Noumenon is a drone synth. It’s kind of inspired by the drones I’m getting out of the Lyra-8 when its oscillators are held open. That said, I didn’t really try to recreate the Lyra… more the feel of the Lyra drones: dense, distorted, lots of feedback and FM and distortion.
So, there are four oscillator voices, formed from sine waves fed back into themselves. The output of each oscillator is fed into a VCA, which is modulated by an LFO. The output of the VCA is then fed back into the voice, to produce FM feedback.
You can turn the oscillators on and off by pressing the buttons on the top left side, which also display the level of a given oscillator at a time. I should note that the voice won’t mute until after the LFO finishes its present cycle, so that they don’t cut off abruptly. Likewise, when you turn on a voice, it will retrigger the LFO that controls its output, so that it won’t jump back in mid-cycle.
The voices have a LEVEL control, which sets the FM intensity as well as the overall volume of the voice. Each LFO has a RATE control that goes from very slow to just at the edge of audio rate. Each LFO also feeds into the subsequent LFO, so there’s always a little wiggle in how the voices are modulated. (Let’s call it organismic.)
The frequency of the oscillators can be sent into sub-audio territory — when you increase the FM intensity at thes frequencies, this can create rhythmic, sort of metallic sounds.
You can change the LFOs from unipolar to bipolar using the BIPOLAR LFO button — this inserts silences between each cycle. You can adjust the SHAPE of the (sine wave) LFOs, pushing the slope one direction or another. As you adjust the shape, the LFOs will become more saw or ramp-like and then square-like with very brief intervals between the cycles. There’s also a BLEED control, which is really an offset for the LFO level; positive amounts will push the LFO range up (in unipolar mode, this means some audio will always “bleed” through, regardless of the LFO’s output; in unipolar mode it can change the ratio of time that the voice is silent); negative amounts will introduce and prolong silence between each LFO cycle.
The voices themselves are pretty simple, but they are really just grist for the mill that is the rest of the signal path. They are sent in a few different places, then, and the amount of each path is mixed in a mixer in the top right corner.
DIRECT LEVEL controls the level of audio coming directly from the voices.
BUFFER LEVEL controls the level of audio coming from a delay line buffer. The BUFFER TIME control determines the length of this buffer — up to 16 seconds. (This was originally the only path, but it can be a little disconcerting to make “blind” adjustments — as in not being able to hear what you’ve done until the output of the buffer, several seconds later, plays.) The buffer is intentionally feeding back at above unity gain to produce distortion and saturation in the buffer regeneration loop. There’s a REGENERATION amount, but I mostly park this at 1. If you set it below ~.7? the buffer can act as a fairly regular delay. But again, I tend to keep it at 1. Instead, I control the feedback with the REGENERATION SUPPRESSION amount, which uses an envelope follower to duck the regeneration, based on the output of the voices. This allows you to blend together old material and new in some surprising (and very dense/thick) ways. Fun fun.
RING LEVEL controls the level of the two sources above, ring modulated against one another. Sort of a klangorous meeting of two worlds.
In parallel to these is a CLIPPER SEND which adds distortion and saturation and picks up on weird harmonics. It’s really just slamming the signal into a couple of maxed out VCAs in series. I originally balanced the signal with buffer delays, but I found that the phasing introduced by the unbalanced parallel signal path added a quality to the sound that I liked, a bit of hollow, metallic resonance (which can be smoothed out with the filter below). So, I did it “wrong” for the right reasons. I repeat myself: fun fun.
At this point, the signal paths converge again and proceed into a compressor, and the COMPRESSOR THRESHOLD sort of sets the level for the patch. But it also affects the sense of “density” in the patch, because the compressor is set to very heavy (1:inf) ratio, so it acts as a limiter. Along with controlling the volume, this also allows you to control the headroom, which can help make the output of the patch sound even denser and more smeared together.
The output of the compressor goes into a multi-filter low-pass filter, which which you can set the FILTER FREQEUNCY of so that it sort of acts as a tone control, but as you increase the Q it can also be more of a sound source, adding a resonant peak to the output.
After the filter, the signal goes into a mono-in/stereo-out tape delay module. You can add even more DELAY FEEDBACK using its feedback circuit, and create different feelings of density by changing its DELAY TIME. As the DELAY MIX is increased, the output has more stereo width, too, which can be fun to play around with.
Finally… there’s a TOTAL FEEDBACK control, which feeds the output of the entire patch back into the FM inputs of the oscillators. Good. Times. Things can get pretty unruly as this happens, but in a good way (this is the number one reason there’s that limiter in the signal path, although there are also numbers two through nineteen for its existence, as well).
That’s it. I think it’s fun and sounds good.
Oh, in case you were wondering: “Noumenon” is a term in Kantian analytic philosophy that refers to the “the thing-in-itself” (ding an sich). This is the sort of real object that we can only access through our subjective, phenomonological experiences, and because of that, it is forever at a remove from our consciousness, which only has access to its representation. This sort of places the noumenon beyond our faculties of reason, which is the closest I can come to a connection to this patch, aesthetically. But it’s also a word that rattles through my head now and again (I spent a lot of time reading Kant as a student… because I was punk), and I put it in as a placeholder while developing the patch and then just decided to keep it.
Another performance of the patch:
https://youtu.be/UCHRRHCMP9Q