Swiss Army Man

df2716_swissarmymanWhen it comes to quirky movies, Swiss Army Man stands alone. Hank (Paul Dano) is a castaway on a deserted island at his wit’s end. In an act of final desperation, he’s about to hang himself, when he discovers Manny (Daniel Radcliffe), a corpse that’s just washed up on shore. At this point the film diverges from the typical castaway/survival story into an absurdist comedy. Manny can talk and has “magical powers” that Hank uses to find his way back to civilization.

Swiss Army Man was conceived and directed by the writing and directing duo of Dan Kwan and Daniel Sheinert, who work under the moniker Daniels. This is their feature length film debut and was produced with Sundance in mind. The production company brought on Matthew Hannam to edit the film. Hannam (The OA, Enemy, James White) is a Canadian film and TV editor with numerous features and TV series under his belt. I recently spoke with Hannam about the post process on Swiss Army Man.

Hannam discussed the nature of the film. “It’s a very handmade film. We didn’t have a lot of time to edit and had to make quick decisions. I think that really helped us. This was the dozenth or so feature for me, so in a way I was the veteran. It was fun to work with these guys and experience their creative process. Swiss Army Man is a very cinematically-aware film, full of references to other famous films. You’re making a survival movie, but it’s very aware that other survival movies exist. This is also a very self-reflexive film and, in fact, the model is more like a romantic comedy than anything else. So I was a bit disappointed to see a number of the reviews focus solely on the gags in the film, particularly around Manny, the corpse. There’s more to it than that. It’s about a guy who wonders what it might be like had things been different. It’s a very special little film, because the story puts us inside of Hank’s head.”

Unlike the norm for most features, Hannam joined the team after the shooting had been completed. He says, “I came on board during the last few days of filming. They shot for something like 25 days. This was all single-camera work with Larkin Seiple (Cop Car, Bleed For This) as director of photography. They shot ARRI ALEXA XT with Cooke anamorphic lenses. It was shot ARRIRAW, but for the edit we had a special LUT applied to the dailies, so the footage was already beautiful. I got a drive in August and the film premiered at Sundance. That’s a very short post schedule, but our goal was always Sundance.”

Shifting to Adobe tools

Like many of this year’s Sundance films, Adobe Premiere Pro was the editing tool of choice. Hannam continues, “I’m primarily an Avid [Media Composer] editor and the Dans [Kwan and Sheinert] had been using [Apple] Final Cut Pro in the past for the shorts that they’ve edited themselves. They opted to go with Premiere on this film, as they thought it would be easiest to go back and forth with After Effects. We set up a ‘poor man’s’ shared storage with multiple systems that each had duplicate media on local drives. Then we’d use Dropbox to pass around project files and shared elements, like sound effects and temp VFX. While the operation wasn’t flawless – we did experience a few crashes – it got the job done.”

Swiss Army Man features quite a few visual effects shots and Hannam credits the co-directors’ music video background with making this a relatively easy task. He says, “The Dans are used to short turnarounds in their music video projects, so they knew how to integrate visual effects into the production in a way that made it easier for post. That’s also the beauty of working with Premiere Pro. There’s a seamless integration with After Effects. What’s amazing about Premiere is the quality of the built-in effects. You get effects that are actually useful in telling the story. I used the warp stabilizer and timewarp a lot. In some cases those effects made it possible to use shots in a way that was never possible before. The production company partnered with Method for visual effects and Company 3 [Co3] for color grading. However, about half of the effects were done in-house using After Effects. On a few shots, we actually ended up using After Effects’ stabilization after final assembly, because it was that much better than what was possible during the online assembly of the film.”

Another unique aspect of Swiss Army Man is its musical score. Hannam explains, “Due to the tight schedule, music scoring proceeded in parallel with the editing. The initial temp music pulled was quirky, but didn’t really match the nature of the story. Once we got the tone right with the temp tracks, scenes were passed on to the composers – Andy Hull and Robert McDowell – who Daniels met while making a video for their band Manchester Orchestra. The concept for the score was that it was all coming from inside of Hank’s head. Andy sang all the music as if Hank was humming his own score. They created new tracks for us and by the end we had almost no temp music in the edit. Once the edit was finalized, they worked with Paul [Dano] and Daniel [Radcliffe] to sing and record the parts themselves. Fortunately both are great singers, so the final a cappella score is actually the lead actors themselves.”

Structuring the edit

Matthew Hannam and I discussed his approach to editing scenes, especially with this foray into Premiere Pro. He responds, “When I’m on Media Composer, I’m a fan of ScriptSync. It’s a great way to know what coverage you have. There’s nothing like that in Premiere, although I did use the integrated Story app. This enables you to load the script into a tab for quick access. Usually my initial approach is to sit down and watch all the footage for the particular scene while I plan how I’m going to assemble it. The best way to know the footage is to work with it. You have to watch how the shoot progresses in the dailies. Listen to what the director says at the end of a take – or if he interrupts in the middle – and that will give you a good idea of the intention. Then I just start building the scene – often first from the middle. I’m looking for what is the central point of that scene and it often helps to build from the middle out.”

Although Hannam doesn’t use any tricks to organize his footage or create selects, he does use “KEM rolls”. This term stems from the KEM flatbed film editing table. In modern parlance, it means that the editor has strung out all the footage for a scene into a single timeline, making it easy to scrub through all the available footage quickly. He continues, “I’ll build a dailies reel and tuck it away in the bottom of the bin. It’s a great way to quickly see what footage you have available. When it’s time to revise a scene, it’s good to go back to the raw footage and see what options you have. It is a quick way to jog your memory about what was shot.”

A hybrid post workflow

Another integral member of the post team was assistant editor Kyle Gilbertson. He had worked with the co-directors previously and was the architect of the hybrid post workflow followed on this film. Gilbertson pulled all of the shots for VFX that were being handled in-house. Many of the more complicated montages were handled as effects sequences and the edit was rebuilt in DaVinci Resolve before re-assembly in After Effects. Hannam explains, “We had two stages of grading with [colorist] Sofie Borup at Co3. The first was to set looks and get an idea what the material was going to look like once finished. Then, once everything was complete, we combined all of the material for final grading and digital intermediate mastering. There was a real moment of truth when the 100 or so shots that Daniels did themselves were integrated into the final cut. Luckily it all came together fairly seamlessly.”

