I just recently finished the video for I See You (I am Empty) and the making of the video went very smoothly, for the most part, hence I have very little to write about the experience. The most obvious problem I had making this video, which was obvious long before I even started, was that the main source was analog audio without a time code. This was exacerbated by the fact the the video was produced as a videoSong, meaning, the actual recording of the song had to be incorporated into the video. I didn't stand a chance of getting true sync for this footage. The sync is okay though, not actually too bad. I had to incorporate some cuts that weren't part of my original plan, because some of the imagery started to creep out of sync with the audio. By making a cut, I could re-align the errant images.
Another thing that was not part of my original plan was to make visual references to the lyrics (like when I say "Yugo" you see a Yugo on the screen). This came about because, after I started cutting away from the performances, I thought the video got a little dull. My original idea was that you would see each of the four tracks performed in their entirety, to really see how the song was made. Once I started cutting, I became selective and started cutting away more, to "justify" the necessary cuts and make them look more intentional.
Most of the added imagery was pretty much predictable and workman like, with one exception, the Space Shuttle Challenger. It was challenging, for sure.
You can actually download footage of the tragedy on NASA's website. I watched it a few times and considered including it. But it seems so wrong to me. I know I included a reference in the song, and that felt genuine, I mean, I remember it, it seems valid to talk about my experience of it. In contrast, it would be weird if I wrote a song about experiencing the Kennedy Assassination, since I wasn't yet born, unless I was working in some historical fiction context.
In referencing it, I'm still telling my story. But if I was to use the footage, suddenly, I'm no longer telling my story, I'm telling an American story about people who are not with us anymore. It stops being about the witness. And it seems wrong to do that in a song about a guy dating a robot. And if it's not wrong, it's very distracting, overshadowing the fictional drama I'm writing about with events that actually happened.
Since many of my visual cues are still images, I decided to go that route and the image of the shuttle I chose is not even the Challenger, but the Space Shuttle Atlantis. And rather than give it an abstract "flight" path like most of the other sprites in the video, I gave the still image of the shuttle a realistic path. However, at the top of the screen, when the lyrics describe the explosion, the image of the shuttle just fades away and disappears and it's so much more effective than the real footage would have been. IMO
They say necessity is the mother of invention, and I explained why I couldn't use the actual footage. But as a result of not using the real thing, I came up with something even better, for my purposes. The challenge of the Challenger, resolved.
One of my friends recently asked how long it took to make my Teenage Dream VideoSong. It's so easy to answer, but it opens a whole panoply of memories for how it all transpired. I decided to record Teenage Dream because I'd been working on a video for Dancing with Myself for a year . . . and I was frustrated.
I started work on the video of Dancing with Myself in the Spring of 2010. One day on my lunch hour, coworker and fellow movie maker Stephen and I traveled to Spanish Harlem to shoot with my friend Cherie, who plays "the one who got away" in the video. The location was great, Cherie was great, the shoot went great, all was rosy.
When I was planing the shoot for Dancing with Myself, I was trying to figure out how to make an HD video without an HD camera, since I didn't have one and didn't want to borrow one for as long as it would take to shoot the video. I realized I could use still images in the beginning of the video, especially if they were still images of people being very still (naturally, sitting, etc). But I really wanted some live action too, so I came up with the idea of doing some rescan, as if one of the characters were watching the video on an old TV. Of course, this is only literally what we see.
In the Video, the TV is supposed to symbolize a kind of "box of memories" for this girl he lost. It's a technique I've used before, but this is the first time I had the character watching the TV, usually the TV is next to their head, like a thought bubble. I have to admit, the way I used it this time is a little more confusing, in terms of communicating what's going on.
Once the shoot in Spanish Harlem was finished, Stephen and I made the still shots for the opening sequence. A cardboard box stood in for the TV until the screen and dials were added later in Photoshop. Reverse shots of me watching the TV were also put in the screen as "reflections". Later still, footage of Cherie was added. Although lots of people could tell the footage was added, most mentioned the lack of a power cord as the element that clued them in, not the fact that it was merely a cardboard box on the ground.
So, I had the guy in the factory watching the TV, I had Cherie and I in Spanish Harlem acting out a little drama, but I really wanted something more, a performance element, the band, actually playing the song. But since Tribrix is essentially a one man band, the idea of a band performance was out . . . or was it.
I didn't have a green screen at the time, but I conceived of a way of using black screen to composite 3 moving images of me, to represent "the band". I wanted to mimic the lighting on the Police in their video for Every Breath You Take. If you set the shots up correctly, you can layer the three images together and create the illusion of having a whole band.
