Working with a Picture Editor – Film Speed and Video Speed
When the film is shot, the speed that the film runs through the sprocket holes of the camera is the “film speed”, 24 fps. Black and white television was originally shot with video at 30 fps.
Color standard definition video in North America is shot at 29.97 fps (Non Drop) for long format material, while standard definition broadcast television is shot at 29.97 (Drop Frame). HD video in North America is usually shot at 23.98 fps, although 24 fps (while less common) is sometimes used as well. Outside of North America, 25 fps is commonly used for both standard definition and high definition video projects. Generally the production audio recorder will run at a frame rate and speed matching the camera. Although 24 fps and 30 fps are different frame rates, their respective audio speeds are identical. While the standard 35 mm film speed protocol is 24 frames per second, the recent 2012 film, The Hobbit , was shot at 48 frames per second using dual RED Epic cameras on a 3D rig. The digital camera can be set to shoot at various speeds, including 23.976 fps, 24 fps, 25 fps, 29.97 fps, 30 fps, 48 fps, etc. Traditionally, and in the not too distant past, filmmakers shot movies at 24 frames per second, and then when the time came to edit the movie, the video and audio needed to be “pulled-down” (or slowed down) by 0.1%. All editorial work was done using videotape, with a standard definition speed of 29.97 fps, which is 0.1% slower than 30 fps. This transfer process from film speed to video speed is known as telecine. Both 29.97 and 23.98 are generally regarded as running at “video speed” in reference to the days when we used video tape for editorial. These video and audio speeds are in fact identical; only the sub-division of each second is different, as you are dividing each second into essentially 24 parts (frames) or 30 parts (frames). The speed difference between the shooting camera and the videotape was basically due to the fact that in order for the videotape to represent color picture, as opposed to black and white, it needed to be slowed down by 0.1%. Because current postproduction editorial no longer uses videotape, you would think that this slowed-down speed would be irrelevant. Not true: We still use the format of 29.97 fps of slowed-down video and audio. In fact, the most common editorial speed currently is 23.976 fps (also referred to as 23.98 fps), which is 24 fps slowed down by 0.1%. One nice thing about 23.98 video speed is that it divides the second into 24 parts or frames in the same way film speed (35 mm film) divided the second into 24 parts or frames, i.e. 24 frames per second. In Pro Tools we can set both the 23.98 timecode ruler and the 23.98 film ruler in the edit window and the frames will line up. Even with the advent of digital, there is still a protocol of shooting at film speed and essentially pulling-down for postproduction.
The final step after editing the entire picture, all of the audio, mixing and print mastering the film, is re-converting (using a 0.1% pull-up) the audio back to its original film speed of 24 fps, so that it can be properly synchronized to the film print or Digital Cinema Package (DCP). Increasingly, more and more films and television shows are being initially shot digitally at 23.976 fps, remaining at 23.976 fps throughout the entire editorial, mixing and print mastering process, avoiding the need for all but one audio pull-up. This takes place at the very end of the postproduction process, when the final printmaster audio stems are converted back to a 24 fps film speed by applying a 0.1% pull-up, so that the audio synchronizes with the final film print or Digital Cinema Package (DCP) master.
When converting audio between 23.976 fps HD video projects and 29.97 fps standard definition video and broadcast television in the U.S., no audio pulls are necessary, despite the frame-rate difference, which greatly simplifies this process, saving both time and money.
Although there are multiple pull-up and pull-down settings in use on film projects worldwide (such as pull-up by 4% for converting from 24 fps to 25 fps, as would be necessary for PAL, the European standard film speed) most film project workflows have transitioned to utilizing fixed-frame rate settings and sample rates with no audio pulls for the entire postproduction process, in order to avoid the confusion and potential mistakes that had occurred in the past.
To check these audio and video settings, first communicate with the picture department or picture editor, then view the file codec type in the QuickTime software to make sure the settings in Pro Tools are correct, and finally verify that the Pro Tools main timescale display matches the window burn on the digital picture (see images below).
Working with film projects that require audio and video pulls can be confusing, and are not the only valid ways of working.
Readers should familiarize themselves with all pertinent technical information dealing with postproduction video. It will also be important for the music editor to fully understand pull-up and pull-down. There are many technical documents online or in books, that explain these two important aspects of postproduction picture and audio in more detail.
Excerpt from Music Editing for Film and Television by Steven Saltzman © 2014 Taylor & Francis Group. All Rights Reserved.
About the Author
Steven Saltzman, MPSE is a music editor and composer based in Los Angeles, CA. He received his Bachelors of Music in composition and film scoring from Berklee College of Music and is a certified Avid Pro Tools instructor. He has been editing music for film and television for the past eighteen years. In addition, Steven has lectured nationally, and he has created and taught numerous music editing courses. A recipient of a Golden Reel Award for music editing, Saltzman is also a member of the Motion Picture Editors Guild, the Society of Composers and Lyricists, and he sits on the board of the Motion Picture Sound Editors Guild.