Postproduction Workflow

Workflow is the technical path your project will take from sound and image acquisition to exhibition. This includes choices about the shooting format, the editing format, editing software, finishing processes, your mastering format, and your final distribution formats. Filmmakers can easily lose their way in the workflow stream, with unexpected and often expensive results. A little bit of research at the very beginning into the ways various production and postproduction phases and formats interface with one another will go a long way toward minimizing nasty surprises. Doing some tests using the various types of footage you will be working with and taking them all the way through postproduction is advisable. At the very least, talk to other filmmakers and postproduction professionals, including any facilities you’ll be working with, before you begin postproduction.

The postproduction workflow is to some extent based on decisions you’ve already made in production, including what camera you used, what format you shot in (and what format other footage such as archival material comes to you in), and how your sound was recorded (double- or single-system). Your postproduction workflow is also based on an understanding of your desired distribution venues (Chapter 23).

At this point in the evolution of the medium, there are a wide variety of postproduction pathways one can take, and each option has a significant impact on the budget of the film, your work process, your documentary’s look and sound, and ultimately on your range of exhibition possibilities. We are also at a time of big transitions in terms of available digital nonlinear editing (NLE) software tools. A few years ago, Apple’s Final Cut Pro (FCP) and AVID’s Media Composer dominated the documentary editing field. Today, with Apple’s move to the more consumer-friendly Final Cut Pro X, documentary makers and editors who used FCP have been migrating, often to Adobe’s Premiere Pro or to Avid’s Media Composer. For the near future, things will continue to be in flux, but a good grasp of postproduction basics, as we are providing below, will help you make the best decisions for your project regardless of the specific system you’re using. For the purposes of this book, we will use examples and figures from both AVID’s Media Composer (Avid) and Adobe’s Premiere Pro (PP).

When devising your workflow, you should have answers to the following questions:

1. What was the film shot on: SD, HD, 2K, or 4K? What codec, resolution, and frame rate? Did you shoot single- or double-system sound? (Chapter 8)
2. What editing software will you use? Will you be using specialized ingest software (p. 274) or syncing software (p. 277)?
3. Will you work at highest resolution throughout, or save space and RAM by using lower-resolution versions of your files (proxies) and linking back the higher-resolution footage at the end? (p. 276)
4. Do you want to transcode all your footage to a single editing format at the beginning of editing, or will you work with mixed formats and standardize at the end? (Each approach has advantages and disadvantages; see p. 275.)
5. Will you edit and mix sound within your editing system, use specialized software to do it yourself, or rely on a specialist to finish the soundtrack? (Chapter 21)
6. Will you do the color correction (also called grading) on your own, or do you plan to use a professional colorist? Will you need specialized third-party software if you do it yourself? (Chapter 22)
7. How do you want to master the project? (Chapter 22)
8. Do you want your master to be mono, stereo, or surround sound? (Chapter 22)
9. How do you want to distribute the project (Chapter 23) and what media types will be required for exhibition?

Finally, if you are delivering the program to a broadcaster or other type of distributor, you need to make sure you understand what their specifications are so you deliver everything correctly.

Let’s look at a typical low-budget project shot on HD that aims to finish on one of the HD digital video formats with the intention of screening at festivals and streaming on the Internet (Figure 17.1).

■ Figure 17.1 Low-budget workflow. Shoot HD video, edit and finish in your nonlinear editing system (NLE), master as a file for festivals, DVD, and web. Master audio is stereo.

■ Figure 17.2 Medium budget workflow. Shoot HD video, edit and lock picture in your nonlinear editing system (NLE), send to professional sound edit and mix, and professional colorist. Master as a file for festivals, DVD, and web. Includes surround sound for Digital Cinema Package (DCP).

For a low-budget or student project, you will likely do some of your own color correction or sound finishing in your NLE. For a higher end project, more suitable for broadcast or theatrical release, you will actually be exporting both sound and picture and taking them to a mix session and a color correcting session, respectively (Figure 17.2).

