Silent Way and Sonar redux

[deastman]

I tested Silent Way with Sonar when it was first released. While I was able to do a lot with it, I was disappointed that Sonar was not flexible enough to enable calibrated polyphonic use of the Voicecontroller plugin. Flash forward to present day, and I have just purchased a copy of Silent Way and an ES-3 interface, knowing full well that the problem remains. I decided to explore and document the way Voicecontroller works with Sonar, and propose a potential solution. Here it is...

Voicecontroller can be used in Sonar as either an instrument or an audio effect.

When you add it as an instrument, Sonar creates a MIDI track which is routed to control the VSTi. The output of the VSTi (configured as mono x6) is then routed to 6 instrument tracks. The output of each instrument track can manually be assigned to a specific output on the audio interface. With this configuration, there is no way to route an input of the audio interface into Voicecontroller. Therefore, the calibration procedure cannot be run.

When you add Voicecontroller as an effect on an audio track, it cannot receive MIDI data, and is therefore of limited usefulness. Alternatively, you can add Voicecontroller as a VSTi into an effects bin of an audio track. Configured this way, it becomes possible to route the output of a MIDI track to the input of Voicecontroller, even though it is an effect on an audio track. The output of Voicecontroller is routed to the audio track's main output, which can either be a bus or a physical output on the audio interface. Because an audio track can only have one output (either mono or stereo), it is not possible to route 6 mono or 3 stereo outputs from Voicecontroller to physical outputs on the audio interface. Only the first two Voicecontroller outputs will be used. Because an audio track does receive input from an audio interface, it is possible to calibrate Voicecontroller when configured this way. However, only the first two outputs are available for calibration.

One possible workaround is to use the audio track configuration described above to create and save a calibration file. Then you can delete the audio track configuration, switch to an instrument track configuration, and load in the calibration file. The problem with this scheme is that only two outputs can be calibrated at a time. With this approach, using Voicecontroller in a polyphonic mode to control a modular synthesizer becomes impractical at best.

Certain other developers have built their own workarounds for this problem. The audio sampling capabilities of FXpansion's Geist comes to mind. Geist is instantiated as one or more VSTi instrument tracks, with their outputs routed as necessary. A separate audio effect, "Spitter", is added to the effects bin of an independent audio track. The input of the audio track is relayed through Spitter directly into Geist, where it becomes available as an input to sample from. If Silent Way included an audio effect plugin capable of transmitting directly into the calibration input of the Voicecontroller instrument, that would completely solve the Sonar signal routing problem.

[os]

Thanks for the detailed post.

[deastman]