What is sleep…?
Earlier this month I had the pleasure of directing a short film workshop for an old script I had found from a few years back, titled Prisoners. Joanna had recently acquired a new camera (Sony A7s) and wanted to shoot some test footage to try it out. We decided we could try a few scenes from this script and she organised a couple of actors to help out, Danxia Yang and Jimmy Sky (who we had previously worked with on Nightfall).
We prepped to shoot the middle scenes of the script (simple dialogue between two characters), totalling around seven pages. On the day, we only had a few hours to setup, rehearse and shoot, and there wasn’t much notice given to the actors, so it was big ask for them to learn seven pages of dialogue. Nonetheless, they picked it up really quickly with minimal rehearsal and we began shooting.
We were able to get through it quickly, with a couple of camera setups for each scene. I didn’t bother with coverage for the second scene as we were short on time, and I wanted it to feel more intimate anyway, so we just did a two-shot, with two takes back-to-back. It would have been nice to have a bit of coverage and some inserts to help with editing the two takes together, but I managed to scale the two-shot image up to become a close-up for each character, since we shot in 4K, so that helped hide cutting between takes.
For the final scene, I wanted it to feel more energetic and unstable, so we shot it mostly handheld with a moving camera. I did want an angle on Danxia with the tripod, so she felt stable and in control, until the turning point in the scene, which is when we swap to handheld for her as well.
After the shoot was done and I got back home, I began reviewing the footage and began the edit. Firstly, as always, the boring stuff. Organising video and sound files, syncing audio tracks (luckily PluralEyes handled most of that for me) and setting up the timeline. We shot 4K at 16:9, so I just used that for the sequence and added my own 2:39 overlay.
The edit itself was fairly straightforward. I just worked on the scenes sequentially, cutting together the best of each take, though we only had a couple of takes for each setup which posed some challenges for certain moments. One small, but annoying thing, was a continuity error with the background in this shot.
After this take, Joanna suggested putting the beads down to help cover the obvious kitchen behind and provide a cleaner, less distracting image. I also wanted Jimmy’s leg in the foreground to add some depth. It definitely improved the shot, but meant it kept changing when cutting between the two takes. I didn’t want to only use take two footage, as there was great stuff in the first one, so I opted to just composite the right side of frame from the first take, on top of the second. This was simple for the most part in After Effects, since both shots were done from a tripod which didn’t move, so it was a simple copy-paste with a mask on top.
The only issue was when Jimmy walks across the frame in one shot, so I had to rotoscope his legs out, but that didn’t take too long.
Early on in the edit, I was playing around with Lumetri to see how much I could manipulate the colour, as I had never shot on this camera/format before (S-Log), and the initial image is super duper flat. Turns out, it is very pliable, even though it’s not a RAW format, which was a great surprise. I actually stumbled upon the grade for this film accidentally. I had started with a much less aggressive day-to-night and colour palette, instead just trying to add contrast and a little colour.
As I was trying things, I accidentally duplicated the Lumetri effect on the shot so it was applying it twice. As I was about to undo my mistake, I noticed that it actually looked really cool, and was far more aggressive than I had even considered trying. I didn’t think it would have been so easy to make it seem so dark when I first reviewed the footage, but the footage held up really well. I changed my grade to match the doubled-up shot and it turned out to be a pretty convincing dank, dark room, as opposed to a bright happy apartment.
You copy that grade across all shots, make some tweaks where needed, and you have a stylin’ film!
Once the grade was done, all that remained was the final sound mixing, which I did in Audition as usual. Some of the levels were a little low for Danxia’s shots, but I was able to boost them well enough. A bit of noise in the background, but definitely high-quality audio for a simple workshop. I am very grateful that Jo was willing and able to provide and operate the lavs, as well as camera.
Then all that was left was the final render! I did have a bunch of issues with that though, as Premiere decided it didn’t want to export properly, but I got there in the end and the final result is embedded above. Watch, enjoy, and hopefully it makes some sense to you, even though there are two scenes missing, one before and one after this excerpt. Hopefully I’ll have the opportunity to shoot the entire film for real sometime.