The Basic NLE System

There are purely consumer programs, like Apple’s iMovie and Sony’s Vegas, which are inexpensive and relatively easy to learn. These programs are not flexible enough, though, for anyone who seriously wishes to make accomplished documentary films. The more professional programs, like Avid’s Media Composer and Adobe’s Premiere Pro, are more expensive but have much greater capability in terms of editing sound and picture, handling large projects, and integrating with professional finishing workflows for color correction and sound mixing. All professional NLE programs tend to work on the same basic principles and use similar interfaces.

The Hardware Setup

While small projects can be completed in on a laptop computer (Figure 17.4), most editing setups use a desktop computer. The hardware for a typical NLE station consists of a computer and monitor(s) (Mac or PC) running NLE software, external hard drives, a media transfer device (camera or card reader), speakers and/or headphones, and a comfortable chair (Figure 17.5). Editing requires a fairly powerful personal computer. Many editing setups will use two monitors, with one (often an HD monitor) dedicated to a full-screen display of the program.

■ Figure 17.4 The convenience of today’s editing software. Two students do some last-minute cutting in the hallway before their class begins.

External hard drives store your media files, and connect to your computer using either Thunderbolt or USB 3.0. Most NLE software instructions recommend not putting media files onto the internal drive of your computer, where your software resides. This is a speed issue and also a convenience issue. Media saved on a portable hard drive can be used with any editing station running the same software. Keep in mind that media files take up a lot of space, so you must calculate storage needs before you start working with your footage. Storage space depends entirely on your editing format. Five minutes of 1080/60i HD footage in codec like Apple ProRes HQ will require 5 GB of drive space, while 5 minutes of AVCHD footage takes up much less room. Any number of free format/storage calculators can help you figure out the storage needs for your format’s codec and data rate (p. 132).

■ Figure 17.5 A typical NLE system setup (in this case, Annukka Lilja editing Deirdre Fishel’s Care): computer (A), external hard drives (B), speakers (C), and a comfortable chair (D).

In addition to holding your original media files, your NLE system will create new media called render files for a variety of reasons. It’s essential that you have plenty of extra space for these files. It’s also important that your hard drive have a spindle speed of 7200 rpm. Many less expensive drives run at the slower 5200 rpm speed, which is inadequate for playing back HD footage. Currently a new type of solid-state drive (SSD) can offer even higher data transfer speeds, but the high price and small size makes this less likely to be found in an editing context.

A media transfer device is used to download your footage into the NLE system. Again, this depends on your shooting format. If you’re shooting on HD tape, then you’ll require a deck interfacing with the computer, or you can use your camera. If you are shooting on memory cards, you’ll need your camera or a card reader. Note that some computers come with a reader installed. Often footage will be transferred or transcoded onto a drive in the field by a Digital Imaging Technician (DIT).

A good set of speakers is important to get a true sense of your audio. If you are working in a room where there are other people, you’ll be forced to edit with headphones. Headphones are fine while you’re constructing the rough cuts of your movie, but the final soundtrack mix should always be done with high-quality speakers to get an accurate sense of how the balance and presence of the audio will sound to an audience who will be listening to your documentary through speakers. In particular, problematic bass sounds that you will be unaware of listening to headphones will become obvious when your tracks are played through good studio speakers.

A comfortable chair is a must. Editing requires that you remain inside an editing room, sitting on your behind, in front of an NLE system for days on end. Under these conditions, it’s not uncommon for repetitive stress injuries to occur. For this reason, an adjustable ergonomically correct chair is important.

Setting Up Your NLE Project

When you create a new project in your NLE, you will need to set it up so that the program understands the basic parameters of your project. One of the most important settings is your format, which relates to the format of the material you will be importing and how you want it to be converted when you bring it into your NLE (Figure 17.6). Your scratch disk (PP) or capture drive (Avid) determines where your NLE will put any files it creates while importing or during editing. It’s important to set these correctly every time you edit, or you run the risk of spraying footage throughout your system and ultimately losing files. Finally, your sequence settings determine the format of any new sequence you create.

