I've got lots of images that I would like to turn 3D via the awesome use of MIR. So far the 3D-ifying is no problem, the problem comes in how complex it is to duplicate a single instance of an image. Normally I would have to create 3 new layers for every instance (the mir layer, the texture layer, and the geometry layer) and then re-link them in MIR.
But I thought it would be much easier if I simply nested one long composition containing all the sequential images, one separate long comp containing all the bump maps, and a single instance of MIR that 3D'd them all as they appeared in their respective nested comps. So far of course this works.
But the whole idea was that I wanted to then nest this entire "3D'd sequential images" into another comp so I could duplicate it and simply use time remapping on each duplicate to select the image it was showing. Again this works but only because the nested comp is essentially being treated as a movie. What I'm looking for is a way to have the nested comps also respond to the lights and cameras of the comp they are sitting in.
I thought I was clever when I expression linked a light and camera in the "3D'd sequential images" to the light and camera in the outer comp...until I tried time re-mapping the nested "3D'd sequential images" comp's and saw that the time remapping takes precedence over the expression linking to the items in the outer comp. Meaning even though the camera in the outer comp hasn't finished zooming in, if I time remap the nested comp's to be further in time, they not only show the next image, but also have finished their zoom already.