CHAPTER 10

THE ART OF EDITING IS CRAFTING A STORY

There are as many ways to photograph an object as there are ways to edit the same footage. We all know that 10 photographers looking at the same landscape will capture 10 different interpretations of it. Give those 10 images to 10 different retouchers and the possibilities are endless.

The same thing happens with video editing. The craft of editing is about telling the story in the most efficient way possible. It is finding which combinations, out of millions available, will help you emphasize emotions, build tension, fix technical mistakes, and hide or reveal information to the viewer.

The iconic German director Werner Herzog stated, “Just learning a few technical tricks doesn’t make someone a filmmaker any more than knowing how to type makes someone a writer.” I’d argue that this is especially true when it comes to video editing. Did you know that since 1981 not a single film has won an Oscar for Best Picture without at least being nominated for Film Editing? And two-thirds of the movies nominated for Film Editing have won Best Picture.

Editing styles can be extremely different—from the very seamless and invisible to the self-aware and intentional, where the edits themselves almost become a character of the story. The genre of every piece and even the emotional charge of a specific scene will often dictate the rhythm. For example, a romantic scene with very a long take or just a few cuts might work better than many fast cuts. For an action scene, the opposite is true.

Editing is arguably the topic that prevents most photographers from really jumping into filmmaking. And with good reason—this is by far the most time-consuming aspect of our productions.

Here’s my quick and simple rule of thumb to estimate editing times:

• 1 hour of footage from one camera without sound = 3 hours of editing
• 1 hour of footage from one camera with external sound = 6 hours of editing
• 1 hour of footage from two cameras without sound = 8 hours of editing
• 1 hour of footage from two cameras with external sound = 16 hours of editing

Of course, there are many more variables to consider, such as the complexity of the edit and the experience of the editor, but these are good ballpark numbers to quickly estimate times and budgets.

Instead of telling you how to select a tool or apply a technique in a software application that most likely will be obsolete by the time you read this line, I’d like to share what I believe are the fundamental concepts in the art of editing. I’ll do my best to explain some terms that sounded like a foreign language to me when I was getting started.

My goal is that after reading this chapter you approach this exciting topic with an open and positive attitude. Ready? Here we go.

CREATING A POST-PRODUCTION WORKFLOW

As photographers we come back from a shoot, launch an app like Adobe Lightroom, and start editing even as the images are being imported. As long we copy the images from our memory card onto an internal or external hard drive and Lightroom sees them, we are good to go.

I wish the same could be said for video. Not taking the time to establish a clear post-production workflow usually ends up consuming a lot more time and eventually costing more money. The difference is that not only are the files much larger in size, but you also need to consider other assets like location audio from different devices, music, graphics, LUTs and Looks, and stills. These might seem like a small or obvious decisions at first, but where and how you organize your assets can have a significant impact on the overall post-production process.

DECONSTRUCTING THE EDITING PROCESS

Post-production workflows vary widely depending on the director, DP, editor, project, budget, genre, and even the cameras used, but the most fundamental steps are as follows:

1. Download the footage.
2. Ingest – Import footage into your video-editing software.
3. View the raw footage.
4. Assembly Cut – Sequence the clips on a timeline in a logical order to see how the story will flow.
5. Rough Cut – Refine your selection of clips. At this stage they will be good enough to be shared with others and get feedback, but they will still be missing other assets, final graphics, music, and color corrections.
6. Picture Lock – Lock the clips in place. Graphics, music, and sound effects have been added at this point.
7. Add titles, music, sound effects, and special effects.
8. Color-correct to match clips.
9. Grade to enhance the footage.
10. Mix everything.
11. Final Cut – Create the finished graded and mixed version.
12. Export and upload your video to an FTP or distribution platform like YouTube or Vimeo.

ORGANIZING YOUR ASSETS

When it comes to organizing the assets before the edit, I tend to follow a very simple rule: organize the project in a way that an editor working without instructions, and without me sitting next to them, could quickly understand my system and find everything they need.

Figure 10.1.1 Adobe Creative Cloud offers all of the tools you need to get started in video editing, color grading, and audio editing.

After downloading all the media to a safe place and importing it into your editing software, a good way to organize your footage and additional assets is by using bins (virtual folders within the editing software) and labels to categorize and subcategorize them. The process varies slightly in each editing system, but the basic premise remains the same.

