I’m in the middle of editing a video that combines an artist interview with event footage. The supplied raw material is 10 minutes of interview footage and 45 minutes of the event, shot from a single camera. From that footage I was able to extract 5 minutes of usable interview, but just 40 seconds of the gig. It’s not that the gig video was badly shot, it was just homogenous. Medium-long shot of people dancing. Medium shot of the DJ. Over the shoulder shot of the DJ. Medium shot of girls dancing. Repeat.
This is sad, because a single camera and half an hour is plenty of time to capture a dynamic performance. The secret sauce? Close ups.
For one of the final Game On performances, Melbourne’s “spiky electro pop” group Flamingo Crash came along and rocked out in front of a bunch of security cameras, and alongside their visual collaborator Simulcast with his Tagtool.
They enjoyed the live footage so much that they commissioned me to edit it as an “offical” music video for their album.
I’m seeing this kind of visualist- and web-savvy behavior coming from bands with increasing regularity. Artists (and audiences) are starting to realise that a music video doesn’t need to be blinged out and post-produced into sterility to be entertaining and valid. Outfits such as La Blogotheque’s “Take Away Shows” and $99 music videos are showcasing consistently high quality releases of low-budget, high-speed concepts. As a rule, visualists - those who can perform, produce, hack, and create new concepts and looks quickly - are going to do well in this environment. In the financially exciting world we have right now. Focussing on “fast and effective” also means that you’re able to release more work, which in turn attracts more people to your work, which allows you to release more work… Personally, I’ve more offers so far in 2009 than in all of 2008, so I don’t see this process reversing itself any time soon.
For further proof that you can make footage in Blender, here’s an example whipped up by Troy James Sobotka. Troy’s approach is one familiar to a lot of us: grab the simplest camera possible, go shoot something, go make something. I think it’s part of what I find appealing about the world of live visualists - exploration is encouraged. The tools in this case:
A Kodak Zi6 camera - cost: US$160. (I’m impressed; sure, it’s broad daylight which is ideal for cheap cameras - but it still looks better than what I’ve seen from the Flip.)
Blender for editing, effects
ffmpeg for export (no capture necessary — thanks, flash memory camera!)
Two hours shooting, four hours editing. (Now, if they did it on the day they had a gig, then they’d be a VJ.
I’m not saying you wouldn’t still prefer your fancy pro HD cam and Final Cut, but that’s not the point - the point is, you can make these tools work if you like. And, hey, if I had to choose, I’d save my money there and go buy my favorite VJ software package / more projectors and gear. More details:
Life is short. You find yourself having to absorb the work techniques of a lot of different software. And some of those divisions — between vector and pixels, 3D and 2D, motion and stills — look increasingly old-fashioned. Since the early 90s, we’ve seen a succession of software try to bridge those gaps. But for the first time, there’s an open-source entrant that promises to bring just about everything involving 3D and motion, minus audio, into a single tool. That means the ability to run on any OS, and a greater sense of a community that can hack the app itself.
Like to cut up and sample video? Sick of all that time-consuming scrubbing, slicing and rearranging in Vegas or Premiere? Well I’ve figured out a workflow using a collection of After Effects scripts which turns lots of tedious editing into a very quick process to output a series of video clips for your VJing pleasure.
My first test run for this technique was on The Arctic Monkeys - Brainstorm:
As you can see, this clip has an extremely choppy edit. My plan was to remove all of the band shots and leave just the dancing girls and UVA’s LED wall animations. With cuts coming at a rate of several per second, this would have been extremely time-consuming using a traditional, manual editing workflow. With the following technique I managed to cut out everything I didn’t want and reassemble the remains into a series of performance-friendly video clips in about 15 minutes.
The Workflow
Install your scripts: ReverseSelectedLayerOrder and Precomp to Layer Duration go to the folder “After Effects/Support Files/Scripts/”. Magnum, RenderLayers and ScriptLauncher go to “After Effects/Support Files/Scripts/ScriptUI Panels” (you may have to make this folder manually if it’s not already there). Restart After Effects, and choose Menu: Window > rd_ScriptLauncher.jsx to get the handy Script Launcher panel up, and Menu: Window > Magnum-The Edit Detector.jsx to get your Magnum panel.
Now run Magnum - TheEditDetector on your source clip (choose sensitivity, click the layer, then click “DO IT!”). This took about a minute to run through Brainstorm, and at the default sensitivity level it found 529 edits! You may have to experiment with this level depending on your source, and of course it’s designed to find cuts, not fades, so this won’t work for all videos.
After Magnum is finished, you’ll have a series of layers. Select all of them, then use the ReverseSelectedLayerOrder script to, yes, reverse the layer order. This will be useful later when we come to render individual layers.
8