■ Figure 17.6 Settings dialog box in Premiere Pro (left) and Avid (right). Here you can see the choices available for setting up a project in a variety of formats.

Ingesting Your Video

These days, you will most likely be shooting on memory cards (SD, CF, P2, or SxS) or onto a hard drive. You will bring your material into your NLE environment via a process called ingesting. Once you’ve inserted a card into your computer or card reader, you have several options. One is to use the computer’s operating system to copy the entire contents of the card, including its file structure, onto your hard drive. You can then bring the media files right into the NLE working environment, or convert them into another format for editing using your NLE software or another application. In other situations, manufacturers will provide software that reads the file structure on the card, and copies the files onto the hard drive, wrapping them in the process. An example is the Canon XF utility. Wrapping involves putting the media into a file format that the NLE can understand and work with, without changing the original shooting format encoding. Premiere Pro prefers files wrapped in Quicktime or .WMV (the Windows container format) wrappers. Avid, on the other hand, saves all media in the Material Exchange Format (.mxf) container format.

■ Common Source Media Formats

Many cameras that shoot in the AVCHD format use a file structure based on the one developed for Blu-ray discs. Material shot in this format is actually not saved as discrete files, but as streams, which lack the structure of normal files. Some of the information, like time code, is actually stored in a separate folder (Figure 17.7). This means that the media typically has to be wrapped in a container format—in other words, turned into individual files (.mov for Premiere, .mxf for Avid). Different cameras generate media with different file structures, and they may also have to be rewrapped or transcoded for use. NLE and camera manufacturers make it their business to offer updated software addons and plug-ins that will allow the newest camera files to be ingested. Before you start trying to bring in any media, you should make sure your software is up-to-date and set up to bring in the media you are working with.

■ Figure 17.7 The file structure of an AVCHD disk. The media is on the right in the Stream folder as .mts files.

Depending on our workflow, your media may be need to transcoded. Transcoding means converting the media to a new codec that the NLE is optimized to work with (from AVCHD to Apple Prores, for example).

Once your material has been ingested, it will appear in your NLE as clips (p. 284).

Working Natively vs Transcoding

Increasingly, NLEs (including PP) can handle a variety of codecs natively. This means that they can be ingested as they are, without transcoding. When you determine your workflow, you will need to decide whether to work natively or transcode all of your footage so that it is in one consistent codec. Documentary editors often face a “codec salad,” meaning that some footage was shot on one camera, other footage on another, archival material is in another format altogether, and so on. Avid will always convert to its own proprietary codec and wrap in an .mxf container.

The advantage to working natively is that you don’t have to spend a lot of time early on transcoding to a single format. The disadvantage of mixing formats is that long sequences can bog down your computer as it tries to process the many codecs in real time. Your finishing may also be more complicated and can reveal unsuspected problems. There is no consensus among editors at this time as to which approach is better. We suspect that, as software and processing speeds improve, working natively will become more common.

Frame Rate Considerations

If you are working in NTSC, you will have shot your footage either at 24 fps (actually, 23.97 in most cases) or 30 fps (actually 29.97). Hopefully you will not have combined both in one project! You now have to decide whether to edit in 23.97 or 29.97. The important thing is to work backwards from your intended delivery formats. You cannot mix frame rates within one sequence, so you will have to either convert your 24 fps material to 30 fps, or vice versa.

When video cameras capable of shooting 24 fps first came on the market, they were incapable of actually recording at anything except the broadcast standard of 29.97 fps. This problem was solved using an approach inherited from the days of film-to-video transfers, called 3:2 pulldown. The DSP in the camera would repeat certain fields (half a frame, remember!) so that the 24 frames filled the 29.97 timespace. Today, many HD cameras have the capacity to shoot 24p native (24Pn). XDCam and AVCHD, for instance, both include Pn formats. The implications of this can be a bit complex. The most important thing to think about is your workflow, which can go a couple of different ways for productions that shoot 24Pn. One alternative is to edit in 24P and do a pulldown at the end to create a 29.97 fps master for broadcast. Alternatively, you could do the pulldown right away, and edit in 29.97 fps to make a broadcastable master. Then you can reverse the pulldown process to make 24 fps versions of your show for DCP or Blu-ray, both of which support that frame rate.