Here’s an example: I always create a bin for Original Footage, another bin for Stills, and a few more for Music, Graphics, Timelapses, and so on. The names of each bin should be self-explanatory. Then inside my Original Footage bin I create another bin for each day of shooting. Inside each day I create one bin for Camera A, another bin for Camera B, and a third bin for Sound, for example. Some people choose to organize the material by scene and takes. It all depends on your personal preferences and project.

There is nothing worse than putting together a sequence, having the revelation of “I know the perfect shot for this!,” and then wasting a lot of time trying to find that clip. Scrolling through clips looking for a specific moment simply because we didn’t label our footage properly does not pay off.

Figure 10.1.2 This is the typical folder structure on my hard drive. I have one master folder with subfolders. At the beginning of each project I simply copy and rename the master folder structure.

Figure 10.1.3 I also use a master folder (bin) structure in Adobe Premiere Pro. I created all of the bins I could possibly need and saved that empty project as a master project. As soon as I’m ready to start editing, I simply copy and rename the master project, which saves me valuable time.

In most editing systems, you can change the name of a clip without altering the original source file. This feature can be a double-edged sword. I suggest creating a naming system that works for you early on, making sure everybody involved in the edit understands the system, and sticking with that system throughout the project.

After bins and filenaming, a third organizational technique is using color labels. Color coding helps you to quickly identify and associate assets and maintain an orderly timeline. For example, I might use blue for footage, green for stills, white for graphics, and yellow for voiceover. Just like with bins, some people customize their labels to reflect different scenes or different characters in a script.

A fourth organizational tool is the Search Bins feature in Adobe Premier Pro, a feature I use often. Search Bins are very similar to Smart Galleries in Adobe Lightroom—they’re essentially virtual folders with predefined search criteria.

The Search Bins update automatically when you are importing assets, deleting assets, or modifying asset metadata. Just like keywords in Adobe Lightroom, if the metadata for a clip is changed or a clip is deleted, the item is automatically removed from the Search Bin.

I typically use Search Bins with criteria like “good takes” and “best takes,” or even “wide shot” and “closeups.” Renaming a Search Bin is very easy, and the new name will not affect your previous search criteria. Similarly, deleting a Search Bin only deletes the aliases of the items inside and not the original files.

SYNCING CLIPS AND SOUND

Syncing the audio from your camera with other audio devices or syncing up multiple cameras without the use of a clapboard was a very arduous manual process until very recently. Thankfully, today we have software applications that take care of that in minutes, and most nonlinear editing systems (NLEs) also do a fairly decent job.

Figure 10.2 PluralEyes from Red Giant is one platform I use to sync media.

Depending on the platform, you can sync media based on in and out points, timecode, clip markers, and even sound. Which one works best depends entirely on your project, software, and workflow.

If I’m using more than one camera, which is almost always the case, I like to create “multicam” clips to quickly toggle between different camera angles rather than editing the source material individually.

I hope these simple tricks will help you keep your assets organized and save you valuable time.

BUILDING A SEQUENCE

Every video editor I’ve worked with has a slightly different process when it comes to building a sequence. Some start by putting together a rough version of the story, and then invest more time fine-tuning it. In my opinion, this approach makes a lot of sense for most occasions because you can see how the story flows and where the gaps might be before spending a lot of time adding and enhancing technical aspects like color and sound. This approach also makes sense if you have a specific duration target—for example, with music videos or news. If you know your video must be 90 seconds long, you cut for that, and then start fine-tuning.

Other editors prefer to begin working with a specific scene, often a key moment in the story, and fine-tune it until it is polished enough to be used as the reference for the rest of the project. In other words, all subsequent scenes should match the look and feel of that original master scene.

Obviously, there are as many editing approaches as there are shooting styles. Lately, if I am editing my own work, my first step is to watch all the footage. If I know there’s a very good take I start with that one; otherwise, I watch everything and start assembling a rough cut as soon and as quickly as possible while the images are still fresh in my mind.

SELECTING THE BEST TAKES

Regardless of the approach you use to build your sequences, you are inevitably tasked with a serious mission: deciding which takes are best. It sounds simple, but if as a photographer you have worked on architectural or fashion projects, you already know that the differences between one image and the next can be negligible. The same can happen between one take and another.

A different challenge with video is that you usually have to watch a complete clip to decide whether it’s working. So, let’s assume you are trying to select the best take out of three. The scene just has two people having a conversation, and let’s also assume that each take is 3 minutes long. You will be watching 9 minutes of footage before you can even begin to fine-tune the best take. In reality, what is most likely to happen is that you don’t have a single best take, but there are good parts in one take and good parts in another take. This is why editing can take so long!