“Having finished the movie, I look back at it and I’m full of warm feelings. We kind of just dove into it as a big team. The two Dans, Kyle and I were in that room kind of just operating as a single unit. We shifted roles and kept everything very open. I believe the end product reflects that. It’s a film that took inspiration from everywhere and everyone. We were not setting out to be weird or gross. The idea was to break down an audience and make something that everyone could enjoy and be won over by. In the end, it feels like we really took a step forward with what was possible at home. We used the tools we had available to us and we made them work. It makes me excited that Adobe’s Creative Cloud software tools were enough to get a movie into 700 cinemas and win those boys the Sundance Directing prize. We’re at a point in post where you don’t need a lot of hardware. If you can figure out how to do it, you can probably make it yourself. That was our philosophy from start to finish on the movie.”

Originally written for Digital Video magazine / Creative Planet Network.

©2016 Oliver Peters

The On Camera Interview


Many projects are based on first person accounts using the technique of the on camera interview. This approach is used in documentaries, news specials, corporate image presentations, training, commercials, and more. I’ve edited a lot of these, especially for commercials, where a satisfied customer might give a testimonial that gets cut into a five-ish minute piece for the web or a DVD and then various commercial lengths (:10, :15, :30, :60, 2 min.). The production approach and editing techniques are no different in this application than if you are working on a documentary.

The interviewer

The interview is going to be no better than the quality of the interviewer asking the off camera (and unheard) questions. Asking good questions in the right manner will yield successful results. Obviously the interviewer needs to be friendly enough to establish a rapport with the subject. People get nervous on camera, so the interviewer needs to get them relaxed. Then they can comfortably answer the questions and tell the story in their own words. The interviewer should structure the questions in a way that the totality of the responses tells a story. Think in terms of story arc and strive to elicit good beginning and ending statements.

df4115_iview_2Some key points to remember. First, make sure you get the person to rephrase the question as part of their answer, since the audience won’t hear the interviewer. This makes their answer a self-contained statement. Second, let them talk. Don’t interject or jump on the end of the answer, since this will make editing more difficult.

Sometimes in a commercial situation, you have a client or consultant on set, who wants to make sure the interviewee hits all the marketing copy points. Before you get started, you’ll need to have an understanding with the client that the interviewee’s answers will often have to be less than perfect. The interviewees aren’t experienced spokespersons. The more you press them to phrase the answer in the exact way that fits the marketing points or to correctly name a complex product or service in every response, the more stilted their speaking style will become. Remember, you are going for naturalness, honesty and emotion.

The basics

df4115_iview_1As you design the interview set, think of it as staging a portrait. Be mindful of the background, the lighting, and the framing. Depending on the subject matter, you may want a matching background. For example, a doctor’s interview might look best in a lab or in the medical office with complex surgical gear in the background. An interview with an average person is going to look more natural in a neutral environment, like their living room.

You will want to separate the interview subject from the background and this can be achieved through lighting, lens selection, and color scheme. For example, a blonde woman in a peach-colored dress will stand out quite nicely against a darker blue-green background. A lot of folks like the shallow depth-of-field and bokeh effect achieved by a full-frame Canon 5D camera with the right lens. This is a great look, but you can achieve it with most other cameras and lenses, too. In most cases, your video will be seen in the 16:9 HD format, so an off-center framing is desirable. If the person is looking camera left, then they should be on the right side of the frame. Looking camera right, then they should be on the left side.df4115_iview_7

Don’t forget something as basic as the type of chair they are sitting in. You don’t want a chair that rocks, rolls, leans back, or swivels. Some interviews take a long time and subjects that have a tendency to move around in a chair become very distracting – not to mention noisy – in the interview, if that chair moves with them. And of course, make sure the chair itself doesn’t creak.

Camera position

df4115_iview_3The most common interview design you see is where the subject is looking slightly off camera, as they are interacting with the interviewer who sitting to the left or the right of the camera. You do not want to instruct them to look into the camera lens while you are sitting next to the camera, because most people will dart between the interviewer and the camera when they try to attempt this. It’s unnatural.

The one caveat is that if the camera and interviewer are far enough away from the interview subject – and the interviewer is also the camera operator – then it will appear as if the interviewee is actually looking into the lens. That’s because the interviewer and the camera are so close to each other. When the subject addresses the interviewer, he or she appears to be looking at the lens when in fact the interviewee is really just looking at the interviewer.

df4115_iview_16If you want them looking straight into the lens, then one solution is to set up a system whereby the subject can naturally interact with the lens. This is the style documentarian Errol Morris has used in a rig that he dubbed the Interrotron. Essentially it’s a system of two teleprompters. The interviewer and subject can be in the same studio, although separated in distance – or even in other rooms. The two-way mirror of the teleprompter is projecting each person to the other. While looking directly at the interviewer in the teleprompter’s mirror, the interviewee is actually looking directly in the lens. This feels natural, because they are still looking right at the person.

Most producers won’t go to that length, and in fact the emotion of speaking directly to the audience, may or may not be appropriate for your piece. Whether you use Morris’ solution or not, the single camera approach makes it harder to avoid jump cuts. Morris actually embraces and uses these, however, most producers and editors prefer to cover these in some way. Covering the edit with a b-roll shot is a common solution, but another is to “punch in” on the frame, by blowing up the shot digitally by 15-30% at the cut. Now the cut looks like you used a tighter lens. This is where 4K resolution cameras come in handy if you are finishing in 2K or HD.

df4115_iview_6With the advent of lower-cost cameras, like the various DSLR models, it’s quite common to produce these interviews as two camera shoots. Cameras may be positioned to the left or the right of the interviewer, as well as on either side. There really is no right or wrong approach. I’ve done a few where the A-camera is right next to the interviewer, but the B-camera is almost 90-degrees to the side. I’ve even seen it where the B-camera exposes the whole set, including the crew, the other camera, and the lights. This gives the other angle almost a voyeuristic quality. When two cameras are used, each should have a different framing, so a cut between the cameras doesn’t look like a jump cut. The A-camera might have a medium framing including most of the person’s torso and head, while the B-camera’s framing might be a tight close-up of their face.

