dealing with interface complexity - encapsulation?

Hi, how do you deal with interface complexity in generative systems in particular? For example, say I am using Blocks - Toybox in particular - and want to construct a system with 6 piano rolls all running with their own probabilistic automation, AB automation plus internal CC automation, their own individual speeds etc. Then I want to be able to feed each piano roll into its own instrument. And record individual tracks or a stereo mix
I can make that as some monolithic monster that would not fit on my screen but wondered how people with more experience than me deal with large scale complex generative systems so that components become encapsulated and only what is required at the time is exposed to the user
thanks for sharing your methods
gregh

 

I don't think there is an easy way to do this in Racks, but one way could be if you open Blocks in an Ensemble, you can turn on 'Show AB Buttons' in the Instrument properties which allows you to minimise them to just the header.

 

Check out Reaktor's stacked macros feature (the manual describes how to use them). You can drop the contents of entire blocks into a stacked macro and then flip between which one is displayed.

 

great - thanks for both those suggestions - gives me a good start

 

It's good to think carefully about the hierarchy and how to do more with less - other nice way is to run Reaktor on a high-res display so you have plenty of room to breathe

 

I try to limit the number of modules I am using - there's no way I would be using six simultaneous toybox piano rolls in one generative patch.
Lots of carefully tweaked feedback, re-using modulation in multiple places (including 'modulation' from sequencers). Quantizers, clock dividers, shift registers...
There's a really good thread on Muffwigglers dedicated to generative patches, and many are relatively concise patches with amazing complex evolving results.
Of my personal generative experiments, the ones I was happiest with used no sequencers. Although one I liked did use two simple 8-step sequencers, running at very slightly different speeds.
Most of my generative projects have been in Eurorack, not Reaktor, but the concepts are mostly transferable I think.

 

thanks colB and ehdyn - not sure i can pony up for another display I get the idea of simple systems developing complex results and that is a compositional form I have used for a very long time. So that is a good recommendation - I will find the muffwiggler thread and check out some of the patches. I am at the early stage of seeing what Reaktor is useful for and what is more suited to my DAW ie finding out ways to think that are native to Reaktor and not DAW-centric

 

The thread in question is this one, There's lots of interesting ideas, a few audio and video examples, but maybe not quite so many patch diagrams - more like explanations of techniques. Good stuff though.

 

Too late! I have already followed your advice and found that thread, read some posts and bookmarked it as well - once again thanks for your time and your very useful help. I am already starting to think more Reaktor native, with a problem to work through on my test rack. I am working like a small child with their Blocks - building things up then smashing them down, learning each time and not being too precious about what I make

 

Another thing not mentioned yet is the idea of panelsets. You can make a number of them and switch between them. Some can save mixed A and B views. Some can hide instruments.

There is a Factory ensemble called Photone that shows how to hide a lot of this.

Here is what it looks like on full display:

 

excellent, thanks