Working with Proxies (“Off-Line” and “Online” Editing)

Not many documentaries are shot on high-resolution formats like 4K, but this is likely to change in the near future. And even in HD, many documentaries have more than 100 hours of footage, resulting in terabytes of needed hard drive space. Sometimes, your NLE will have a hard time handling your high-resolution files. In any of these cases, it may be advantageous to create proxies, which are lower resolution versions of your high-resolution media files. Proxies take up less storage space and are less demanding in terms of computing power. The idea behind proxies is that you edit with them, and then can easily go back to the high-resolution original files and cut them in after you have finished making your editorial decisions and your cut is locked (p. 292). Using proxies for editing is also called off-line editing, which refers to an earlier time period (before about 1995) when all editing was done this way. Similarly, recreating your cut using the high-resolution footage is called online editing or conforming.

■ Ingesting with Specialized Software

Many workflows involve using special software, either in the field or as a first step toward setting up postproduction. These software systems allow you to ingest clips, transcode footage, create subclip markers, add metadata, and even back up your material while importing your media into your NLE system (Figure 17.8).

■ Figure 17.8 Adobe Prelude is a software companion to Premiere Pro that allows for ingest, transcoding, and logging in the field.

If you are working with AVID, the system will auto matically transcode your source material into the AVID DNxHD proprietary format. You can select from a variety of compression levels, expressed as bandwidths (Mbps). A high quality level would be DNxHD 220 (10-bit, full image quality, with a 220 Mbps bandwidth), while a compressed proxy could be DNxHD36 (8-bit, lower image quality, with a 36 Mbps bandwidth).

Cameras are increasingly capable of shooting high-bandwidth files (4K, for example) that are too big for edit systems to handle, so the use of proxies will likely continue for the near future.

Ingesting Audio

Ingesting audio may seem easy, but in fact there are several things you need to be aware of to avoid problems. The first is to make sure that you are ingesting at the same sample rate that you are using in your project’s main sequences. If you recorded single-system, it is likely that your audio will have a sample rate of 48 kHz. But some cameras have other options and if you are using a separate audio recorder, you may have elected to record at a higher sample rate, such as 96 kHz, and your editing software may or may not be capable of converting this on the fly. Stereo is another key issue. If you recorded in stereo, you need to make sure the import settings match this. If you didn’t record in stereo (Chapter 14), you don’t want to import your two channels of audio as a stereo clip. All nonsync audio sources, music, and sound effects must also be ingested with care. If you bring in a music track, which will often have a sample rate of 44.1 kHz, the NLE may handle it, and it may sound okay in the sequence, but cause trouble when you go to a mix. In addition, sound effects downloaded from the Internet may be in a compressed file format such as MP3, which you should also plan to convert before importing them. In fact, MP3 files are generally problematic for use in a documentary soundtrack, so try to avoid them where possible. The best formats to use are the two uncompressed digital formats, .wav and .aiff.

Syncing Picture and Sound

If you have shot single-system, then the audio will be captured in sync along with the picture. However, if you have shot double-system, then your audio was captured separately and stored as .wav or .aiff files. Now, each scene must be synced in your NLE system. There are different ways of doing this:

1. Using a Slate to Sync

   Open the image clip in the viewer and find the exact frame where the clapper arm on the slate closes. Place an in-point on that frame.

   Open the corresponding sound clip. You will see a sound wave representing the audio on that clip. Find the exact point where the “clap” occurs (looking at a waveform of the sound clip will help tremendously) and mark an in-point. Audio scrub, which allows you to hear frame-by-frame audio as you drag the viewer playhead, is also helpful.

   Select the video and audio tracks and, using the autosync (Avid) or merge clips (PP) command, create a single synced clip. Be sure you select the dialogue box that tells the program to use the in-point as a sync-point.

