Weird bus behavior in Supercollider?

anteater

Hello,

I am using the Bela Mini with Audio Expander with Supercollider. I am allocating some software busses for some internal routing, but finding that the index of the busses isn't what I expect. I am doing the following:

s = Server.default;

s.options.numAnalogInChannels = 8;
s.options.numAnalogOutChannels = 8;
s.options.numDigitalChannels = 16;
s.options.numInputBusChannels = 8;
s.options.numOutputBusChannels = 8;

s.options.blockSize = 512;

// todo, have to make fake busses for some reason? 

~mockBusses = 4.do({Bus.audio(s,1)});

~analysisBus1 = Bus.audio(s,1);
~analysisBus2 = Bus.audio(s,1);
~analysisBus3 = Bus.audio(s,1);
~analysisBus4 = Bus.audio(s,1);

~analysisBus1.index.postln;

But finding that the ~analysisBus1 index was being assigned to "4" which is still within the realm of the hardware outputs, thus outputting to the speakers. I had to as you can see assign some "fake" busses to push it past these hardware outputs. Is this expected behavior?

giuliomoro

I don't know anything about supercollider, but the documentation says that Bus.audio should "Allocate a private audio bus on the server", so I don't see why that should ever return an audio bus that's one of the input or output busses. I don't know why it starts at 4, but have you tried sending a signal to the first bus that's created (the one that has index 4)? Does that come out of output 4?

anteater

giuliomoro It does, which was bad because I was hearing that out the actual hardware out (where my headphones were connected) instead of just as internal software routing.

This "dummy bus" does fix it, but does seem a bit unusual - not sure if it's a bug somewhere, or if I'm doing something wrong. Again, not a big deal, but just confirming I'm not doing something unusual.

giuliomoro

Can you post the full patch as you run it on Bela?