While it’s nice to have two matched cameras and lens sets, this is not essential. For example, if you end up with two totally mismatched cameras out of necessity – like an Alexa and a GoPro or a C300 and an iPhone – make the best of it. Do something radical with the B-camera to give your piece a mixed media feel. For example, your A-camera could have a nice grade to it, but the B-camera could be black-and-white with pronounced film grain. Sometimes you just have to embrace these differences and call it a style!


df4115_iview_4When you are there to get an interview, be mindful to also get additional b-roll footage for cutaway shots that the editor can use. Tools of the trade, the environment, the interview subject at work, etc. Some interviews are conducted in a manner other than sitting down. For example, a cheesemaker might take you through the storage room and show off different rounds of cheese. Such walking-talking interviews might make up the complete interview or they might be simple pieces used to punctuate a sit-down interview. Remember, that if you have the time, get as much coverage as you can!

Audio and sync

It’s best to use two microphones on all interviews – a lavaliere on the person and a shotgun mic just out of the camera frame. I usually prefer the sound of the shotgun, because it’s more open; but depending on how noisy the environment is, the lav may be the better channel to use. Recording both is good protection. Not all cameras have great sound systems, so you might consider using an external audio recorder. Make sure you patch each mic into separate channels of the camera and/or external recorder, so that they are NOT summed.

df4115_iview_8Wherever you record, make sure all sources receive audio. It would be ideal to feed the same mics to all cameras and recorders, but that’s not always possible. In that case, make sure that each camera is at least using an onboard camera mic. The reason to do this is for sync. The two best ways to establish sync is common timecode and a slate with a clapstick. Ideally both. Absent either of those, then some editing applications (as well as a tool like PluralEyes) can analysis the audio waveform and automatically sync clips based on matching sound. Worst case, the editor can manually sync clips be marking common aural or visual cues.

Depending on the camera model, you may have media cards that don’t span and automatically start a new clip every 4GB (about every 12 minutes with some formats). The interviewer should be mindful of these limits. If possible, all cameras should be started together and re-slated at the beginning of each new clip.

Editing workflow

df4115_iview_13Most popular nonlinear editing applications (NLE) include great features that make editing on camera interviews reasonably easy. To end up with a solid five minute piece, you’ll probably need about an hour of recorded interview material (per camera angle). When you cut out the interviewer’s questions, the little bit of chit chat at the beginning, and then repeats or false starts that an interviewee may have, then you are generally left with about thirty minutes of useable responses. That’s a 6:1 ratio.

The goal as an editor is to be a storyteller by the soundbites you select and the order into which you arrange them. The goal is to have the subject seamlessly tell their story without the aide of an on camera host or voice-over narrator. To aid the editing process use NLE tools like favorites, markers, and notes, along with human tools like written transcripts and your own notes to keep the project organized.

This is the standard order of things for me:

Sync sources and create multi-cam sequences or multi-cam clips depending on the NLE.

Pass 1 – create a sequence with all clips synced up and organized into a single timeline.

Pass 2 – clean up the interview and remove all interviewer questions.

Pass 3 – whittle down the responses into a sequence of selected answers.

Pass 4 – rearrange the soundbites to best tell the story.

Pass 5 – cut between cameras if this is a multi-camera production.

Pass 6 – clean up the final order by editing out extra words, pauses, and verbal gaffs.

Pass 7 – color correct clips, mix audio, add b-roll shots.

df4115_iview_9As I go through this process, I am focused on creating a good “radio cut” first. In other words, how does the story sound if you aren’t watching the picture. Once I’m happy with this, I can worry about color correction, b-roll, etc. When building a piece that includes multiple interviewees, you’ll need to pay attention to several other factors. These include getting a good mix of diversity – ethnic, gender, job classification. You might want to check with the client first as to whether each and every person interviewed needs to be used in the video. Clearly some people are going to be duds, so it’s best to know up front whether or not you’ll need to go through the effort to find a passable soundbite in those cases or not.

There are other concerns when re-ordering clips among multiple people. Arranging the order of clips so that you can cut between alternating left and right-framed shots makes the cutting flow better. Some interviewees comes across better than others, however, make sure not to lean totally on these responses. When you get multiple, similar responses, pick the best one, but if possible spread around who you pick in order to get the widest mix of respondents. As you tell the story, pay attention to how one soundbite might naturally lead into another – or how one person’s statement can complete another’s thoughts. It’s those serendipitous moments that you are looking for in Pass 4. It’s what should take the most creative time in your edit.

Philosophy of the cut

df4115_iview_11In any interview, the editor is making editorial selections that alter reality. Some broadcasters have guidelines at to what is and isn’t permissible, due to ethical concerns. The most common editorial technique in play is the “Frankenbite”. That’s where an edit is made to truncate a statement or combine two statements into one. Usually this is done because the answer went off into a tangent and that portion isn’t relevant. By removing the extraneous material and creating the “Frankenbite” you are actually staying true to the intent of the answer. For me, that’s the key. As long as your edit is honest and doesn’t twist the intent of what was said, then I personally don’t have a problem with doing it. That part of the art in all of this.

df4115_iview_10It’s for these reasons, though, that directors like Morris leave the jump cuts in. This lets the audience know an edit was made. Personally, I’d rather see a smooth piece without jump cuts and that’s where a two camera shoot is helpful. Cutting between two camera angles can make the edit feel seamless, even though the person’s expression or body position might not truly match on both sides of the cut. As long as the inflection is right, the audience will accept it. Occasionally I’ll use a dissolve, white flash or blur dissolve between sections, but most of the time I stick with cuts. The transitions seem like a crutch to me, so I use them only when there is a complete change of thought that I can’t bridge with an appropriate soundbite or b-roll shot.

df4115_iview_12The toughest interview edit tends to be when you want to clean things up, like a repeated word, a stutter, or the inevitable “ums” and “ahs”. Fixing these by cutting between cameras normally results in a short camera cut back and forth. At this point, the editing becomes a distraction. Sometimes you can cheat these jump cuts by staying on the same camera angle and using a short dissolve or one of the morphing transitions offered by Avid, Adobe, or MotionVFX (for FCPX). These vary in their success depending on how much a person has moved their body and head or changed expressions at the edit point. If their position is largely unchanged, the morph can look flawless. The more the change, the more awkward the resulting transition can be. The alternative is to cover the edit with a cutaway b-roll shot, but that’s often not desirable if this happens the first time we see the person. Sometimes you just have to live with it and leave these imperfections alone.

Telling the story through sight and sound is what an editor does. Working with on camera interviews is often the closest an editor comes to being the writer, as well. But remember that mixing and matching soundbites can present nearly infinite possibilities. Don’t get caught in the trap so many do of never finishing. Bring it to a point where the story is well-told and then move on. If the entire production is approached with some of these thoughts in mind, the end result can indeed be memorable.