2. Syncing When Your Picture Files Include a Single-System Scratch Track

   Often, when you are shooting double-system, your camera will have recorded audio that is not good enough to use in your soundtrack, but which can be useful for syncing. (This is often the case with DSLR shooting, for example). If you do have a scratch track, just expand the vertical height of the track to make the waveform more visible and line up the waveform of the double-system track with the waveform of the single-system scratch track. Play the track. If it sounds “reverb-y,” it’s not quite in sync. Once the clip plays back without any echo or reverb, it’s in sync and you can merge the clips.

3. Using Time Code to Sync

   If you are fortunate enough to have shot with a camera and sound recording device that had locked time code (p. 116), you will be able to use the time code on the clips to sync your audio and video.

4. Using Pluraleyes to Sync

   Pluraleyes is a software program by Singular Software that synchronizes audio and video clips automatically without the need for time code or slates. It integrates with Premiere Pro and AVID Media Composer.

The NLE Software Interface

In standard editing mode, most NLE systems divide the edit environment (or desktop) into four main windows (Figure 17.9): the project window (Avid) or project panel (PP), (A); the preview monitor (Avid) or source panel (PP), (B); the timeline, (C) ; and the sequence monitor (Avid) or program panel (PP), (D). The names may differ, but their function is basically the same regardless of the particular NLE program you are using.

The Timeline Window

Once you’ve decided on the specific parameters of the shot you’d like to insert into your film, you place it into the timeline. The timeline is where you truly edit your movie by inserting, deleting, arranging, rearranging, and fine-tuning your clips as you build your documentary one cut at a time.

The timeline is divided into discrete video and audio tracks for maximum creative flexibility. A typical timeline will automatically provide one video track and two audio tracks when you start a project, but you can add as many tracks as you need. You have the option to cut, rearrange, or apply effects to only the picture, or only the sound, or any combination of picture and sound tracks. Audio tracks can be layered to create complex sound design. Multiple video tracks can be used to create superimpositions and other layering effects (Figure 17.13).

■ Figure 17.9 The four main editing windows are essentially the same in Avid (top) and Premiere Pro (bottom): the project window (Avid) or project panel (PP), (A) ; the preview monitor (Avid) or source panel (PP), (B); the timeline, (C); and the sequence monitor (Avid) or program panel (PP), (D).

■ Figure 17.10 The organization of project files in Avid Media Composer. The Project Window contains folders (A) for project organization, and bins (B) which hold sequences (C), clips (D), subclips (E), audio files (F), and other elements.

■ Figure 17.11 Bins have many columns that contain useful information for organizing and finding clips. This example is from an Avid project.

As you lay down clips in the timeline, you are creating a sequence, which is a graphical representation of your edited movie. Sequences are saved in the project window or panel, along with your clips. Sequences should be clearly named and saved frequently. One of the great advantages of digital editing is that you can create multiple sequences, copy sequences, create versions of sequences, and even insert part of a sequence into another sequence.

■ Figure 17.12 The preview monitor in Avid is the left half of the composer window (shown in its entirety here). This is where you view your source clips, and set in-points and out-points.

■ Figure 17.13 The timeline consists of the picture track(s) (A), sound tracks (B), and playhead (C).

Inside the timeline is a playhead, which is a horizontally scrolling vertical line running through all edited tracks. The playhead tells you where you are in the timeline and is used to move through your sequence quickly. You can also use the playhead to determine where edit points are placed and where shots are inserted. If you hit play, the playhead moves across the sequence in real time.

In all NLE programs, the number of tools and the possibilities for working in the timeline are staggering. Some capabilities, like the ability to trim shots shorter or make them longer with frame accurate precision, are essential tools for editing. Other timeline functions you will use only rarely. It is not possible to cover the functions in detail in this book, so we refer you to the software instruction manuals. All NLE systems have a Getting Started manual, which is the best place to begin. Also, there are many online tutorials, user forums, and books readily available.

■ Figure 17.14 Avid’s Tool Palette (top) allows you to access commonly used tools for editing, and the command palette (center) allows you to map functions onto your keyboard (bottom) for faster editing.

