Five Mistakes I Made Shooting My First (Mini) Documentary

Update: Content has been moved to feedthesensor.com. If you’d like to continue reading about my experiences with cameras, please head there. Trying to consolidate platforms. Thanks!

Back in March, I got the bug to shoot a documentary. COVID had just begun to take hold in my area, Connecticut, and I was anxious at the sudden emptiness on my calendar. Where had all the clients gone? When would they come back? And what could I do now that I had been “given” too much time to handle?

I think many filmmakers out there are facing that same dilemma: what do I do with all this time?

A documentary, of course, would fill in that gap (somewhat). And more than that it would be something that I could call mine, that I could be proud of once all was said and done. In the three year journey into filmmaking that I had been on, I had not given myself the opportunity to put together a project, and so now, more than ever, was the chance to do something about that.

Also around this time, I had just picked up a new camera, the Z Cam E2 S6, and was on the go go to use it wherever I could. I wanted to test its dynamic range, its battery life, to really get knee deep in to the use of the cube form factor of this camera that was so different from the DSLRs that I had previously owned and used. Is it like a rental cine camera? No, not really. But the image is on par. I wrote a review about it here.

I set out on making the documentary by forming a list of potential subjects on a yellow notepad. I went to Google, searched across Connecticut for interesting people and happenings that could turn into a full-fledged story on camera, and mulled over different ideas that felt something real until one stuck.

The one that stuck was about Matthew Parkinson at Dragon’s Breath Forge. He’s a two-time Forged in Fire champion and Damascus-steel bladesmith who teaches classes and makes swords, axes, tools, and kitchen knives for clients all over the world.

I got in contact with him. We ended up shooting in March and April and I edited the twenty-three minute doc through August this year. It was a great experience and I learned a ton, which is why I am writing this post here. This post is about a few of the mistakes that I made and things that I would have done differently.

Also, if you have the time and would like to see how hand forged Damascus steel knives are made and get to know Matthew Parkinson, please take a look at “How Damascus Steel Knives Are Made With Matthew Parkinson.”

So let’s get into it.

Stop talking and be aware of audio

One of the first things I learned when getting an education in broadcast media was to “edit audio first,” meaning to build the narrative of whatever you are producing from the audio before even thinking about touching the video portion. The video complements the audio and the audio is the narrative, not the other way around.

In the corporate world of video, things are a little different. At the base level, you get your talking heads and some B roll and slap on a royalty free track. Audio from B roll generally does not get used, in my experience (often it’s disabled in the export process when making a catalog of selects for a client).

What became immediately apparent when I dove into this edit was that I talk, a lot. I make mouth noises like “hmm” and “uh huh” as if I’m Michael Barbaro on The Daily. I engage in conversation while shooting B roll as if the audio of that B roll is unimportant. I wonder why that might be the case?

I did not realize while recording that I would need the sound of that drill press sans my voice asking, “How’s the weather been?” Or that I would need the end of something Matthew was saying without me going, “Right, right, and can you talk about the hammer?” I cannot tell you how many sniffles I had to mask out and the number of shots that did not get used because I made the mistake of opening my mouth.

Some shots were recoverable, sure. I found audio bits and pieces and massaged them until they fit to mask out the bad portions. This kind of masseusery, while challenging (and something of a lame humblebrag in the vein of “yeah I dabble in audio wizardry”), is a time waste that takes away from the edit.

In short, when the tally light is on, keep your mouth shut. Be aware of the sounds you are making and limit as much as possible things that are difficult to control, like sniffling in a cold room. You will need all of the audio in the B roll shots you choose to include, so make sure that it does not include you in it (unless that is a “style thing”).

And always make sure you are getting audio on B roll, whether through the in camera audio or a shotgun mic mounted on the camera. If your camera supports it, I highly recommend using an on-camera shotgun mic with a secure cable connection. Even a slightly bumped 3.5mm-type connection will record no audio whatsoever, so make sure what you get is secure. There have been times where a small bump of a 3.5mm cable resulted in the loss of audio for twenty minutes of selects (and I mean selects, the shots I pulled from uncut footage).

Hold on action

Knowing when to hold on the action of the subject and when to move to a new position is a learned skill. The rookie mistake is to forget to let shots breathe.

Not two seconds after settling on a shot of the hammer coming down on the hot iron, I could have been off again getting closer and lower for that closeup of the metal shedding off when the hammer hits. Then back to a wide with Matthew in the right with lots of negative space on the left. Then a closeup of his boots. In the edit, the whole recording here might have looked like three minutes on the track, but in reality it would have been just sixteen seconds of usable footage — four for the medium shot of the hammer, four for the closeup, four for the wide, and four for the boots.

