For the last few hours I’ve been sitting here attempting to recreate three simple transitions/effects that Wright uses across all of his work: the crash zoom, the whip pan transition, and the ‘obscured wipe’ (where someone or something moves in front of the frame, taking the scene with them and revealing a new one). I’ve been doing well, in the sense that I now understand how to do all of these things and am confident that with the right computer, I could use these effectively as a part of a scene. I’ve not been doing so well in the sense that up until now I’ve been using my laptop to learn these things, and whilst After Effects is already extremely frustrating to learn about as someone with almost no patience at the best of times, it’s even worse when it’s running too slow to provide a proper preview of what you’re editing. Add to that a very serious laptop failure a couple of hours ago (system failure, unable to reboot, panic… miraculous recovery? Constant fear of events repeating themselves), I’m calling it quits. At least until I can get into the editing suites and attempt it all there instead. For now though, just to prove that those hours of work have not been for nothing, I’ll write a brief description for each one.
Editing the crash zoom is pretty simple, and I really wish that Premiere had a motion blur effect, because that’s really all I need After Effects for. Basically, all you need to do is set a key frame for position and scale, and then second key frames on your ending frame. Then you reposition the image so that it’s zoomed in/focussed on the subject, and add motion blur. Done!
The whip pan transition actually doesn’t need After Effects, since it’s just a matter of cutting during the pan for each shot, so I was able to edit a test for this one:
So then there’s the ‘obscured wipe’. This is what killed my laptop. I’d actually finished the edit too, so I completely understand what I’m doing here, it just overworked my computer. Basically there are two ways to do this. Firstly, you can just have two shots, where in one of them the object moves across frame. You then use a mask (drawn with the pen tool) to tell the program that you want to the frame of the first shot to end where the object intercepts the frame, and frame by frame you draw the mask so that as the video moves, so does the mask, and hence the transition can work since your second video would be layered underneath. This takes forever. The other option is to have a third clip which is the shot of something moving across the frame, preferably in front of a green screen, and then rotoscoping the object so that is sits overtop of the two other clips. Then you use masking on the first clip and hide the edge under the object as it moves, and it has the effects of taking the scene with it. This is quicker, however you do have to consider that you might want to ‘fix’ the third clip, as in, the edges might be rough from rotoscoping, so you’ll have to fix that to make it blend with the scene. You might also want to blur it or darken it since its not the focus of the scene.