Menu, Icon, or Keyboard: Take Your Pick

One thing you’ll discover on all NLE systems is that there are usually three ways of doing exactly the same thing. You can find any given command inside pull-down menus, or you can trigger the same command by clicking an icon on the desktop or using a keyboard shortcut. For example, in Premiere Pro, setting in- and out-points on a source clip can be accomplished three ways: by hitting the “i” or “o” button on the keyboard, by clicking on the mark in/mark out icons in the source panel, or by scrolling down the “mark” menu to the “mark in” or “mark out” command (Figure 17.15). You don’t need to memorize all three ways of doing every task. Professional editors tend to use more keyboard commands as they are the quickest, but most people use a combination. Customize your process, and find the easiest and most comfortable route for yourself.

■ Figure 17.15 Many functions can be accomplished in multiple ways. For example, you can set in- and out-points on the keyboard (left), through a menu (A), or by clicking icons (B).

There are two special keyboard functions that are extremely handy to know:

1. The J, K, and L buttons are universal transport control buttons. J plays in reverse at normal speed, K is pause, and L is play forward at normal speed. Pressing J and L multiple times increases or decreases the play speed from normal speed (one press) to 2x to 4x to 8x (four presses). Holding the K and pressing J or L will give you slow motion.
2. The command-Z keystroke combination instantly undoes your last action. Inevitably, as you edit, you will click a wrong icon or drag and drop something where it doesn’t belong, or even accidentally delete an entire sequence. If you’ve made a mistake, simply hit command-Z, and all is put back where it was before your blunder.

Split Edits

In the exercise above, we relied exclusively on straight cuts. These edits cut the picture and sound on the same frame, at the beginning (also called the head) and end (the tail) of each shot. In many instances, documentary editors combine sounds and images in more complex ways. Many of these fall into the category of split edits. In an L-cut, the picture cuts to the next shot first, and the audio cut occurs later. An example would be an interview that starts in sync, but as the person continues to talk we see images unrelated to their location. These cuts are useful when you are condensing dialogue, as you can use the incoming images to conceal cuts in the interview and avoid a visual jump cut. The J-cut is the opposite: the audio cuts out first, while the picture plays a bit longer before cutting to the next shot. The timing of these edits depends on a number of factors, some creative and some practical. A split edit can seem more “natural” than a hard cut, perhaps because it mimics the way we often perceive real events, hearing something first, and then turning our heads to look. It is a useful exercise to take a sample documentary scene and take it apart conceptually so that you can understand how split edits contribute to the whole. There is an example of a documentary scene that uses split edits on pp. 308–310.

Reviewing Raw Footage

Everything you shoot in production constitutes your raw footage. While you will likely review your material as you shoot it, your first task when you begin postproduction is to sit down and watch it all again. As you watch your footage, you are evaluating each and every scene as though you are seeing it for the first time. As you view your material, take detailed notes on which shots you think you might use. These are called selects (shots you won’t use are called outtakes). You need to develop your own shorthand for note-taking. There are various tools available in your NLE, including color coding and markers, or as previously explained you can use columns in the project window to make notes. Often documentary cinematographers shoot long continuous takes. Although your material comes as clips, these may or may not need to be further broken down. Some editors like to break these down into subclips, excerpted sections of the original long take that act as independent clips. Some editors create “selects” sequences by shoot day or topic. These are good to look at for a quick reminder of what you thought was good material when you first reviewed it.

As you review and evaluate your footage, you should also be keeping notes with any ideas for structure, themes, or relationships relevant to your project, or even ideas about how shots or scenes may work together. These notes are rough and spontaneous, but they help you make connections and identify the storytelling potential of your footage. Even if you have transcripts of your interviews, you should be watching that footage with the same fresh set of eyes. Sometimes sound bites seem great on paper, but the way they are delivered is less compelling.

