Not the most impressive thing in the world, but anyway. I’d had to play with a lot more settings than I expected on this. It uses the step distribution, which then gives you the option to set spacing between the dominoes.
So I had this idea that the CS tools Mo-cam rig might be a great basis for some camera projection as it’s effectively a series of cameras that guide the path of one ‘master camera’, as shown here:
So I got a bunch of random images from Google, and projected them randomly from each ‘Waypoint’ in the Mo-cam, onto a Cloner Object filled with cubes. One problem I did have (then eventually stopped caring about as it was Friday) was the projection on the cubes once the dynamics kick in. It would have been way nicer to work out a way to fix the projected textures to the UVW of each cube, but it wasn’t happening for me inside the cloner object… any ideas?