©2015 Oliver Peters

Tips for Production Success – Part 2

df2015_prodtips_2_smPicking up from my last post (part 1), here are 10 more tips to help you plan for a successful production.

Create a plan and work it. Being a successful filmmaker – that is, making a living at it – is more than just producing a single film. Such projects almost never go beyond the festival circuit, even if you do think it is the “great American film”. An indie producer may work on a project for about four years, from the time they start planning and raising the funds – through production and post – until real distribution starts. Therefore, the better approach is to start small and work your way up. Start with a manageable project or film with a modest budget and then get it done on time and in budget. If that’s a success, then start the next one – a bit bigger and more ambitious. If it works, rinse and repeat. If you can make that work, then you can call yourself a filmmaker.

Budget. I have a whole post on this subject, but in a nutshell, an indie film that doesn’t involve union talent or big special effects will likely cost close to one million dollars, all in. You can certainly get by on less. I’ve cut films that were produced for under $150,000 and one even under $50,000, but that means calling in a lot of favors and having many folks working for free or on deferment. You can pull that off one time, but it’s not a way to build a business, because you can’t go back to those same resources and ask to do it a second time. Learn how to raise the money to do it right and proceed from there.

Contingencies at the end. Intelligent budgeting means leaving a bit for the end. A number of films that I’ve cut had to do reshoots or spend extra days to shoot more inserts, establishing shots, etc. Plan for this to happen and make sure you’ve protected these items in the budget. You’ll need them.

Own vs. rent. Some producers see their film projects as a way to buy gear. That may or may not make sense. If you need a camera and can otherwise make money with it, then buy it. Or if you can buy it, use it, and then resell it to come out ahead – by all means follow that path. But if gear ownership is not your thing and if you have no other production plans for the gear after that one project, then it will most likely be a better deal to work out rentals. After all, you’re still going to need a lot of extras to round out the package.

Shooting ratios. In the early 90s I worked on the post of five half-hour and hourlong episodic TV series that were shot on 35mm film. Back then shooting ratios were pretty tight. A half-hour episode is about 20-22 minutes of content, excluding commercials, bumpers, open, and credits. An hourlong episode is about 44-46 minutes of program content. Depending on the production, these were shot in three to five days and exposed between 36,000 and 50,000 feet of negative. Therefore, a typical day meant 50-60 minutes of transferred “dailies” to edit from – or no more than five hours of source footage, depending on the series. This would put them close to the ideal mark (on average) of approximately a 10:1 shooting ratio.

Today, digital cameras make life easier and with the propensity to shoot two or more cameras on a regular basis, this means the same projects today might have conservatively generated more than 10 hours of source footage for each episode. This impacts post tremendously – especially if deadline is a factor. As a new producer, you should strive to control these ratios and stay within the goal of a 10:1 ratio (or lower).

Block and rehearse. The more a scene is buttoned down, the fewer takes you’ll need, which leads to a tighter shooting ratio. This means rehearse a scene and make sure the camera work is properly blocked. Don’t wing it! Once everything is ready, shoot it. Odds are you’ll get it in two to three takes instead of the five or more that might otherwise be required.

Control the actors. Unless there’s a valid reason to let your actors improvise, make sure the acting is consistent. That is, lines are read in the same order each take, props are handled at the same point, and actors consistently hit their marks each take. If you stray from that discipline, the editorial time becomes longer. If allowed to engage in too much freewheeling improvisation, actors may inadvertently paint you into a corner. To avoid that outcome, control it from the start.

Visual effects planning. Most films don’t require special effects, but there are often “invisible” fixes that can be created through visual effects. For example, combining elements of two takes or adding items to a set. A recent romantic drama I post-supervised used 76 effects shots of one type or another. If this is something that helps the project, make sure to plan for it from the outset. Adobe After Effects is the ubiquitous tool that makes such effects affordable. The results are great and there are plenty of talented designers who can assist you within almost any budget range.

Multiple cameras vs. single camera vs. 4K. Some producers like the idea of shooting interviews (especially two-shots) in 4K (for a 1080 finish) and then slice out the frame they want. I contend that often 4K presents focus issues, due to the larger sensors used in these cameras. In addition, the optics of slicing a region out of a 4K image are different than using another camera or zooming in to reframe the shot. As a result, the look that you get isn’t “quite right”. Naturally, it also adds one more component that the editor has to deal with – reframing each and every shot.

Conversely, when shooting a locked-off interview with one person on-camera, using two cameras makes the edit ideal. One camera might be placed face-on towards the speaker and the other from a side angle. This makes cutting between the camera angles visually more exciting and makes editing without visible jump cuts easier.

In dramatic productions, many new directors want to emulate the “big boys” and also shoot with two or more cameras for every scene. Unfortunately this isn’t always productive, because the lighting is compromised, one camera is often in an awkward position with poor framing, or even worse, often the main camera blocks the secondary camera. At best, you might get 25% usability out of this second camera. A better plan is to shoot in a traditional single-camera style. Move the camera around for different angles. Tweak the lighting to optimize the look and run the scene again for that view.

The script is too long. An indie film script is generally around 100 pages with 95-120 scenes. The film gets shot in 20-30 days and takes about 10-15 weeks to edit. If your script is inordinately long and takes many more days to shoot, then it will also take many more days to edit. The result will usually be a cut that is too long. The acceptable “standard” for most films is 90-100 minutes. If you clock in at three hours, then obviously a lot of slashing has to occur. You can lose 10-15% (maybe) through trimming the fat, but a reduction of 25-40% (or more) means you are cutting meat and bone. Scenes have to be lost, the story has to be re-arranged, or even more drastic solutions. A careful reading of the script and conceiving that as a finished concept can head off issues before production ever starts. Losing a scene before you shoot it can save time and money on a large scale. So analyze your script carefully.

Click here for Part 1.

©2015 Oliver Peters

Tips for Production Success – Part 1

df1915_prodtips_1_smThroughout this blog, I’ve written numerous tips about how to produce projects, notably indie features, with a successful outcome in mind. I’ve tried to educate on issues of budget and schedule. In these next two entries, I’d like to tackle 21 tips that will make your productions go more smoothly, finish on time, and not become a disaster during the post production phase. Although I’ve framed the discussion around indie features, the same tips apply to commercials, music videos, corporate presentations, and videos for the web.

