Читать книгу The Making of a Motion Picture Editor - Thomas A. Ohanian - Страница 3

A Brief History of Motion Picture Editing

Traditionally, for theatrical motion pictures, footage was captured on film. For television, footage was captured on either film or videotape. Eventually, of course, technological developments and improvements changed these workflows. Content for theatrical films can be acquired on film, videotape, or digital. Camera sensor technologies have and will continue to improve to the point that there will be no difference between film and digital acquisition. All of these changes are well documented in those countless resources. To keep this section somewhat brief, I have glossed over the specifics of the editorial process, but what is included herein is sufficient for a better understanding of the interviews.

Structurally, film “editing” was first done without actually physically cutting the film. The term, “editing in the camera” meant that in whatever order the final images were desired, they would have to be shot in exactly that order. In other words, you had to plan the shots that would be photographed in their final order. For example, actor “Ace” sees an apple, takes a bite, and his wife “Sue” reacts. First, Ace would be filmed. Then the shot of the apple. Then the shot of Ace taking the bite. Then the shot of Sue reacting. The film would then be developed (now you could see the film’s negative) and then printed to a positive (read: workprint) which resulted in a viewable image. That film was projected, and you would see the shots in the same order that they were filmed: 1. Ace, 2. Apple, 3. Ace bites the apple, 4. Sue reacts.

If any of the shots were too short or too long, so be it. They were shot that way and for that length. But if the shot of Sue was what we wanted to see first, how could that be accomplished? The answer is physically cutting the film. The four individual shots would be cut out of the printed film roll. Now, we would have four strips of film, each at the exact length they were when they were shot. At that point, we could rearrange the shots in any order that we desired.

Ah, but not so fast… What order should we choose for the shots? We have four of them. Those four shots yield 24 possible permutations:

1: 1 2 3 4; 2: 1 2 4 3; 3: 1 3 2 4; 4: 1 4 2 3; 5: 1 3 4 2; 6: 1 4 3 2; 7: 2 1 3 4; 8: 2 1 4 3; 9: 3 1 2 4; 10: 4 1 2 3; 11: 3 1 4 2; 12: 4 1 3 2; 13: 2 3 1 4; 14: 2 4 1 3; 15: 3 2 1 4; 16: 4 2 1 3; 17: 3 4 1 2; 18: 4 3 1 2; 19: 2 3 4 1; 20: 2 4 3 1; 21: 3 2 4 1; 22: 4 2 3 1; 23: 3 4 2 1; 24: 4 3 2 1

And that is just four shots. What about eight shots? Now there are 40,320 possible combinations. And what happens when the director has decided to do three takes of each shot? While we still have eight shots, we now have three choices for each of those shots.

Splicing film involved joining the film strips together using a liquid referred to as film cement or film glue. A small section of emulsion of both shots to be spliced together would be slightly scraped in order to make the “cement” adhere to the film workprint. Film cement was eventually supplanted by tape splicing where adhesive tape was used to join both film ends. A film splicer—known as a guillotine splicer—was used to cut the film. A piece of film would be laid into this metal guillotine splicer and a razor blade affixed to the vertical arm would be pressed down, cutting the strand of film into two pieces. At that point, two different film strands could then be spliced together, affixed with tape.

Obviously, cutting the film then led to not only the reordering of the different film shots, but also to be able to adjust the length of the shots.

The Magnasync Moviola from the 1960’s. Film viewer on the right and the two magnetic tracks (audio) on the left.

A two-picture head playback flatbed film editing system, particularly useful for A and B (two camera) scene coverage.

The film synchronizer is used to keep reels of film in synchronization with other reels. Picture and audio track(s) with common starts are thus kept in sync. Any change to the length of any track (shortening or lengthening) will result in the audio to picture synchronization being affected.

Videotape Editing

In 1956, Ampex, a U.S.-based company introduced a videotape recorder. Videotape editing was similar to film editing in that the videotape was actually cut. The recorded track could not be seen with the human eye. By applying a liquid to the videotape, the recorded tracks became visible. The videotape would then be cut in the same manner as film, but the joining of two pieces had to be done quite precisely. A solution was applied to the videotape, a microscope was used to see the tracks, and then they were aligned while being held in the splicing block, and then spliced together.

The Manual Splicing Jig from Ampex Corporation. Note the arrow indicating placement of the editing pulse.

Over time, the editing of videotape transitioned to electronically re-recording segments. Videotape was no longer cut and spliced. Instead, most often using the original source recordings, shot by shot was re-recorded onto a new videotape, creating the desired final sequence. While this became easier than physically cutting the videotape, the process was now linear—any change in the recorded sequence could not easily be undone—or reordered. Instead, the changes would have to be re-recorded due to the now linear process.