In the amount of free time and space I have, it was tough to arrange all of this, backdrop, lighting, camera setups. But I managed to shoot all three moving images of myself, playing drums, bass, and guitar. It had taken a few months but it looked like I'd be finished soon.
When I was working on my first music video, for the song Vice, I had discovered something interesting: digital devices have made time coding almost obsolete. In the old days, if you shot sound and images at different times, you could use time code to sync them up later. The time code would ensure that everything was chronologically stable. Without it, the audio and video tracks would drift apart in time, and people's lips would start to look like random rubber bands, almost like watching a movie that has been dubbed, except, it's not supposed to look dubbed, it's supposed to look synchronized.
My original plan for Vice was to do a flash animation, but there were so many delays getting that off the ground I had to forget it. Cheney's term as Vice President was coming to an end, and I wanted to do this video before that happened. I can't say if flash would have made the video better or not, but it wasn't an option. Instead of flash, I decided to go with a paper bag puppet for the singing and two After Effects animations for the "action" shots.
From a content perspective, it's a much more interesting choice. But I quickly learned that puppeteering is much harder than it looks. And ones hands and arms start to hurt quickly over multiple takes. Eventually I got some usable footage and started to edit it together.
I had not worried about time coding because it was a puppet, and I assumed if the lip sync was off a little, no one would notice, since it was a puppet. So you can imagine my surprise when I lined up the song and the visual puppet performance and they stayed in perfect sync for the whole song. I was shocked. It shouldn't be possible, yet it was working. I imagined Nagra was going to have a much harder time selling those $2000 tape recorders used for recording sync sound on movies.
The sound source for the puppet performance on Vice had been the cheapest little MP3 player you could buy. Not exactly "professional", sync ready, design. When it came time to record "the band" for my Dancing with Myself video, the most convenient source to lip synch to was a sony CD player I had laying around (the battery on my iPod needed replacing and the cheap MP3 player was MIA). CDs are digital, so I imagined the same rock solid sync ability as the cheap MP3 player. But I was wrong.
I don't know how, I don't know why, but the CD player drifted all over the place, chronologically speaking. It wasn't just playing the song too slow, it was playing the song with a completely inconsistent speed. Slowing down, then speeding up. I still wasn't worried though, remember, I was only working with guitar, bass and drums, so a little drift probably wouldn't be noticed. As long as I did a sync at the start of each lyrical stanza . . . or so I thought.
I spent weeks and weeks and weeks massaging this footage (which was easier than a reshoot), all to no avail. It was a mess, and not getting any better. Further more, for Christmas that year I got an HD camera and a green screen. A reshoot was in order, but I was in no mood to go back an work on it anymore. I decided I needed a change, some quick project to take my mind off the chaos of the current project.
Katy Perry's Teenage Dream was gaining traction and I really liked the song. Not just as a listener, but for some reason, I really wanted to record it. Who knows why these feelings hit us, but it hit me hard, and I wanted to do it while the song was still current.
I'd first heard the term "VideoSong" listening to an interview with the male half of Pomplamoose. I was intrigued and thought my Katty Perry cover would be perfect for doing a VideoSong. I mean, how hard could it be to record video while you recorded the song? Because, I had learned nothing from the Dancing With Myself experience.
My Teenage Dream video started off okay. I set up the camera and started performing the various tracks. It was actually going almost a quickly as a regular audio song. I figured, given the genre, I aught to work on the video simultaneous to the audio. And that's when I hit the first snag.
The footage was so, so, so boring. I mean snorzeville.
I had to start over.
When I first moved to New York, my roommate-to-be had a disco ball. We hung it up and I put a couple of lights on it. Then I added some Christmas lights, then some police flashers and a lighting controller. By the end we had gobo scanners and a huge assortment of other lights, all in our living room, all on with a single switch, computer controlled to pattern on and off with the music. It was like our own private nightclub, which was the point.
I decided to pull out the old lights and controller for my Teenage Dream video. If nothing else, at least I could spice up the video footage a little. I took a long time, but eventually I came up with a set up for the vocals. Other times, I used one of the gobo scanners alone.
Of course, setting all this up and re-recording, and making sure to wear something decent, and do something with my hair meant, recording a VideoSong takes a lot longer than recording an audio only song. In the end, my little side project to distract me from Dancing with Myself took four months. But I learned things doing it, so when I returned to finish the first video, it was completed quickly and I had the benefit of HD and green screen for creating my "band". I said I'd never do another VideoSong, but I'm already working on one for I See You (I am Empty). Look for it soon.
