The “Bandwidth Hydra Send” proposal (terrible name, come up with something better?) is actually pretty neat.
Instead of slider bars, I would prefer a different interface (like in the AU Multiband Compressor image above, where I can drag 4 pointers/lines to cover a range of the spectrum). The ability route parts of the signal to different send tracks is rad.
yeah… I was thinking the same. But like Signal Follower is more flexible than sidechain compression, this will be more flexible than multiband compression. Still keeping the flexibility in mind I think there should be a switch or option in this device whether to link previous high pass with next receivers low pass or not.
Edit.
On the other hand. If the purpose will be multiband compression for most of the people, it’s going to be a pain in the a’'e to set it up every time. Especially when you want to have mbc on master you will need to route everything to one send-track and create a send-track for each band which will be routed to master after that…
I’d call it “#crossover send” device. And in practice it should work so when moving the crossover points and all the send channels are dry and same level, you shouldn’t hear any difference in the sound. Doing seamless crossover with current renoise filter devices is bit impossible unfortunately.
Sorry if I hijack your topic but in theory you can have this mutltiple-something send device for a lot of FX.
To make this more modular (and this reusable), I think it is time to have the dsp chains in a graph format.
For the mutiple band-filter send device, typically the graph will look like:
The split-device will simply replace the send device. The add(er) device will be decided how much percentage (weight) of each input is mixed to the output. Mutliple inputs can be created on the adder device.
Now you only need to route tracks to the signal input of the graph.
It is the simplest but it is also the less modular. Also the team have to make some assumptions about the hardcoded filer settings (MAybe you want to split on HP and LP, Q-factor?).