In other words, let’s say that you started to put together your program in this order:

Shot A then Shot B then Shot C

And then, you wanted to switch the order to:

Shot A then Shot C then Shot B

With electronic videotape recording, recall that videotape is no longer physically cut. As a result, the master tape (which has the re-recordings of the original tape), would be wound back to the end of Shot A (since that shot is not changing its position). At that point, Shot C would be recorded at the spot where the previously recorded Shot B had been and at the point where Shot C ends, Shot B would then be recorded. It is this “recording over” process that is necessary with linear, electronic videotape editing. It’s easy to see how time-consuming this can be when shots need to be re-ordered. And that brings us to some important terms that are specific to the editing of motion picture images:

Analog

Digital

Linear

Nonlinear

Random Access

Sequential Access

In the case of film, for example, we can classify the editing of film as being analog and nonlinear but not random access. Certainly, the film was not in digital form, so it’s analog. Editing the film could be done in a nonlinear fashion (it could be cut and re-ordered) but it was not random access. Because you need to move either forward or back through the film roll (and could not jump from place to place within the film roll), film is sequentially accessed.

Using these terms, we can classify videotape as being either analog or digital. It is linear (isn’t physically cut), and sequentially accessed (can’t jump around).

Electronic Nonlinear Editing Systems

To provide a solution which would attempt to combine the best of film editing (nonlinear) and videotape (ability to easily erase and record over), electronic nonlinear systems appeared. These were not digital, and they typically consisted of multiple videotape machines. The videotape cassettes in each machine contained the same material. Thus, by using Machine 1 to play back Shot A and Machine 2 to play back Shot B and Machine 3 to play back Shot C, the nonlinear aspect of film was achieved. We would see Shot A, then Shot B, and finally Shot C. If we wanted to change the order to Shots A, C, and B, Machine 1 would play back Shot A and Machine 2, instead of playing back Shot B would, instead, play back Shot C, and Machine 3 would play back Shot B.

The Ediflex Nonlinear Editing System. Note the bank of multiple videotape machines which provided a limited amount of random access to content.

The Laserdisc-based Systems

These systems introduced multiple laserdisc players instead of videotape players. Because the read head could jump around the analog video that was recorded onto the discs, fewer machines were necessary than the videotape-based alternative. Another significant benefit was that if an editor was creating a sequence which necessitated a greater number of cuts, there was a much better chance of the laserdisc machines being able to move quickly to the required shots. The laserdisc-based systems can be classified as electronic, analog, nonlinear and random access.

The CMX 6000 Nonlinear Editing System. Here, the bank of multiple videotape machines has been replaced with laserdisc machines which provided random access.

The Digital Nonlinear Systems

By the late 1980’s, the combination of computer technology, video compression, hard disk and optical disc storage systems led to the development of digital nonlinear editing systems. Video compression was used to reduce the size of each frame due to the fact that computer storage was quite expensive—$15 per megabyte. The images were compressed at 250:1 and the resulting resolution was quite pixelated.

The CL-550 JPEG compression chip from C-Cube Microsystems, 1990. This chip provided the hardware JPEG compression that was used by the first set of digital nonlinear editing systems.

Because they are digital, the methodology of creating a sequence out of the various shots is akin to how a word processing application functions. You can cut, copy, and paste words to modify a sentence, and a digital editing system enables shots to be easily rearranged and trimmed. These systems can be classified as electronic, digital, nonlinear and random access. Within a 7-10-year period after their introduction, these digital systems became the standard for editing motion picture content.

The Avid/1 Media Composer, circa 1990. The videotape machines to the right were used to play back footage which was then converted from analog to digital form and stored as digital files onto the computer hard drives to the left of the machines. The editing software ran on the Macintosh computer under the central monitor which displayed content in the form of thumbnail images.

As you read through these interviews, you will come across the various systems and different forms of workflows which these editors had to employ on their films. Knowing how manually intensive film editing is, or how time consuming it is to change your mind in editing film or videotape makes it all the clearer as to the amazing work that these editors accomplished. Imagine a couple of million feet of film, winding through it, cutting out the takes, splicing and trimming them and doing all of it manually. Two hundred, three hundred hours of film—10, 20, 50 takes of a shot—and now the unique aspect that is the craft of editing becomes clearer.

And through it all, the editor is busy cataloging and making mental or physical notes—the slight moment that an actor made an eye movement that may be of use later on or a shot that was stolen from another take to make a scene work.

There are many fine books and online resources that track the history of how motion picture images are edited. Rather than duplicate those efforts, this brief section provides the necessary background information to better understand the references over the course of the interviews.