Every now and then we know we have our “money shot,” where everything worked out great. Whether the actors nailed their lines, the camera’s movement was nice and smooth, or the lighting was just right, it is clearly the best we are going to get. With interviews and educational content, that money shot is usually, but definitely not always, the last one we shot. Because of this I review the last take first. If it has everything I need, I label it as “good take” and move on to the next one.

The easiest way to eliminate takes is by looking for obvious technical issues like unwanted camera shake, noises in the background, a boom peeking into the frame, or an actor missing his mark.

The next step is to look for continuity issues—for example, that the actor’s eyeline is consistent from one cut to another or that the props are in the same position. Most people are very forgiving with minor continuity differences, but if they are too obvious or abrupt they can pull the viewer out of the world we are working so hard to create.

On most shoots things go well, especially if you consistently work with the same crew and develop a good workflow. That’s great, but it means that instead of selecting your best takes based solely on continuity or technical considerations, you will have to gauge highly subjective things like the quality of an actor’s performance.

Every film is different and every performance is unique from genre to genre. Over time, you will develop a sense of when to pick a delivery that might not be technically perfect but is more believable over one that does not have any obvious issues but might not engage the viewer or preserve the authenticity of the piece.

Lastly, if you have a choice of different camera angles or different compositions for the same or similar takes, try to use a little of all of them. This usually adds an inherent freshness to the edit and provides the viewer with additional visual information. Whether it is using a closeup to hone in on the reactions of a character or using a wider shot to provide the context of a scene, if you have it, use it.

Figure 10.3 Examples of time-lines for two very different projects

ADDITIONAL SHOOTING, AKA RESHOOTING

After finishing your rough cut, it becomes clear if you are missing content that would help you to tell the story successfully. This is when reshooting, or as I prefer to call it, additional shooting, might be required. To clarify, missing content doesn’t necessarily mean that the director or DP screwed up. More likely, when assembling the rough cut the editor or director might notice that a scene could use an extra shot to be more effective, or that by adding a few lines of dialogue the story could be stronger.

Interestingly, on large-budget productions reshooting is an expected step of the process and is often budgeted accordingly. In my world, additional shooting is a real luxury. Most of the time we are calling in a lot of favors: from securing the location and borrowing props, to assembling the cast and crew. Typically, reshooting a scene is financially or logistically impossible for the projects we work on.

If reshooting is completely out of the question, the editor has to come up with creative solutions, like crafting a montage to conceal a missing transition or altering the structure of a dialogue scene to cut out some subpar acting. If we need additional content to help establish the context or environment of a scene, stock footage might be a viable option.

BACKING UP

I trust that as a photographer you have your backup strategies down to a science. Most likely, the software you are currently using to back up your photos will work to back up your video assets.

A slight but important difference is your hard drive’s speed, because it will significantly impact your post-production workflow. Even if the other pieces of the puzzle like video cards and RAM are adequate, slow hard drives become a major bottleneck for real-time performance.

Solid-state drives (SSDs) store digital data just like hard disk drives (HDDs), but they are very different in terms of construction and performance. HDDs are electromechanical devices with spinning disks and movable read and write heads. The device needs to wait for these moving parts to get in place before they can read or write any information. This makes them slower and more likely to fail.

Figure 10.4.1 My standard setup in my home office includes multiple external hard drives. Each is labeled with a unique name and contains different media. For example, “Piazzolla” contains only music and sound effects while “Deakins” contains only video files.

SSDs contain no moving parts, which means they’re faster, more reliable, and potentially more durable than HDDs. SSDs do not need to be defragmented, and they stand a better chance of surviving a drop or impact. Strong magnetic fields, like speakers or even cell phones, can corrupt the data on an HDD, whereas SSDs are unaffected by magnets. The advantages of reliability and durability are obvious.

SSDs can generally read and write twice as fast as mechanical HDDs, which becomes a very clear advantage when booting a computer, downloading large video files, opening heavy software applications (like Adobe Premiere Pro, Adobe After Effects, and Adobe Photoshop), or working on highly demanding tasks like video editing and grading.

Another great advantage of SSDs is that they consume less power and produce less heat, which greatly expands the life and battery performance of laptops and other portable devices like external monitors and recorders.

Remember, all drives, regardless of their price, brand, interface, or age, will eventually fail and die. It is not a matter of if but when, and a backup strategy should be in place regardless of which device you choose for your workflow.

