I got it working!!! Finally I’ve managed to record audio along with the video. It’s been a four-hour coding extravaganza this evening, getting a recording module working using jit.vcr and sorting out a raft of other problems, but I think the patch is pretty much complete. So here’s the first improv out of the final form. It’s basically still just a test, but to me it feels like it’s gained more form than it had previously. In particular, I think the changes I made this evening to tie the volume control to the inverse of the crossfade setting so that when the river footage is showing you can’t hear it, and to smooth out the value changes for the volume control and also the automation of zoom, tolerance and fade, have made the piece feel more… musical. To me at least. I feel I’m on more solid ground now, working with a structure that feels more developed than before which was super-rough and rather hit-and-miss. Obviously there’s still work to be done – in particular (be warned!) the volume comes in very suddenly at max level as soon as I start to mix in the river footage at the beginning (forgot about the inverse relationship between the crossfade and volume. oops), but overall, I think I’m liking it and that this is the version of the patch I’m sticking with.