Another thing worth mentioning in an era when many people edit their own films is that, as an editor, you are wearing a different hat than you did as a director. You need to be detached, even ruthless. Ask yourself, “What is significant about what I have here?” and “What story can this material tell?” The film that can be made at this point is very often different from the one you had in your mind while shooting. Try and see this as an opportunity instead of as a disaster! Every editor, on every project, confronts this reality.

Paper Edit

Once you have a sense of what material you have and what story it might tell, you can start to conceptualize a structure. A paper edit is one tool that allows you to organize your documentary on paper before you start actually editing. It can save you much time and effort. Here is one way to approach creating a paper edit.

Consider the bare-bones structure of your story. What is a possible beginning, middle, and end? Consult the dramatic arc in Chapter 3, and think about possible themes and takeaways from your film for the audience. Then, look through your observational material and find the scenes or events could become potential building blocks. A scene is something that unfolds in one place at one time. If you shoot at a location for half a day, there may be a dozen incidents that could be the basis of a scene, but which one you choose depends on the story you are telling. Say you’ve been shooting in a middle school classroom. You might have a few moments that stand out as being interesting: a meltdown by a student, a learning moment with some shared reading where the teacher shines, or a personal interaction at the end of class with a student who is having trouble.

If your film is about the fairness of evaluating teachers by their students’ test scores, the most significant scene might be the one where the teacher shines, because it shows that learning happens in ways that aren’t reflected on tests. Or the meltdown might be useful, as it could show the challenges teachers face in the classroom.

Whichever scene you choose to develop, it’s important to understand that a scene generally has some type of conflict, and ends when that conflict is resolved, even in a small way. So, with the hypothetical meltdown, the scene will end when the student calms down, or when the teacher decides he needs to be brought to a counselor’s office. Also important to consider is how the scene might contribute to the progression of ideas and information in your film.

The reason we start with observational scenes, if we have them, is that interviews are almost always better used when married to a concrete unfolding situation than as stand-alone elements. In addition, the vast majority of student documentaries suffer from an overabundance of talking heads and a shortage of other types of material, so it makes sense to focus on finding your best observational material first.

We suggest putting each scene you have identified on an index card. Include the content, the characters involved, and what you expect the scene to deliver in terms of an “event” or point delivered. Arrange these cards into some kind of order with an eye to the dramatic arc of your film. You should be able to identify a beginning, a climax, and a conclusion. Now add interview bites, ideas for montage sequences, text cards, animation, or voice-over on additional cards and you have a paper edit!

Your paper edit can be taped on a wall, laid out on the floor, or put into a notebook. It is a roadmap for your editing.

Rough Cuts

Now that you have a solid idea of how you want to structure your documentary, you can start to build a rough cut, following the structure you developed on paper. Start with the opening of the film and work from there. As you build the film, your main characters and events will come to life. You will get a clear idea of which interviews work really well and which don’t. It is guaranteed that some ideas that seemed great on paper will not work at all. There may be glaringly obvious ways to rearrange the material to improve your storyline and thematic threads. All this is good and part of the process. Paper edits are written on paper, not in stone. As you put the film together, take advantage of the evolving connections and chemistry that happen when you actually put your material together, and feel free to reorganize things along the way. You can expect this rough cut to be longer than your final film, but with practice it will give you a fairly clear idea of what that final length might be.

At this point, you should use the basic tools available in the edit system, like straight cuts instead of dissolves. Avoid complex split edits (p. 288). It is important not to over-polish the film at this stage. It may be irritating to view material that isn’t perfectly finessed, but you want to feel free to be able to move material around, and complex transitions between shots will only get in the way. There is another danger to overcutting: sometimes the magic of a fine-tuned cut will make you think something is working on a structural level when, in fact, it is not.

If you are planning to use narration, here is where you begin writing it. It is a good idea to start actually inserting scratch narration, whether recorded straight into the edit system or from a separate recorder. You will record final narration as part of your sound mix, so don’t worry about the quality or performance at this point. Graphic elements like titles and animated sequences should also be scratch at this early stage. You can insert title cards as placeholders to indicate any footage or scenes that remain to be filmed.