Huge thanks to Jimmy and Danxia for coming out for this shoot and lending their talents and time, as well as one-woman-band Joanna for doing lighting, audio and camera. Couldn’t have done it without you all, and I’m super pleased with the end result. A great learning experience for me and a good trial run for the future. Special thanks to Billie as always for her hospitality and for all the fruit!
TLDR; I edited the above music video for a competition. Enjoy.
So a little while back, Adobe announced an editing competition called Make the Cut, in which it challenged editors to put together their own cut of the Imagine Dragons music video for their new song Believer. They provided the song file, and all of the uncut video assets and let us make our own edit. It sounded like it’d be a fun challenge, give me a chance to flex my rusty editing muscles and didn’t hurt that they offered a nice cash prize.
I hadn’t even heard of Imagine Dragons before, but apparently they’re really popular, and I decided not to watch their official music video before attempting mine, as I didn’t want to influence my edit. I started the process by listening to the song a few times with the lyrics up to get a sense of the pace and story, as even in music videos I prefer to focus on storytelling, rather than just cool shots and lip-syncing. That is why I opted to use slower cuts and edit the video much like a short film, leaving time to breathe and take in what’s happening, rather than just assaulting the senses with quick flashes of imagery.
After I had done that, I went through all of the video footage and watched each clip, making notes if I saw something interesting or had an idea for a specific moment. The footage had a lot of clips of the lead singer and an older gentleman (who is actually Dolph Lundgren) boxing with each other. There was also some great footage of the band playing the song at various focal lengths and a couple of angles of the vocalist singing the song. Then there were a couple miscellaneous shots of him sketching in a notebook, as well as a young kid doing the same. It was clear that they were each meant to be the same person, so I decided to run with that as a through-line.
It seemed most obvious to me to make the lead singer the protagonist, as his vocals almost narrate the piece, and the footage seemed to skew that way. From there, I just began assembling the footage. The song became a guide for the pace and timing of the edits, and the rest started to fall into place. I just started with what was logical to me (the two approaching each other, challenging, and then beginning the spar). Our hero is outmatched and losing at first, but is determined and persists through the (PAIN!… I’ve listened to this song waaaaay too many times at this point).
As the song progresses, we introduce the child hero and start to lay the groundwork for the “reveal” at the end. Things start to heat up and our hero is being fueled and driven by the pain (much like a Saiyan!). He is eventually driven to a breaking point, where he goes into a frenzy and overpowers Dolph, unleashing his anger (basically goes Super Saiyan…). At this point, the speed of the cuts has progressed and becomes much faster here, as the song reaches its peak. Our hero is victorious! Now to wrap things up, we match cut to Dolph defeated on the floor, and we get our final “reveal” as we cut between the three ages of our hero.
The edit took about four sessions (maybe 4-5hrs per session) to complete, and I submitted it fairly early on. After I had finished mine, I checked out the official one and noticed it was much flashier and more VFX heavy than I was imagining. It turned out a lot of submissions were like that too.
After the competition closed, they selected 25 finalists to showcase on their YouTube channel for public voting to select a fan-favourite. I didn’t make the cut (haha… ha…) but still glad to have the opportunity to work the source material and make something cool. Most submissions selected had very “out there” effects, but there were some really interesting and creative ones very deserving of it. It’s cool to see what other people came up with and how they approached it differently. I found out afterwards that the criteria they used for rating the submissions leaned heavily on creativity/originality, so I probably wouldn’t have scored much for my standard story-focused edit, but again, it was great fun to put together so I don’t really mind.
And that’s the story behind this music video – it’s embedded above if you haven’t already watched it. I hope you enjoy.
Few games have ever provided so much replayability and fun times consistently. Today we honour those many hours spent with a highlight reel and random funny moments. Dedicated to all the Jeff Kaplan mains out there.