If you are interested in learning more about my personal backup strategy (Figure 10.4.1), which I’ve fine-tuned over 15 years or so, read this article: http://bit.ly/HDdeadNowWhat.

BACKING UP ON LOCATION

We all know that creating new habits isn’t easy. If you are used to shooting an assignment and waiting to get home or to your studio to back up, I have news for you: you need to develop a new habit of backing up on location (Figure 10.4.2). There are many things that can go wrong: a LUT is being applied (or not applied) to your footage; the sound is not being recorded to the chosen destination; one mic is working but the second one isn’t. Backing up on location as often as possible can eventually save a project or even a client.

Figure 10.4.2 This is my typical DIT station setup on location. It includes a laptop, at least two external hard drives, fast card readers, and plenty of USB hubs and battery chargers.

Before a shoot I label each and every card, not only with a label (which could fall off), but also by giving each card a unique name (Figure 10.4.4).

Since we almost always use more than one camera, we name a 64 GB card A1 for Camera A, a second card with the same exact capacity B1 for Camera B, and a third card C1 for Camera C. The same goes for A2, B2, C2, and so on.

This is the key during production: when we replace one memory card, say C1, we also replace B1 and A1, even if there’s space left on those two cards.

This step might seem wasteful, and it is, but it will make the process of downloading the footage later much smoother. After a 12-hour day of shooting with three cameras, one external recorder, and one audio recorder, it is extremely easy to confuse what has been downloaded and what hasn’t.

Figure 10.4.3 I’ve been using ChronoSync for many years to back up multiple hard drives.

Figure 10.4.4 To prepare for a shoot I lay out my gear and label it clearly.

In the film industry, ShotPut Pro seems to be the standard software application to download footage (Figures 10.4.5–10.4.8). I use it and like it, but you can continue using whichever application you currently have.

I name my folders following the year/month/day format followed with the time in the 24-hour format. For example, a folder created on September 25 of 2017 at 9:10 p.m. would read: 20170925-2110.

With ShotPut Pro you can choose when to start the data transfer. The options are Manual and Automatic. Automatic means that as soon as the application detects a hard drive, SSD, or memory card it will start the downloading process. Manual simply means you need to start the process manually. ShotPut Pro provides a fairly accurate download time, which helps you determine if you should grab something to eat or wait a little longer.

Figure 10.4.5 Another great application is ShotPut Pro. The first step after launching the app is to drop the files, folders, or devices you want to back up into the app.

Figure 10.4.6 The next step is to define your naming convention and backup destination.

Figure 10.4.7 Click on Begin to start downloading or backing up.

Figure 10.4.8 Upon completion, you will get a message confirming that everything went well. If the process was not successful, you will get a log with possible issues.

ShotPut Pro can back up to seemingly endless destinations, so you can select two or three hard drives and instantly have multiple copies with a single click. ShotPut Pro can also format your hard drive, SSD, or memory card as soon as the download is finished and verified, but I don’t feel comfortable formatting the original media right away. When possible, I prefer to wait until we have triple checked and confirmed that everything is properly backed up.

A minor weakness of ShotPut Pro is that if you try to back up the same card again, even to the same destination, it will let you (unlike Adobe Lightroom’s “prevent importing suspected duplicates” feature). My simple workaround is put away the downloaded cards, grab fresh ones, and keep shooting.

BACKING UP IN THE STUDIO

A “RAID system” may sound like a military term, but it can be your favorite “geek” term. If we oversimplify, a RAID (Redundant Array of Inexpensive Disks) is a system of two or more hard drives that are connected to improve performance, reliability, and protection. Because you are using more than one drive to store your data, you greatly decrease the chances of losing such data. RAID systems have been growing in size and features, while coming down in price, and provide an excellent backup solution for photographers and filmmakers.

Most current portable and desktop computers can handle Full HD and even 4K video editing. However, when you need to mix audio, stills, footage, and motion graphic assets while multitasking with software applications like Adobe Premiere Pro, Apple Final Cut Pro, Adobe After Effects, and Adobe Photoshop, hard disk bandwidth requirements increase so dramatically that laptop-based, multistream, real-time editing is virtually impossible. Add to this high-end codecs like Apple’s ProRes 422 codec and Apple’s Intermediate Codec (AIC), and the needs for very fast, very large, and expandable storage solutions are painfully clear.

The great news is that most current RAID systems offer capacities from 2 to 8 TB, and are compatible with Apple’s iMac, MacBook Pro, and Mac Pro, as well as Linux and Windows workstations. Your ideal capacity obviously depends on many factors like capture format, resolution, and budget.