Avoid white. Modern digital cameras handle white elements within a shot much better than in the past, but hitting a white shirt with a lot of light complicates your life when it comes to grading and directing the eye of the viewer. This is largely an issue of art direction and wardrobe. The best way to handle this is simply to replace whites with off-whites, bone or beige colors. The sitcom Barney Miller, which earned DP George Spiro Dibie recognition for getting artful looks out of his video cameras, is said to have had the white shirts washed in coffee to darken them a bit. The whiteness was brought back once the cameras were set up. The objective in all of this is to get the overall brightness into a range that is more controllable during color correction and to avoid clipping.

Expose to the right. When you look at a signal on a histogram, the brightest part is on the righthand side of the scale. By pushing your camera’s exposure towards a brighter, slightly over-exposed image (“to the right”), you’ll end up with a better looking image after grading (color correction). That’s because when you have to brighten an image by bringing up highlights or midtones, you are accentuating the sensor noise from the camera. If the image is already brighter and the correction is to lower the levels, then you end up with a cleaner final image. Since most modern digital cameras use some sort of log or hyper gamma encoding to record a flatter signal, which preserves latitude, opening up the exposure usually won’t run the risk of clipping the highlights. In the end, a look that stretches the shadow and mids to expose more detail to the eye gives you a more pleasing and informative image than one that places emphasis on the highlight portion.

Blue vs. green-screen. Productions almost ubiquitously use green paint, but that’s wrong. Each paint color has a different luminance value. Green is brighter and should be reserved for a composite where the talent should appear to be outside. Blue works best when the composited image is inside. Paint matters. The correct paint to use is still the proper version of Ultimatte blue or green paint, but many people try to cut corners on cost. I’ve even had producers go so far as to rig up a silk with a blue lighting wash and expect me to key it! When you light the subject, move them as far away from the wall as possible to avoid contamination of the color onto their hair and wardrobe. This also means, don’t have your talent stand on a green or blue floor, when you aren’t intending to see the floor or see them from their feet to their head.

Rim lighting. Images stand out best when your talent has some rim lighting to separate them from the background. Even in a dark environment, seek to create a lighting scheme that achieves this rimming effect around their head and shoulders.

Tonal art direction. The various “blockbuster” looks are popular – particularly the “orange and teal” look. This style pushes skin tones warm for a slight orange appearance, while many darker background elements pick up green/blue/teal/cyan casts. Although this can be accentuated in grading, it starts with proper art direction in the set design and costuming. Whatever tonal characteristic you want to achieve, start by looking at the art direction and controlling this from step one.

Rec. 709 vs. Log. Digital cameras have nearly all adopted some method of recording an image with a flat gamma profile that is intended to preserve latitude until final grading. This doesn’t mean you have to use this mode. If you have control over your exposure and lighting, there’s nothing wrong with recording Rec. 709 and nailing the final look in-camera. I highly recommend this for “talking head” interviews, especially ones shot on green or blue-screen.

Microphone direction/placement. Every budding recording engineer working in music and film production learns that proper mic placement is critical to good sound. Pay attention to where mics are positioned, relative to where the person is when they speak. For example, if you have two people in an interview situation wearing lavaliere mics on their lapels, the proper placement would be on each’s inner lapel – the side closer to the other person. That’s because each person will turn towards the other to address them as they speak and thus talk over that shoulder. Having the mic on this side means they are speaking into the mic. If it were on their outer lapel, they would be speaking away from the mic and thus the audio would tend to sound hollow. For the same reasons, when you use a boom or fish pole overhead mic, the operator needs to point the mic in the direction of the person talking. They will need to shift the mic’s direction as the conversation moves from one person to the next to follow the sound.

Multiple microphones/iso mics. When recording dialogue for a group of actors, it’s best to record their audio with individual microphones (lavs or overhead booms) and to record each mic on an isolated track. Cameras typically feature on-board recording of two to four audio channels, so if you have more mics than that, use an external multi-channel recorder. When external recording is used, be sure to still record a composite track to your camera for reference.

Microphone types. There are plenty of styles and types of microphones, but the important factors are size, tonal quality, range, and the axis of pick-up. Make sure you select the appropriate mic for the task. For example, if you are recording an actor with a deep bass voice using a lavaliere, you’d be best to use a type that gives you a full spectrum recording, rather than one that favors only the low end.

Sound sync. There are plenty of ways to sync sound to picture in double-system sound situations. Synchronizing by matched timecode is the most ideal, but even there, issues can arise. Assure that the camera’s and sound recorder’s timecode generators don’t drift during the day – or use a single, common, external timecode generator for both. It’s generally best to also include a clapboard and, when possible, also record reference audio to the camera. If you plan to sync by audio waveforms (PluralEyes, FCP X, Premiere Pro CC), then make sure the reference signal on the camera is of sufficient quality to make synchronization possible.

Record wild lines on set. When location audio is difficult to understand, ADR (automatic dialogue replacement, aka “looping”) is required. This happens because the location recording was not of high quality due to outside factors, like special effects, background noise, etc. Not all actors are good at ADR and it’s not uncommon to watch a scene with ADR dialogue and have it jump out at you as the viewer. Since ADR requires extra recording time with the actor, this drives up cost on small films. One workaround in some of these situations is for the production team to recapture the lines separately – immediately after the scene was shot – if the schedule permits. These lines would be recorded wild and may or may not be in sync. The intent is to get the right sonic environment and emotion while you are still there on site. Since these situations are often fast-paced action scenes, sync might not have to be perfect. If close enough, the sound editors can edit the lines into place with an acceptable level of sync so that viewers won’t notice any issues. When it works, it saves ADR time down the road and sounds more realistic.

Click here for Part 2.

©2015 Oliver Peters

24p HD Restoration


There’s a lot of good film content that only lives on 4×3 SD 29.97 interlaced videotape masters. Certainly in many cases you can go back and retransfer the film to give it new life, but for many small filmmakers, the associated costs put that out of reach. In general, I’m referring to projects with $0 budgets. Is there a way to get an acceptable HD product from an old Digibeta master without breaking the bank? A recent project of mine would say, yes.