You will edit several rough cuts of your film, reorganizing things to improve the flow of the story with each version. Rough cuts are where you spend the majority of your time and do most of your creative thinking. There is a lot of cutting and recutting during this stage as you discover the final style, shape, and rhythm of your film.

When you are in rough cut stage, it is a good idea to transcribe your cut periodically. This makes it easier for you to talk through your structure with your editor or other people who give you feedback.

Rough Cut Screenings

It is hard to maintain a fresh eye when you spend months editing your documentary. For this reason, it is important to have rough cut screenings for sample audiences. These can be trusted friends, larger groups of people who are similar to your intended audience, or other filmmakers who you believe may have constructive criticism or helpful contributions to make. It can also be useful to screen rough cuts for experts on your subject matter, to make sure you aren’t misrepresenting things or framing issues in a way you ultimately won’t be comfortable with. If you expect your film to be used widely by a certain group of people like, for example, doctors, then pulling together a screening for a group of doctors is an excellent idea.

Rough cut screenings can be harrowing for directors and editors. Try to listen, take notes, and refrain from arguing with the audience. Remember, not everybody’s comments will be useful or on target. More often than not, people suggest solutions that are incorrect, but are pointing to a real problem that you may solve by other means. Take some time to digest and evaluate all of the feedback you received from the screening before running back to the cutting room and pulling everything apart. You do not need to use all of the suggestions, but you should use whatever can make your film stronger. And remember to thank people for their important contribution to your work.

Building Your Soundtracks

In the rough cut stage, working with sound is just as important as working with picture. As you develop your cut, you should plan to keep your sequence audio split out into tracks that reflect the main sound sources you’ve recorded. As an example, you could create four tracks for interviews, four tracks for ambience and audio from observational scenes, two tracks for music, two tracks for narration, and so on. Any scene will have its original sound, often on more than one track. Keep as much production audio as you have in the timeline, even if that means muting it temporarily so you can hear other sounds. If you are working with a composer, they may give you sketches of music you can lay in to figure out what they should develop further. More commonly, you will place temp music pulled from any source (regardless of whether you have the right to use it), just to get a feel for the style and tone that will work for the scene. Later, the composer will create something that works in a similar fashion, or you can acquire other music that you have rights to use (Chapter 21).

For screening rough cuts, you will have to create a scratch mix. This means getting your audio levels into a comfortable range without spending too much time developing a full mix.

The Fine Cut and Picture Lock

Once you’re happy with the structure of your film, and all of the sequences are working the way you want them to, and you’ve determined that there will be no more big changes to the film, you can start to fine-tune the rough cut. The fine cut involves finessing all of the edits one by one. This is the time to make those small adjustments to, for instance, get that cut on action just right, add the dissolve between two scenes, or trim a few frames off a shot to get the timing just perfect. The fine cut is also where you replace temporary graphics, archival footage, and stills with high-resolution versions that you have cleared the rights to. At this stage, make sure you have a complete and accurate list of lower thirds (IDs) and end credits (Chapter 22).

This stage is also where you add additional sound elements like sound effects, extra ambiences, final music, and final narration. Once the placement of all your elements is perfect, all of the editing decisions have been made, and you’ve decided you will not trim a single frame more, you have arrived at picture lock.

Finishing

Picture lock does not mean all of the creative work is over. The film still needs finishing, which means that you turn your attention to three areas: sound design, graphics, and color correction. Depending on your workflow, you may handle these within your NLE system, as in Figure 17.1. Alternatively, you may plan to export your sound and picture for mixing and color correction with outside professionals (Figure 17.2). Professional sound design is covered in Chapter 21, and color correction and graphics are covered in Chapter 22.

Mastering

The final step in the editing process is to master the film. This means outputting the final film to a sturdy, archivally sound digital form from which distribution copies can be made. Although most NLE systems offer a wide range of output options, you will, no doubt, want to master to a format that does not involve any further video data compression. These days some people continue to master on digital HD tape, while others retain files on a hard drive that is backed up. And, of course, many do both for safety.