Duplicating Duplicators is super useful if, for example, you want to make your own custom distribution modes. At the end, it is all about understanding Index Context.

What we are doing in the above tutorial, is using data from Duplicator B, to influence how Duplicator A is behaving. We are using Dup B Index, to change how each copy of a Duplicator A behaves – it grows exponentially. You don’t have to use this data to only create copies in Dup A. You can modify other settings, like Corner Radius on an underlying shape, as an example.

One thing to note, is that you want to avoid “red noodle of no connections”. Or feedback loop, as it is officially called by Cavalry developers. This happens, when Value A is defendant on Value B, and Value B is trying to feed back into Value A. For example, Shape Scale of Duplicator A is linked to Shape Scale of Duplicator B. You won’t be able to “link back” Shape Scale of Dup B, back to Dup A, as you would create a feedback look.

Once you stay away from “feedback loops”, then you can create more complex systems, where single cells (shapes), create molecules (Dup A), which in turn build organisms (Dup B, containing Dup A). It is all about making things more and more complex, with each level, while maintaining procedural controls.

Overall set up is simple – keep everything on 1 Null Layer. All the controls for thickness of lines, how quickly everything fades, colours, position etc. Everything on 1 layer makes it easier to quickly adjust everything from 1 place.