track for 70mm film which should hit the market this year.

The 70mm sound's specification—44.1 KHz sampling using 10 bit word size, and achieving a frequency range of 20 to 20,000 Hertz—is not, nor should it be, the design goal for 16mm stock, which simply needs an escape route from simple magnetic sound ­ tracks to be freed up considerably. But the technology used to create the

70mm version may be migrating downward, first to 35mm, then to 16mm stocks, depending upon the demand. "It is possible," says Richard Wood, ORC president, "in fact, highly probable, if there is a market for it."

Frustrated filmmakers know all too well what that market is—film festivals, large public presentations, special screenings, for which 16mm mag stripe has often sounded insuffi­ cient. Wood says that to create the sound for 70mm, requires about 5 megabytes per second of information to be optically read and translated. And all that must occur within a narrow channel width—a tenth of an inch wide by 24 frames long. The 16mm version, he says, will probably not be six channels, so it will need less than a megabyte of information, but warns that there is a certain amount of redundant data that must be carried by a typical sound system, regardless of the print size. And, he says, ORC, which has already spent S4.5 million on research and development in this area, would have to build custom chips for a 16mm project. Because the track width is significantly smaller, and the length of 24 frames of 16mm shorter than larger formats,the translation of the code containing the sound must be quicker, the sound head reader must be more precise. The computer processing task is uneviable. Other crucial issues concerning 16 mm's robustness as a format have yet to be resolved.

Still; it might be only a year and a half away. "And, as a technology, optical printing has its advantages, too," Woods adds. "For instance, it's relatively inexpensive. And magnetic striping tends to delaminate and lose its highs after 200 passes or so. The success of the compact disk has driven a market beyond the analog track."

Finally, and again from aggressive Kodak, came an announcement at the fall conference of the Society of Motion Picture and Television Engineers, of a startling breakthrough in t transferring conventional film to digital high definition tape. Principally

driven by a new linear array of charge coupled devices (CCDs) developed in Rochester and Kodak's labs in Harrow, England, the new telecine was producing high-definition images from film many thought rivaled those created by high-d ef video cameras, which were themselves improved considerably by their manufacturers.

The revolution here may be economic in base. Whereas high-def video cameras have gotten more expensive as they've been improved, the Kodak experimental telecine promises that any film imaging system, including 16mm, can be utilized for high-definition production. In fact, Kodak demonstrated the result of a transfer from 16mm negative to high-def tape, and the resulting image was quite acceptable, even under less than ideal conditions. (The footage had not received secondary color correction—which can rectify a number of small image sins—and the transfer was made only after the negative had been blown up to 35mm.) One of Kodak's engineering team leaders, Brad Hunt, remarked that a 16mm gate, which would facilitate a transfer without generation loss, was upcoming, and that 16mm production experiments were underway. The potential this holds for documentarians is obvious. Aesthetically, they may begin to rethink their allocated space to accommodate high ­ def s. wider aspect ratios. (Since television has served as their principal source of revenue in recent history, not shooting in 4:3 may have seemed pointless, even reckless.) And, as the telecine promises to output to any international standard, filmmakers cognizant of (and reliant upon ) foreign markets can rest assured that 16mm will prepare them for the advent of HD-MAC, or another version of advanced television broadcast.

The implications bring an almost giddy relief: an old Bolex camera rental receipt may be our ticket into the next millennium.


Gregory Solman is a freelance film critic and the West Coast Editor of Millimeter magazine.