How we got here

I had a rather storied history with this film. It was originally shot on 35mm negative, framed for 1.85:1, with the intent to end up with a cut negative and release prints for theatrical distribution. It was being posted around 2001 at a facility where I worked and I was involved with some of the post production, although not the original edit. At the time, synced dailies were transferred to Beta-SP with burn-in data on the top and bottom of the frame for offline editing purposes. As was common practice back then, the 24fps film negative was transferred to the interlaced video standard of 29.97fps with added 2:3 pulldown – a process that duplicates additional fields from the film frames, such that 24 film frames evenly add up to 60 video fields in the NTSC world. This is loaded into an Avid, where – depending on the system – the redundant fields are removed, or the list that goes to the negative cutter compensates for the adjustments back to a frame-accurate 24fps film cut.

df_24psdhd_5For the purpose of festival screenings, the project file was loaded into our Avid Symphony and I conformed the film at uncompressed SD resolution from the Beta-SP dailies and handled color correction. I applied a mask to hide the burn-in and ended up with a letter-boxed sequence, which was then output to Digibeta for previews and sales pitches to potential distributors. The negative went off to the negative cutter, but for a variety of reasons, that cut was never fully completed. In the two years before a distribution deal was secured, additional minor video changes were made throughout the film to end up with a revised cut, which no longer matched the negative cut.

Ultimately the distribution deal that was struck was only for international video release and nothing theatrical, which meant that rather than finishing/revising the negative cut, the most cost-effective process was to deliver a clean video master. Except, that all video source material had burn-in and the distributor required a full-height 4×3 master. Therefore, letter-boxing was out. To meet the delivery requirements, the filmmaker would have to go back to the original negative and retransfer it in a 4×3 SD format and master that to Digital Betacam. Since the negative was only partially cut and additional shots were added or changed, I went through a process of supervising the color-corrected transfer of all required 35mm film footage. Then I rebuilt the new edit timeline largely by eye-matching the new, clean footage to the old sequence. Once done and synced with the mix, a Digibeta master was created and off it went for distribution.

What goes around comes around

After a few years in distribution, the filmmaker retrieved his master and rights to the film, with the hope of breathing a little life into it through self-distribution – DVDs, Blu-rays, Internet, etc. With the masters back in-hand, it was now a question of how best to create a new product. One thought was simply to letter-box the film (to be in the director’s desired aspect) and call it a day. Of course, that still wouldn’t be in HD, which is where I stepped back in to create a restored master that would work for HD distribution.

Obviously, if there was any budget to retransfer the film negative to HD and repeat the same conforming operation that I’d done a few years ago – except now in HD – that would have been preferable. Naturally, if you have some budget, that path will give you better results, so shop around. Unfortunately, while desktop tools for editors and color correction have become dirt-cheap in the intervening years, film-to-tape transfer and film scanning services have not – and these retain a high price tag. So if I was to create a new HD master, it had to be from the existing 4×3 NTSC interlaced Digibeta master as the starting point.

In my experience, I know that if you are going to blow-up SD to HD frame sizes, it’s best to start with a progressive and not interlaced source. That’s even more true when working with software, rather than hardware up-convertors, like Teranex. Step one was to reconstruct a correct 23.98p SD master from the 29.97i source. To do this, I captured the Digibeta master as a ProResHQ file.

Avid Media Composer to the rescue


When you talk about software tools that are commonly available to most producers, then there are a number of applications that can correctly apply a “reverse telecine” process. There are, of course, hardware solutions from Snell and Teranex (Blackmagic Design) that do an excellent job, but I’m focusing on a DIY solution in this post. That involves deconstructing the 2:3 pulldown (also called “3:2 pulldown”) cadence of whole and split-field frames back into only whole frames, without any interlaced tearing (split-field frames). After Effects and Cinema Tools offer this feature, but they really only work well when the entire source clip is of a consistent and unbroken cadence. This film had been completed in NTSC 29.97 TV-land, so frequently at cuts, the cadence would change. In addition, there had been some digital noise reduction applied to the final master after the Avid output to tape, which further altered the cadence at some cuts. Therefore, to reconstruct the proper cadence, changes had to be made at every few cuts and, in some scenes, at every shot change. This meant slicing the master file at every required point and applying a different setting to each clip. The only software that I know of to effectively do this with is Avid Media Composer.

Start in Media Composer by creating a 29.97 NTSC 4×3 project for the original source. Import the film file there. Next, create a second 23.98 NTSC 4×3 project. Open the bin from the 29.97 project into the 23.98 project and edit the 29.97 film clip to a new 23.98 sequence. Media Composer will apply a default motion adapter to the clip (which is the entire film) in order to reconcile the 29.97 interlaced frame rate into a 23.98 progressive timeline.

Now comes the hard part. Open the Motion Effect Editor window and “promote” the effect to gain access to the advanced controls. Set the Type to “Both Fields”, Source to “Film with 2:3 Pulldown” and Output to “Progressive”. Although you can hit “Detect” and let Media Composer try to decide the right cadence, it will likely guess incorrectly on a complex file like this. Instead, under the 2:3 Pulldown tab, toggle through the cadence options until you only see whole frames when you step through the shot frame-by-frame. Move forward to the next shot(s) until you see the cadence change and you see split-field frames again. Split the video track (place an “add edit”) at that cut and step through the cadence choices again to find the right combination. Rinse and repeat for the whole film.

Due to the nature of the process, you might have a cut that itself occurs within a split-field frame. That’s usually because this was a cut in the negative and was transferred as a split-field video frame. In that situation, you will have to remove the entire frame across both audio and video. These tiny 1-frame adjustments throughout the film will slightly shorten the duration, but usually it’s not a big deal. However, the audio edit may or may not be noticeable. If it can’t simply be fixed by a short 2-frame dissolve, then usually it’s possible to shift the audio edit a little into a pause between words, where it will sound fine.

Once the entire film is done, export a new self-contained master file. Depending on codecs and options, this might require a mixdown within Avid, especially if AMA linking was used. That was the case for this project, because I started out in ProResHQ. After export, you’ll have a clean, reconstructed 23.98p 4×3 NTSC-sized (720×486) master file. Now for the blow-up to HD.