When Matthew inevitably talks at length about the process of the hammer coming down on the iron, I would have been in a tight place with no long shots to let the viewer take in his words uninterrupted.

Even the 10–10–10 shot style of 10 second hold, 10 second pan-and-back, and 10 second tilt-and-back gives only 10 seconds of usable footage per shot. While people say “never linger on a shot for longer than 10 seconds” as a rule of thumb, that rule applies when the information that the shot is giving is given fully in under 10 seconds. Some shots have a lot of information spread out over an extended period, and those shots if left intact at thirty to sixty seconds in length can be mesmerizing to watch.

This is a weak spot for me. When I went about shooting this doc, I made it a goal to practice holding on action. I frequently fought and sometimes lost the urge to jump in closer. Once in the edit, I found myself both patting and kicking myself on the back: the extended shots I managed to get amplified the narrative in a way that no sequence of short shots could have and the extended shots that I broke for a closeup or some other angle lost important information in the move to the new shot.

Mix it up on repeatable actions

A shot is not boring if the information from action in it is new.

Much of Matthew’s work on the hammer is repetitive. He heats the block in the forge, carries it to the anvil, hammers it, returns it to the forge, and then begins the set of actions again once the block is back up to temperature. Since this process would be shown more than once, I needed to find different camera angles for each set to keep up the flow of new visual information.

Maintaining an awareness of repetition is easier said than done. When Matthew was flattening the blocks so that he could stack and then weld them without gaps in between, I repeated a near identical series of shots that I had gotten a week prior when he was flattening the first blocks that had been forged. The single-point perspective medium of Matthew on the machine was the same as the single-point perspective on the second go around. In the edit, I could not use this second series of shots because though the blocks were different, the visual lacked new information.

Luckily, I caught the repetition early (when playing with the footage in DaVinci at the end of the day). In another round of flattening, I focused on aspects of the machine and Matthew for some cutaways that I then inserted in the previous sets.

Consider a second interview for anything you missed

If there’s a universal statement that can be made about giving interviews, it’s that the process of transforming the interview into the narrative will stir questions that had gone unasked. If it is possible to do a second interview, do it.

For me as a first-time interviewer for a doc, the question was a simple “Why?” Had I asked that question, or a form of it (“tell me more”), as often as I made Barbaro noises in the first interview, I would have had a deeper, more complex narrative.

But, I didn’t. So I went back to ask “Why?” and I kept quiet so that I could hear Matthew’s answers. The doc was a skeleton before this second interview. Those answers filled out the skeleton into a full form that I could call, with some confidence, a documentary.

Write down these missed questions as you edit the first draft. When the draft is ready to send to some friends for review, ask those friends to write down any questions they have while watching it.

Send a polished draft, not a rough cut

Before you send off that draft to your friends, make sure the draft is a polished draft. Do not send a rough cut. How many times has sending a rough cut to a client with a disclaimer ever stopped that client from telling you the audio is bad? Or that the colors are washed out (a “digital negative” I tell them, referring to LOG)?

How are your friends going to make accurate critiques of your doc if they must first define and then see beyond the rough edges that you put a disclaimer on? How are they to know the difference between an element worth critiquing and an element of the rough cut stage?

From my experience with sending rough cuts, more energy is put toward sorting between rough cut elements and elements worth critiquing than actually watching the cut.

I almost sent an early rough cut. I had exported it and had the email ready to go. I was excited. I wanted someone to watch and go “Wow, you made this!” in a flurry of egotism. But then I remembered all of the times when sending a rough cut resulted in a positive outcome — not once. Each time, what I had rashly thought would be an ego-boosting exercise with a laid-back client turned into a self-esteem withering series of explanation and clarification emails about the complicated nature of the editing process.

I held off on sending this doc until I had locked in the audio and visuals as best I could. When I finally sent it off to some friends, I got substantive responses with valuable critiques. Other than for a lone comment about a two-second shot that had somehow forgone color in the rendering process, the people I sent the doc to were not distracted by any rough “edges” — they were able to devote more attention to watching and critiquing the narrative.

So, save yourself the time and frustration. Save your friends the time, too. Send them a polished draft, not a rough cut.

To sum it up, keep quiet when the tally light is on, hold on action, vary your shots, get a second interview, and send a polished draft. These are just a few of the things I learned while shooting this doc.

I know this piece will be rudimentary for some, but there are plenty of other people out there like me who would benefit from seeing where I went wrong before diving into the making of a documentary.

If you want to try your hand at spying the bandaids I talked about in the post, “How Damascus Steel Knives Are Made With Matthew Parkinson” is linked below.

And if you want to see what I have since been up to, I have been starting up a new video production company called Dovetail Filmworks.