“Spells and hymns in Greek magical papyri refer to the goddess (called Hecate, Persephone, and Selene, among other names) as “triple-sounding, triple-headed, triple-voiced…, triple-pointed, triple-faced, triple-necked.'” (From the Wikipedia article on triple deities.) Just to put you in the mood for this patch.
Triple Goddess is a three-voice drone synth, designed for wild, otherworldly tones. Each voice is made from two sine wave oscillators — named Body and Soul, respectively — which are multiplied against one another (ring modulated). The oscillators can go from audio to sub-audio rates, allowing for a multitude of timbres and rhythms as their pulses produce sidebands and complex harmonic content.
To simplify controls, the two oscillator groups — Bodies and Souls — are ganged together. So, the “Body FM feedback” control adjusts the feedback for all three of the Body oscillators in each of the three voices simultaneously (and there is a corresponding “Soul FM feedback” control for that oscillator group). Along with FM feedback, waveshaping and frequency modulation can be added to each oscillator group separately. A noise modulation affects all oscillators simultaneously, adding a grainy, noisy flavor.
The voices — which are named Phoebe, Hekate, and Diana, after the three faces of the triple goddess from Roman antiquity — are panned across the stereo field: Hekate is in the center; Phoebe is panned left ~50%, and Diana is panned right ~50%. Then, the voices pass through a stereo low-pass filter to darken the sound and soften some of the spikier harmonic content all that FMing and waveshaping and ring modulating can produce, before a plate reverb smears everything together and brings cohesion to the sound (along with atmosphere and ambience).
The whole patch was sort of inspired by a sideways turn I took, working on a patch for my livestream, that was inspired by the work of La Monte Young with drones/sustained tones (I may/probably will take another stab at this for a Saturday morning patch). You might be able to hear the reference? But also maybe definitely not, because I did not really do a good job with representing the tonal work of Young?
Controls:
Across the top row are five buttons (and one value module).
There are buttons named “Phoebe,” “Hekate,” and “Diana.” Each of these controls the on/off for that voice. The buttons are based on a flexi-switch design, so if you hold them, they change states momentarily, but if you tap them, they latch in their new state.
There are also two buttons “Body max fdbk” and “Soul max fdbk” which will max out the FM feedback amount for that given oscillator group. These buttons are only momentary.
All of the buttons are controlled by an “Attack-release” parameter, that allows the voices (and feedback) to fade in and out. This control sets the amount of time for those fades, from 0 to 60 seconds.
Beneath each voice button is a set of controls that are unique to each voice. “Body frequency” and “Soul frequency” set the frequency of the two oscillators at the core of each voice. The oscillators can go into sub-audio rates, too, which allows them to apply rhythmic/tremolo effects to their paired oscillator. (If both oscillators are sub-audio, some weird things can happen, but they can’t always be heard.) These controls are set to show frequency/Hz, but you could set the value modules to note intervals if you wanted to tune them to specific harmonies (I like the playful exploration that comes from using Hz, personally, but whatever works, works).
There is also a dedicated level control for each voice, so you can mix them as you wish.
On the next row are sets of modifiers. As I explain above, the Body and Soul oscillators are ganged together, so these controls affect each oscillator group (and therefore affect all voices at once).
“Body” and “Soul waveshape” control the waveshape of the sine oscillators (also called their duty cycle). In the middle (0), they are the most mellow. As you move the control toward either end (-1 or 1), they will begin to pick up more harmonic complexity.
“Body” and “Soul FM feedback” control the amount of FM feedback each oscillator in that group has. A little will produce a richer, fuller sound, while a lot will produce something approaching noise.
“Body” and “Soul” mod rate and depth control the modulation affecting the frequencies of the oscillators in that group. The modulation is based on multiplied triangle wave LFOs (the LFOs also affect one another’s rate), and at slow speeds and depths, it is meant to mimic (to a degree) the sound of a chewed up tape — at more intense settings, the modulation becomes… well, more intense (shocker). The two oscillator groups’ modulation sources are ever so slightly different.
Across the bottom row are sort of global effects.
“Noise mod” affects all oscillators. It sends digital noise (unclocked random module) to all of the oscillators, introducing noise, distortion, and flutter.
“LPF freq” sets the frequency for the low-pass filter. It can be used to soften some of the complex (high frequency) harmonics produced by all of these interactions between the oscillators.
“Reverb decay” and “reverb mix” control the decay and mix of the plate reverb. The EQ has been slightly adjusted to cut low-end (there is lots and lots of low-end in this patch).
That’s all the controls, but all of the controls are really interactive and often unexpected outcomes arise. I would definitely recommend just sort of messing around with stuff — there’s nothing that can go wrong, unless you think drones are wrong to begin with, and then, well… probably not the patch for you.