DaVinci Resolve

df_24psdhd_1_smThere are many applications and filters that can blow-up SD to HD footage, but often the results end up soft. I’ve found DaVinci Resolve to offer some of the cleanest resizing, along with very fast rendering for the final output. Resolve offers three scaling algorithms, with “Sharper” providing the crispest blow-up. The second issue is that since I wanted to restore the wider aspect, which is inherent in going from 4×3 to 16×9, this meant blowing up more than normal – enough to fit the image width and crop the top and bottom of the frame. Since Resolve has the editing tools to split clips at cuts, you have the option to change the vertical position of a frame using the tilt control. Plus, you can do this creatively on a shot-by-shot basis if you want to. This way you can optimize the shot to best fit into the 16×9 frame, rather than arbitrarily lopping off a preset amount from the top and bottom.

df_24psdhd_3_smYou actually have two options. The first is to blow up the film to a large 4×3 frame out of Resolve and then do the slicing and vertical reframing in yet another application, like FCP 7. That’s what I did originally with this project, because back then, the available version of Resolve did not offer what I felt were solid editing tools. Today, I would use the second option, which would be to do all of the reframing strictly within Resolve 11.

As always, there are some uncontrollable issues in this process. The original transfer of the film to Digibeta was done on a Rank Cintel Mark III, which is a telecine unit that used a CRT (literally an oscilloscope tube) as a light source. The images from these tubes get softer as they age and, therefore, they require periodic scheduled replacement. During the course of the transfer of the film, the lab replaced the tube, which resulted in a noticeable difference in crispness between shots done before and after the replacement. In the SD world, this didn’t appear to be a huge deal. Once I started blowing up that footage, however, it really made a difference. The crisper footage (after the tube replacement) held up to more of a blow-up than the earlier footage. In the end, I opted to only take the film to 720p (1280×720) rather than a full 1080p (1920×1080), just because I didn’t feel that the majority of the film held up well enough at 1080. Not just for the softness, but also in the level of film grain. Not ideal, but the best that can be expected under the circumstances. At 720p, it’s still quite good on Blu-ray, standard DVD or for HD over the web.

df_24psdhd_4_smTo finish the process, I dust-busted the film to fix places with obvious negative dirt (white specs in the frame) caused by the initial handling of the film negative. I used FCP X and CoreMelt’s SliceX to hide and cover negative dirt, but other options to do this include built in functions within Avid Media Composer. While 35mm film still holds a certain intangible visual charm – even in such a “manipulated” state – the process certainly makes you appreciate modern digital cameras like the ARRI ALEXA!

As an aside, I’ve done two other complete films this way, but in those cases, I was fortunate to work from 1080i masters, so no blow-up was required. One was a film transferred in its entirety from a low-contrast print, broken into reels. The second was assembled digitally and output to intermediate HDCAM-SR 23.98 masters for each reel. These were then assembled to a 1080i composite master. Aside from being in HD to start with, cadence changes only occurred at the edits between reels. This meant that it only required 5 or 6 cadence corrections to fix the entire film.

©2014 Oliver Peters

Film editing stages – Sound

df_filmsoundeditLike picture editing, the completion of sound for a film also goes through a series of component parts. These normally start after “picture lock” and are performed by a team of sound editors and mixers. On small, indie films, a single sound designer/editor/mixer might cover all of these roles. On larger films, specific tasks are covered by different individuals. Depending on whether it’s one individual or a team, sound post can take anywhere from four weeks to several months to complete.

Location mixing – During original production, the recording of live sound is handled by the location mixer. This is considered mixing, because originally, multiple mics were mixed “on-the-fly” to a single mono or stereo recording device. In modern films with digital location recordings, the mixer tends to record what is really only a mixed reference track for the editors, while simultaneously recording separate tracks of each isolated microphone to be used in the actual post production mix.

ADR – automatic dialogue replacement or “looping”. ADR is the recording of replacement dialogue in sync with the picture. The actors do this while watching their performance on screen. Sometimes this is done during production and sometimes during post. ADR will be used when location audio has technical flaws. Sometimes ADR is also used to record additional dialogue – for instance, when an actor has his or her back turned. ADR can also be used to record “sanitized” dialogue to remove profanity.

Walla or “group loop” – Additional audio is recorded for groups of people. This is usually for background sounds, like guests in a restaurant. The term “walla” comes from the fact that actors were (and often still are) instructed to say “walla, walla, walla” instead of real dialogue. The point is to create a sound effect of a crowd murmuring, without any recognizable dialogue line being heard. You don’t want anything distinctive to stand out above the murmur, other than the lead actors’ dialogue lines.

Dialogue editing – When the film editor (i.e. the picture editor) hands over the locked cut to the sound editors, it generally will include all properly edited dialogue for the scenes. However, this is not prepared for mixing. The dialogue editor will take this cut and break out all individual mic tracks. They will make sure all director’s cues are removed and they will often add room tone and ambience to smooth out the recording. In addition, specific actor mics will be grouped to common tracks so that it is easier to mix and apply specific processing, as needed, for any given character.

Sound effects editing/sound design – Sound effects for a film come from a variety of sources, including live recordings, sound effects libraries and sound synthesizers. Putting this all together is the role of the sound effects editor(s). Because many have elevated the art, by creating very specific senses of place, the term “sound designer” has come into vogue. For example, the villain’s lair might always feature certain sounds that are identifiable with that character – e.g. dripping water, rats squeaking, a distant clock chiming, etc. These become thematic, just like a character’s musical theme. The sound effects editors are the ones that record, find and place such sound effects.

Foley – Foley is the art of live sound effects recording. This is often done by a two-person team consisting of a recordist and a Foley walker, who is the artist physically performing these sounds. It literally IS a performance, because the walker does this in sync to the picture. Examples of Foley include footsteps, clothes rustling, punches in a fight scene and so on. It is usually faster and more appropriate-sounding to record live sound effects than to use library cues from a CD.

In addition to standard sound effects, additional Foley is recorded for international mixes. When an actor deliveries a dialogue line over a sound recorded as part of a scene – a door closing or a cup being set on a table – that sound will naturally be removed when English dialogue is replaced by foreign dialogue in international versions of the film. Therefore, additional sound effects are recorded to fill in these gaps. Having a proper international mix (often called “fully filled”) is usually a deliverable requirement by any distributor.

Music – In an ideal film scenario, a composer creates all the music for a film. He or she is working in parallel with the sound and dialogue editors. Music is usually divided between source cues (e.g. the background songs playing from a jukebox at a bar) and musical score.

Recorded songs may also be used as score elements during montages. Sometimes different musicians, other than the composer, will create songs for source cues or for use in the score. Alternatively, the producers may license affordable recordings from unsigned artists. Rarely is recognizable popular music used, unless the production has a huge budget. It is important that the producers, composer and sound editors communicate with each other, to define whether items like songs are to be treated as a musical element or as a background sound effect.

