For example, should I put the logic for movement, needs and e.t.c of a character in one Flow Macro or divide everything into multiple?
Are you asking specifically for your movement management? Or all elements?
There are instances where I put all control in one macro tied to one object - but then when it gets too crazy, start splitting the reusable sections of code off into individual macros and call them as super units. So it is kind of the same thing as having it all in ‘one large macro’, but kind of with smaller macros inside of the main macro - just to help keep it clean and neat and readable.
Just my newbie perspective.
I have the same. One large macro with lots of super units. It’s very easy and readable. But my main concern is the performance. People say that it’s better to have lots of C# scripts then use 1 for eveything. So I thought this logic applies to Bolt, too.
Every layer you go deeper with superunits adds a bit of initialization cost. So don’t go too nuts with superunits 30 layers deep. That said, the impact of embedding superunits into other graphs is pretty negligible in the grand scheme of things. It’s a micro optimization. And multiple FlowGraph/ScriptGraph workflow on the same gameobject is not well supported right now. It works, but viewing the right graph when the object is selected or in Play mode often has to be done by manually pressing the Edit Graph button in the Inspector. A lot more clicking and buzzy work that way. So I do one graph per GameObject.