This is my test scene and its the result i expected.
Its just one shader using 2 UVs
This is my shader network. Just a checker and picture piped into the uv_transform nodes
and layered over each other. So far so good.
The ugly part is how i produced the cutout or that i had to produce a cutout. Actually i hoped
i can just clamp the UV repetition and can layer them over each other. But that didn't work.
So i used the "color" mode in the uv_transform node and picked a color wich doesn't exist in the picture (in my case white) so i could manipulate it (inverse and then desaturate and push the gamma) to get a black and white cutout for the mix slot.
How can i do that more efficient / smart?
Of course in this case i could just use one UV map but its just a test for more complex models where the aim is to have a dirt shader painted in substance running over multiple objects with different native arnold materials underneath. Like a dirt splash over multiple objects. Or is my whole approach wrong?
Here the Scene File
thanks in advance !!
A slightly better option to produce the mask needed for the Mix is to duplicate the uv_transform of the image, but with a white input color. The problem is of course that you have to maintain both uv transform nodes.