The best situation is when an experienced film composer delivers all completed music that is timed and synced to picture. The composer may deliver the score in submixed, musical stems (rhythm instruments separated from lead instruments, for instance) for greater control in the mix. However, sometimes it isn’t possible for the composer to provide a finished, ready-to-mix score. In that case, a music editor may get involved, in order to edit and position music to picture as if it were the score.

Laugh tracks – This is usually a part of sitcom TV production and not feature films. When laugh tracks are added, the laughs are usually placed by sound effects editors who specialize in adding laughs. The appropriate laugh tracks are kept separate so they can be added or removed in the final mix and/or as part of any deliverables.

Re-recording mix – Since location recording is called location mixing, the final, post production mix is called a re-recording mix. This is the point at which divergent sound elements – dialogue, ADR, sound effects, Foley and music – all meet and are mixed in sync to the final picture. On a large film, these various elements can easily take up 150 or more tracks and require two or three mixers to man the console. With the introduction of automated systems and the ability to completely mix “in the box”, using a DAW like Pro Tools, smaller films may be mixed by one or two mixers. Typically the lead mixer handles the dialogue tracks and the second and third mixers control sound effects and music. Mixing most feature films takes one to two weeks, plus the time to output various deliverable versions (stereo, surround, international, etc.).

The deliverable requirements for most TV shows and features are to create a so-called composite mix (in several variations), along with separate stems for dialogue, sound effects and music. A stem is a submix of just a group of component items, such as a stereo stem for only dialogue.The combination of the stems should equal the mix. By having stems available, the distributors can easily create foreign versions and trailers.

©2013 Oliver Peters

Film editing stages – Picture


While budding filmmakers have a good idea of what happens during the production phase of shooting a film, most have little idea about what happens in post. Both picture and sound go through lengthy and separate editorial processes. These often become a rude awakening for new directors when it pertains to the time and budget requirements. These are the basic steps every modern film goes through in getting to the finish line.

First cut – This stage goes by many names – first cut, first assembly, editor’s cut, etc. In general, this is the first version of the fully-assembled film, including all the scenes edited according to the script. Depending on the editor and the post schedule, this cut may be very rough – or it might be a reasonably polished edit. If the editing happens concurrent to the studio and location filming, then often there will be a “first assembly” and a subsequent “editor’s cut”. The former is a quick turnaround version, so that everyone can make sure the coverage is adequate. The latter is a more refined version.

Some productions employ an on-set editor who is the person generating this “first assembly”. That editor is then often replaced by the main film editor, who starts after all production is completed. In that situation, the “editor’s cut” might be completely different in style, pace and technique from the first version. No matter how you get there, the intent of this step is to properly represent the intention of the script without concern for length or solving any content or script challenges.

Director’s cut – Once the editor has completed the first cut of the film, then the director steps in. He or she works with the editor to complete the cut of the film. Directors often deviate from the written scene. Sometimes this is sufficiently communicated to the editor to show up that way in the first cut. Sometimes it isn’t, because it lives in the director’s mind as the production proceeds. During the “director’s cut” phase, the director and editor work closely to adjust the cut to reflect the director’s vision.

Many directors and editors repeatedly work together on films and form a partnership of sorts. In these situation, the editor has a good idea of what the director wants and often the director only needs to give notes and review the cut periodically. Other directors like to be very “hands on” and will work closely with the editor, reviewing every take and making adjustments as needed.

Depending on the film and whether or not the director is DGA (Directors Guild), this stage will take a minimum of 20 days (DGA low budget) or 10 weeks (DGA standard) or longer. The goal is for the director and editor to come up with the best film possible, without interference from outside parties, including the producers. At this point, the film may go through severe changes, including shortening, losing and/or re-arranging scenes and even the addition of new content, like insert shots and new voice-over recordings.

Producer’s cut – After the director has a shot at the film, now it’s time to make adjustments according to studio notes, producer comments and feedback from official and unofficial test screenings. If the director hasn’t yet brought the film into line – both story-wise and length-wise – now is the time to do that. Typically most indie films are targeted at the 90-100 minute range. If your first cut or director’s cut is 120 minutes or longer, then it’s going to have to be cut down by a significant amount.

Typically you can shorten a film by 10% through trimming and shortening scenes. A reduction of 25% or more means that shots and whole scenes have to go. This can be a painful experience for the director, who has suffered through the agony, time and expense of getting these scenes and shots recorded. The editor, on the other hand, has no such emotional investment and can be more objective. Whichever way the process moves forward, the point is to get the cut to its final form.

Depending on the production, this version of the film might also include temporary sound effects, music and visual effects that have been added by the editor and/or assistants. Often this is needed to fully appreciate the film when showing it in test screenings.

Locked picture – The goal of these various editing steps is to explore all creative options in order to end up with a film that will not go through any further editing changes. This means, no revisions that change time or selected shots. The reason for a “locked picture” is so that the sound editing team and the visual effects designers can proceed with their work without the fear that changes will undo some of their efforts. Although large budget films have the luxury of making editorial changes after this point, it is unrealistic for smaller indie films. “Locking the cut” is absolutely essential if you want to get the best effort out of the post team, as well as stay within your budget.

Visual effects – If your film requires any visual effects shots, these are best tackled after picture lock. The editors will hand off the required source elements to the visual effects company or designers so they can do their thing. Editors are typically not involved in visual effects creation, other than to communicate the intent of any temp effects that have created and to make sure the completed VFX shots integrate properly back into the picture.

Sound editorial This will be covered in depth in the next blog post. It has its own set of steps and usually takes several weeks to several months to complete.

Conform and grade – Prior to this step, all editing has been performed with “proxy” media. During the “finishing” stage of the film, the original camera media is “conformed” to the locked cut that was handed over from the film editor. This conform step is typically run by an online editor who works in tandem with the colorist. Sometimes this is performed by the colorist and not a separate individual. On very low budget films, the film editor, online editor and colorist might all be the same person. During conforming, the objective is to frame-accurately re-create the edit, including all reframing, speed ramps and to integrate all final visual effects shots. From this point the film goes to color correction for final grading. Here the colorist matches all shots to establish visual consistency, as well as to add any subjective looks requested by the director or director of photography. The last process is to marry the sound mix back to the picture and then generate the various deliverable masters.

©2013 Oliver Peters