“Milo” is a short documentary filmed as a midterm assignment for a film class at Princeton University. My family acquired Milo about 1.5 years ago when he was just under a year old. The family that got him originally didn’t have the ability to manage such a smart & high-energy dog and so decided to look for a new home for him. My family, the Powell’s, decided to bring him in and it was one of the best decisions we could have made. He is an awesome dog and brings a lot to our family – as with all the strays we have brought in.
I am happy with this short-documentary. I think it really captures the daily spirit & attitude of Milo. If I had to do another documentary for class, I don’t think I would choose an animal. I shot over 250gb of RAW footage mostly for the fact that you never know when an animal is going to do what it’s going to do and so you have to keep the cameras rolling far longer than you would with a human subject. One of the biggest things I learned was an appreciation for people who have the skill & patience to film animals. For me, it was nearly impossible to storyboard for the film as I didn’t know what Milo would do during the three days I had to film him. It was a bit of a reversal from the traditional way of filming something. I also had to do the voiceover for Milo, and I learned very quickly how hard it is to really capture the “feel” you are going for in an animal’s internal dialogue. This has really given me an appreciation for actors that provide the voice for animals in movies.
For a class assignment, I would have rather shot in Canon’s standard h.264 codec so as to save space, not worry about rendering cDNG sequences into a ProRes, etc. However, I quickly realized after pulling up the footage that Milo isn’t an easy dog to film. His hair is BLACK. If I exposed for detail in Milo, everything else would be blown out. If I exposed for the scene, he would just be a moving black blob on the screen with some white eyes protruding out. Shooting in RAW format allowed me to shoot for the middle of the two exposures and on almost every clip I pulled the highlights down completely & lifted the shadows 50-80% in Adobe Camera Raw to enable me capturing detail in both Milo & the scene. It was then rendered into ProRes 422.
If I had more time, I would have spent more time color grading the footage to be more warm overall but then adding a silvery blue to the shadows as this blue really brings out the detail in Milo’s hair. For the sake of sanity during my busy season here at Princeton, I instead used a film-preset in Premiere that was more cool in nature so as to keep the focus on Milo and not have to dive into too much color grading.
One thing that annoys me is the framing of the interviewees. We didn’t have a boom pole nor a stand to allow for the Zoom H1 to be about the subjects head or cleanly off to the side. This required Shelby to sit with the mic and hold it as close to the base of the frame as possible. We had to compromise between the “headspace” over the subject & how close we wanted to mic to the interviewee. We opted to err on the side of better audio as audio makes or breaks a documentary. I also did not have a tripod with me on this trip, so I actually just held the camera on a gimbal for the shooting of the interview. Not a setup I’d use on a professional gig, but for a class documentary, I got the job done.
Equipment, Settings, & DP Decisions:
-Canon 5D Mark III w/ Magic Lantern Raw
-Canon 35mm F/2 IS
-Beholder EC1 Gimbal
-Zoom H1 using its built in mic.
This was shot in an aspect ratio of 2.35:1 @ 50 FPS & usually a shutter speed of 1/100. If I had to do it again we would have shot at 30fps. But knowing I would be shooting unpredictable animals and that I would likely be jerking the camera around to follow the action, I opted for a higher FPS to as to minimize motion blur & jello effect from Canon’s rather crappy rolling shutter. I also thought we might do slowmo, but I liked the fast-paced feel of the song, the fast FPS, and overall vibe of the film and so decided not to slow down any footage.
The majority of the footage used here was stabilized using the gimbal. However, for some tight spaces & a lack of time to capture a moment, I used handheld.
The ZoomH1 was set to 48 & 24. In Adobe Audition, I normalized the audio to -.1, then set a hard limiter at -.1, then used the Tube Modeled Compression Tool & set the baseline slightly over the average peak, then compressed the peaks over the average by usually 3:1, but sometimes as high as 10:1. Gain was then added until audio would peak. The before & after audio is night & day different.
(Tennessee Hayride by Audionautix)