Higher-end RAID systems offer additional features on top of increased speed and reliability. For example, they can constantly monitor the hard drives’ temperature and health, offer visual and audible alarm signals in case of a drive failure, provide noiseless fans, and include a secondary power supply that enables the system to continue working in the event of a power-supply failure. Some systems allow users to monitor and manage the RAID storage remotely, and even send error notifications via email.

My typical video workflow for Full HD and 4K involves at least three devices. The computer’s operating system (OS) and any software applications we need live on one hard drive. I suggest using an SSD for this because applications will launch much faster. All our original assets live on an external RAID configured as RAID 0 for faster performance. And all our video projects are exported onto a third hard drive, typically a portable and relatively inexpensive USB 3.1 or Thunderbolt drive.

An overview of my recommended system looks like this:

• Hard Drive 1 = Solid State Drive = Operating System and Applications
• Hard Drive 2 = RAID = Master Video and Audio Files
• Hard Drive 3 = Portable Hard Disk Drive = Exported Final Projects

A RAID system can be configured in several different ways. In the photo and video industries, we generally use RAID levels 0 (Fastest), 1 (Mirror), 5 (Safe), and 6 (Safest). Each configuration offers different advantages and disadvantages.

RAID 0 = FASTEST: A RAID 0 configuration provides maximum performance and speed by accessing multiple drives simultaneously. Keep in mind that this configuration does not provide any data redundancy, which is very high risk, but offers the fastest speed of all the levels. It achieves this speed by breaking up the data into smaller blocks and then writing a block to each drive in the array.

The probability of a RAID 0 failure increases in direct proportion to the number of drives, which means that your chances of losing data with an eight-drive RAID 0 array are twice as high as with a four-drive RAID 0 array.

RAID 1 = MIRROR: RAID 1, also known as disk mirroring, greatly increases reliability over RAID 0.

Because you are mirroring the data on other drives, you cut your chances of data loss in half. When one drive fails, the data is preserved on the other drive.

The disadvantage is that storage capacity is also cut in half, and since all the data has to be written twice, the write speed is much slower.

RAID 5 = SAFETY: If you are currently using a RAID system for your photography, you are most likely using a RAID 5 configuration. I like this configuration because I can play a video and pull a drive from the system, and the video continues playing without any issues or data corruption. This magic happens simply because the data is written to all of the drives instead of being concentrated on a single dedicated disk.

When one drive in the system fails, the parity information can be used to reconstruct the data from the failed drive. All drives in the array can be used at the same time, which greatly increases performance, but the speed will never be the same as with RAID 0. Also with RAID 5, if two drives fail simultaneously you are in big trouble.

RAID 6 = SAFEST: RAID 6 provides an extremely high fault tolerance, and can sustain two simultaneous drive failures without downtime or data loss.

Parity data requires disk space, so you might see a significant reduction of available capacity. For example, on RAID 1 the amount of available space is roughly 50 percent of the total capacity. On a RAID 5 or 6, the amount of available space is approximately 75 percent of the total capacity.

Figure 10.4.9 I don’t have a “typical” setup for working in recording studios because every place has vastly different gear and configurations.

THE IMPORTANCE OF TEASERS AND TRAILERS

We have all seen movie trailers, which seem to be getting longer and louder over time. We have also seen teasers, but not until you get into filmmaking will you notice the differences between the two.

Traditionally, a movie trailer is between two and three minutes in length, and it shows the movie’s main plot and the main characters, and ultimately delivers a sense of the kind of story you can expect to watch.

A teaser is usually much shorter, typically under one minute in length and edited specifically to generate interest in the upcoming movie. Teasers don’t necessarily provide a complete sense of the story, and they often show the most impressive shots, including some that don’t make it into the final cut.

Most people, including me, find the post-production phase tedious and a bit frustrating. To me, the longer it takes to edit a project, the less engaged I become. Working on something I shot a month ago is not as enticing as looking at the footage I shot earlier today. Including a teaser as part of your deliverables offers several advantages: it forces the editor to go through all the footage and pay attention to the strongest scenes. It is a great tool to share with other collaborators, like the colorist or composer, so they can start thinking about or even working on the project before a complete rough cut version is available. But perhaps the biggest advantage is that by having something with you at all times (on your phone or tablet), it is easy to show what you are working on, watch the reactions from people who are not connected to the project, receive feedback, and start creating buzz about your project. “When can I see the final piece?” is one of the best compliments you can get.