I’m writing this post mostly to remind myself how it’s done.

Overview: I’ve got a nice setup in Guitar Rig, including a virtual volume pedal,  and I really want to be able to control it using a MIDI controller foot pedal.

Step 1 – Add an audio track for our guitar, to capture the guitar performance.
You’ll also want to select the appropriate input channel on your audio interface.

Step 2 – Put an instance of Guitar Rig in the FX bin, and select our super-awesome patch:

Step 3 – Using the VST2 drop-down menu in the plugin header, select “Enable MIDI Input”:

Step 4 – Add a MIDI track to capture the MIDI controller data.

Step 5 – Assign the appropriate input channel for the MIDI track.

In my case, MIDI control data will be generated from my Roland A-80 controller keyboard with various pedals connected to it, connected to the Midisport 2×2 In A, so I assign the input channel thusly:

Step 6 – Assign the output channel of the MIDI track to the Guitar Rig Plugin.
This is possible because we enabled MIDI input in Step 3 above – we get an option to select the plugin as the destination for the MIDI data in the output channel list:

Step 7 – Define a controller inside the Guitar Rig plugin

Go to the Options > Controller tab in Guitar Rig, and click on “Add Controller”.

At this point, a “New Control” module appears in the Controller list in Guitar Rig.

Step 8 – Click the “Learn” button and wiggle the controller’s foot pedal:

Notes:

• Obviously the controller must be turned on!
• Assuming the foot pedal has been assigned to CC#4 (which is typical), you should see the controller number appear against the control definition in Guitar Rig.
• For diagnostic purposes, it helps to have the meters in Cakewalk displaying incoming MIDI messages.

Step 9 – Assign the controller to the Volume Pedal

In the controller list, click on the Menu button and select Volume Pedal > Pedal

At this point, moving the expression pedal should be reflected in Guitar Rig’s UI as moving the volume pedal level:

If we now arm both tracks for recording, we can record the guitar audio and the MIDI controller data at the same time, capturing the performance.

Modern digital audio workstations offer a plethora of ways to solve your mixing and routing problems. Inspired somewhat by Craig Anderton’s latest column in Sound on Sound, I discovered a new way to apply reverb selectively to multiple tracks, in Cakewalk By BandLab using the Sonitus Reverb VST. Here’s my use case:

I have Lead and Backing vocal tracks, and I want to apply a long-tail reverb to portions of the verse and chorus phrases. (The main sustained notes provide a wash of reverb in the background, but keeping it clean and un-muddled by fricatives and fast syllables.)

In the past I have created a “VOX FX” Buss, put an instance of Sonitus Reverb on it, and directed it to feed the “VOX” main buss. Finally, I add a Send on each vocal track to feed the VOX FX, and use “Send Level” automation on each track:

This works well, and allows the automation curves to be adjusted per-track.

But what if you have more than two tracks, and per-track envelopes isn’t needed? Could there be a way to send audio to the reverb using a single automation curve? I experimented with using an Aux track, intending to add a send on each vocal track to the Aux track, then use the “Automated Send To FX Buss” trick described above.

However, I realized that I could simplify things by putting the reverb effect on the Aux track itself, and then automating the “VST Input Level” instead:

From the Edit Filter selector (displaying “Clips” by default) we can drill down into the Sonitus Reverb and select “Input” from the automation choices:

Now we can draw the automation envelope (only one is needed) to control the amount of audio from all tracks being processed by the reverb:

This gets the job done, and no need for a general purpose fx buss.

As I mentioned above, there are good reasons NOT to do it this way – but it is nice to have the option.

UPDATED 27 Sep 2019

It seems you can’t “freeze” an AUX track, so if you find yourself in the resource crunch and the FX bin on the Aux track is CPU-heavy, well, that’s another good reason not to use this technique.