I’m using the “Image-Switch” node within my node hierarchy and I realized that it seems to handle the input/output as one baked image in “Bitmap” mode and not as “Pass Through” which makes me lose the possibility to nudge elements backwards and forewards.
I would love the “Image-Switch” node have a “Pass Through” mode allowing it to be used within a rig-structure rather than only to be used at the final bottom level where effects are applied globally on precomposed areas handled as one single “Bitmap”.
I think that the solution can only come via an updated version of the “Image-Switch” node or another new node that includes the possibility to set its mode to “PassThrough”.
I am just here to confirm that the Image Switch indeed ALWAYS converts the output to one pass, and composites all content.
So you will indeed loose the PAss Through content.
If you want to use a similar principle while maintaining the Pass through effect, you can either:
Use a peg and scale either x or y to 0 to have the element ‘‘squish on itself’’ and disappear. Turn it back to 1 scale when you need it.
Use a transparency node and have it lower the opacity to 0, of whatever you want to hide at any given time. Turn it back to 100 when you want to see it.
There are advantages and disadvantages to each options. Try them out and see what works best for you !
I prefer the peg option, but some prefer the transparency node.
Alvejado, can this help you out for what you want to do?
I still would find it handy to have an animatable “Gate” node allowing to switch between input-ports, that works like the “Image-Switch” but with the option to choose between different composite modes.
P.S.
The Image-Switch node is definitively miss-understood by riggers.
I know riggers who use that node in production to switch between different types of limb structures without being aware of the the fact that it merges the z-position values.