banner Toyota
January 28, 1996

The All-Digital Movie? Not Quite Yet ...

In an age of microchips and CD-ROM's, it's galling to think of plain old celluloid film -- which was birthed, messy and chemical, more than 100 years ago -- as a high-tech medium. As the millennium approaches, surely we should be shooting our movies with electronic cameras, storing them in digital jukeboxes, sending them to theaters via fiber optics, and displaying them on huge liquid-crystal screens. Right?

Actually, it's surprisingly hard to beat celluloid, even with today's cutting-edge technology. Consider this: A single frame of film contains at least 40 megabytes' worth of information. To store an average feature-length movie digitally -- say it's 100 minutes long -- you would need almost 6 terabytes of computer-disk space.

To anyone but NASA, that's an awful lot. By comparison, many home computers have a hard-disk drive that stores 250 megabytes. Stack a whopping 23,000 of those drives together and you've got the equivalent of a few reels of old-fashioned 35-millimeter film.

And that's just on the storage side. There's no such thing as a viable digital movie camera yet; indeed, still photography is just starting to get digitized, and that's relatively simple. To capture motion pictures, technologists must find a way to gather one gigabyte of data -- that's four home computers' worth -- every single second.

Meanwhile, liquid-crystal-display movie screens are out of the question; it's a struggle even to build television-sized L.C.D.'s. And cinema-quality digital projectors, which are based on military technology, cost more than $100,000 a pop. Think about how few cinemas have installed digital sound -- a far cheaper proposition -- and you'll realize this transition will be no picnic.

Still, it is virtually certain that celluloid is doomed in the long run, and for a very simple reason. Computer technology rides a predictable curve: Every 18 months or so, it gets twice as powerful for about the same price. Viewed another way, the cost of achieving any fixed goal drops by half every 18 months.

Within a decade, then, digital cameras, projectors, and storage will be not only possible but affordable. And from that point on, they'll continue to improve. The only question, really, is how quickly these new methods will be adopted.

For better or worse, the current bet in Hollywood is that it won't happen anytime soon.

Panavision International, for example, is the leading supplier of cameras to the movie industry. It doesn't plan to begin research on digital cameras for five years or so. "It's not at that stage yet," says Phil Radin, executive vice president of marketing.

There's a better chance that movies will be distributed digitally before long: Hollywood studios are eager to stop paying $1,500 or so for each film print they strike. In the United States alone, prints cost $3 million for every wide-release movie. Shipping digital tape -- or zapping movies to cinemas via satellite or fiber optics -- would be much cheaper.

Yet such a move would require theaters to buy new communications gear, some kind of big storage device, and several digital projectors. Upgrading all the world's cinemas could cost more than $100 billion, according to some estimates. Who would pay?

"I find it hard to believe that theater chains will capitalize the next phase of movie technology," says Nick DeMartino, director of strategic planning for the American Film Institute. "When cinema owners upgrade a theater, they usually just buy better seats."

As a result, newfangled distribution will likely be tried at first only by companies like Sony, which owns both ends of the transaction -- a movie studio and a theater chain.

Meanwhile, the biggest push into the digital realm will continue to come between the initial shoot and the final screening. More and more filmmakers are scanning footage into digital form so they can use computers to edit their movies and to mold images a la Forrest Gump. Soon this trend will relegate celluloid to a medium that's used for input and output -- but nothing in between. And once that change has been digested, people inevitably will think about closing the loop with digital cameras and projectors.

Going fully digital will have its drawbacks, however. For one thing, it will require constant, costly upgrades: Like PC's today, movie cameras, projectors, and even screens will become obsolete every few years.

And while film archivists fret about celluloid deteriorating over decades of storage, digital methods could be far worse. After all, you can still run a 1940 reel in a state-of-the-art 1996 projector, while computer disks from 1980 are useful only as mouse pads.

"And if film deteriorates, at least you can play it back and see exactly what the damage is," says Nick DeMartino of the A.F.I. "You might not be able to read a damaged disk at all."

In short, celluloid still has a lot going for it. But even Eastman Kodak, the granddaddy of celluloid, knows a transition is inevitable sometime. While the company keeps improving the technology it developed in tandem with Thomas Edison a century ago, it's also hoping to be a leader in all-digital cinema. Not for a while yet, though.

"People always say, 'One of these days, something will change,' " says Joerg Agin, Kodak's president and general manager of motion picture and television imaging. "That's probably true. But I think it's premature to predict the demise of film anytime soon."

Home | Sections | Contents | Search | Forums | Help

Copyright 1996 The New York Times Company