I’m not planning on going into too much detail on this unless someone asks but after having fun with feeding GR-16 MIDI into Obsidian/Slate, I decided to try and further abuse SonoBus to add hacky AUFX automation to NS2. Note that you can already control a number of AUFX apps that do MIDI learn directly in NS2, e.g. SugarBytes effects.

The basic idea is:

1. NS2 internal audio send from your dry track to an External MIDI track.
2. SonoBus from this track to AUM and insert your FX after SonoBus in AUM. Set up any necessary MIDI controls for the plugin and ensure the MIDI control input port is also selected. You could use the AUM port or AudioBus virtual port or another virtual port.
3. External MIDI is then pointed at the correct MIDI port/channel in AUM.
4. Any automation drawn/recorded in the External MIDI track will be sent to the FX and you should hear it alongside your other NS2 tracks.

I recorded the simultaneous output of NS2 and SonoBus and a waveform comparison showed less than 20ms difference. It was certainly good enough for me to plan automation for a track.

To do an accurate mixdown you can mixdown the dry track (or tracks), load it/them in to the AUM file player on a channel feeding the same FX then sync start AUM and NS2 by hosting both in AudioBus and hitting AUM record. NS2 will control the FX in real time as the file passes through and is recorded. For extra bonus points, send a stop record MIDI control from NS2 to AUM on the final bar for a perfect recording.

I need to see if the automation can be easily imported into other DAWs as it is likely to be simpler to do the mixdown as dry stems and apply the designed automation in another DAW in one pass rather than trying to get the accurate version via AUM. It should just be a matter of exporting the automation track to something like MTS, along with the stem.

Notes:

1. I originally had SonoBus sending back into NS2 but this did not offer any obvious advantage as mixdown did not include the AUM processing.
2. To bypass the SonoBus track in NS2, simply turn off the audio send power button on the dry track.
3. Using the audio send “pre” option will let you bring the volume of the dry track down to zero.
4. Turn the “monitor” down to zero in the sending SonoBus to completely mute the dry signal. You should then only hear the processed AUM signal.
5. Use the SonoBus connection history to quickly reconnect. There is an autoreconnect on startup option which doesn’t seem to work in the AU (reported in SonoBus thread) but will be useful once it is fixed.
6. I feel like the multichannel SonoBus capability should help here somewhere but I have not quite got my head around it.