As promised, here is my patch featuring 81 AirWindows, and nothing else! (Except for the requisite Audio module, plus some light dividers and documenting text dividers). Note - this patch should be run at 48 kHz sample rate.
And here are a couple videos using the patch
Most of the sounds are generated by amp emulators with feedback, and no external inputs - the most elementary form of no input mixing. The Tone controls work well to control the frequency. In particular, the LilAmp works exceptionally well, giving a lovely lead tone that covers nearly 3 octaves. I tuned everything by ear. I didn’t use this feature, but the LilAmp can be tuned to 1 volt per octave for almost 1 octave.
In one voice I implemented a form of FM, with three different self oscillating amps functioning as FM operators. Op 1 modulates the tone of Op 2, which in turn modulates the tone of Op 3 to produce the final signal. I also mix in some varying CV with the Op 1 output to generate different tones at the end of the FM chain.
The Cabs amp can be swept through different tones by changing the model via CV, providing a sort of high pitched arpeggiator, but without much control of the pitch.
The self oscillating BassAmp generates a really interesting LFO signal that I modulate into envelopes to ping a resonant filter, which creates the driving rhythmic bass line.
The Mackity distortion (modeling the Mackie mixer amps) also self oscillates very easily with some feedback, just like the real deal. I use that for a two toned bass drone.
I am really happy with the LFOs that are created using Constant +5 and/or -5 voltage (DCVoltage) feeding the inputs of AutoPan. It has a wide range of frequencies, from very slow, to just below audio rate. They are really well behaved, though the frequency intentionally has a nice subtle drift to it.
There is no way I could create this pure AirWindows patch without copious use of Golem as a mixer. It can only mix two signals into one - so it takes many instances. The PurestGain works great as an amplifier or attenuator, or VCA. I only used the VCA capability paired with a complex LFO in one place to create an undulating voice.
Nearly half the patch modules consisting of nothing but PurestGain and Golem are used to create my 4 stereo mixer with 2 send/return effects loops.
One module I did not use, but I think has a lot of potential, is the LRFlipTimer that swaps left and right channels at regular intervals, ranging from 1 to 10 minutes. It could be used to alternate between two totally different sounds. I only wish it could be set to less than 1 minute - say maybe 15 seconds.
I really enjoyed this challenge! Thanks @ablaut !