Hi there,
I have made a lighting rig composed of a few hundred suspended hanging leds. They are not arranged in a grid, but randomly at different heights to look like a star-scape. For clarification, I have made a panel with 36 light mounts randomly scattered across it, then duplicated this panel 6 times. I have painstakingly measured the height of all my lights, so I now have their positions mapped in xyz space.
I have successfully set up a TOP to CHOP style mapping network, where I can run videos across the set up 2 dimensionally (viewed from above). This looks great but I would like to be able to animate across the vertical axis as well. I have what feels like a work-around solution, but it feels like it could be more elegant. I have two cameras looking at a single sop, I render them which gives me an XY view and an XZ view. I then use the same TOP to CHOP technique with a math CHOP afterwards that means that if a SOP interacts with a mapped pixel in both the XY space and the XZ space then I get a dmx output. This works but it has a few flaws. You can only map solid shapes, as although the math CHOP step means the mapping is sensitive to 3 dimensions, it is only referencing 2d videos so a surface becomes an blocker, and any changes behind the surface along that axis get lost. You can’t show a hollow sphere for example, as the system doesn’t know the inside shouldn’t be mapped. Similarly, I don’t know how to adjust transparency in interesting ways to SOPs. In the TOP environment, when you use a moving circle TOP as the content to animate the lights with, you can adjust the soften parameter to blur the edges, which looks really strong when sent to the lights. However, there is no such soften feature on a sphere SOP, and setting the alpha is global for the whole shape. Is it possible to make a sphere that is opaque in the center, and transparent at the surface? Like a circle ramp TOP but in 3 dimensions?
What is a best practice way of pixel mapping in 3 dimensions?