The 266t Source of Uncertainty is not a voice module by itself, but it is extremely useful for generating pitches, pitch movement, rhythmically changing melodies, and evolving note selection. Think of it as a melody idea generator and pitch processor.
From the manual, the 266t provides:
To make this melodic in a Eurorack patch, you typically combine the 266t with:
This is the most direct melodic section.
This gives you a random melody that can be constrained to a musical scale with the quantizer.
Stored Random Voltage is great when you want random notes, but with bias.
This is excellent for controlling: - pitch - transposition - octave choice - melody contour
Then use the distribution knob: - CCW: mostly low notes - Center: notes cluster around middle register - CW: mostly high notes
This is much more musical than pure equal random because you can choose whether the melody tends to live low, middle, or high.
Patch the left output to one oscillator and the right output to another oscillator.
You get:
- one line with fully even note choices
- one line with biased note choices
That creates natural contrast between two melodic voices.
This is the classic “random note generator” patch.
Use the internal noise outputs: - White noise: brightest, more jumpy/random feel - Pink noise: more balanced feeling - Blue noise: low-frequency biased according to the manual description, so try it for somewhat different contour behavior
Now each trigger creates a new random pitch.
This method feels more open-ended than the Quantized Random section.
It works especially well if:
- you want a quantizer to define the scale
- you want to sample other voltages besides noise
- you want to alternate notes using the alt outputs
The manual notes: - pulse alt divides incoming pulse into alternating pulse outputs - CV output alt divides voltages into alternating outputs
This is extremely musical.
Or: - pulse alt outputs trigger two separate envelopes/voices on alternating beats
You get: - alternating notes between two voices - call-and-response lines - stereo melodic ping-pong - interleaved melodies from one random source
This is one of the most interesting compositional features on the module.
This section outputs continuously moving random CV rather than stepped values.
By itself, this is usually not ideal as direct pitch CV for tonal music, unless you want gliding/experimental lines. But it becomes very useful when combined with: - a sample and hold - a quantizer - an integrator - a precision adder/transposer
Now the underlying voltage is always moving, but notes are only captured on the clock.
This creates a melody that feels organic and connected, unlike pure white-noise randomness.
This is a very musical way to create evolving melodies without losing structure.
The integrator smooths stepped voltages.
This is perfect after any stepped random source: - Quantized Random Voltages - Stored Random Voltages - Sample & Hold output
Now instead of instant note jumps, you get: - portamento - glissandi - sliding between notes
Modulate the integrator CV input so that some phrases glide more than others: - slow envelope to integrator CV - random CV to integrator CV - manual performance control
This makes the melody feel much more expressive.
A very usable patch for actual songs.
Connections - Master clock → Quantized Random pulse in - Quantized Random n+1 out → integrator in - Integrator out → scale quantizer → oscillator 1V/oct - Master clock → envelope trigger - Oscillator → filter → VCA
Result - random stepped notes - locally grouped pitch movement - optional glide from the integrator - much more “musical phrase” feel than chaotic jumps
Connections - Fluctuating Random Voltage out → Sample & Hold CV in - Clock → Sample & Hold pulse in - S&H main out → quantizer → oscillator A - S&H alt out → quantizer → oscillator B - Pulse alt outputs → envelopes for voice A and B
Result - alternating notes split across two voices - excellent for stereo patches or dialogue melodies - one source creates a surprisingly composed result
Connections - Clock → Stored Random pulse in - Right Stored Random output → quantizer → oscillator 1V/oct - Use the probability/distribution knob to bias the register - Slow LFO or another random source → Stored Random CV in
Result - melody tends to favor low, center, or high notes - feels like intentional contour rather than flat randomness - great for generative lead parts
If you already have a sequencer:
Connections - Sequencer pitch CV → precision adder input 1 - Stored Random or Fluctuating Random → quantizer/transposer → precision adder input 2 - Precision adder out → oscillator 1V/oct
Result - your base melody stays recognizable - the 266t adds octave jumps, phrase transpositions, or tonal shifts - ideal for generative variation that still sounds composed
Connections - Pink noise → Sample & Hold CV in - Clock → Sample & Hold pulse in - S&H out → quantizer → integrator → oscillator
Result - classic random notes - scale-constrained - glide softens harsh jumps - excellent for ambient, Berlin-school, and experimental melodies
If you want melodies that fit a song: - use any stepped CV output - then send it into a quantizer - optionally through the integrator
That gives the strongest results.
Random voltage often spans a wide range. To keep melodies usable: - use an attenuator - use a precision adder with fixed offset - use the Stored Random distribution control - use the n+1 output for tighter intervals
The pulse inputs are where the melody becomes musical. Try: - quarter notes for sparse lines - eighth notes for active melodies - irregular clocks for more human phrasing
For example: - Quantized Random → pitch - Stored Random → filter cutoff or decay - Pulse alt → alternating envelopes
That makes a melody feel intentional and alive.
The Quantized Random section becomes more phrase-like when the number of steps is reduced. Fewer steps often means: - more repeated notes - stronger motifs - less “aimless” randomness
The Stored Random right output is especially good for musical results because you can shape the probability of low/mid/high voltages. This is one of the best features on the module for melody writing.
The 266t is especially strong for:
It is less about writing a fixed melody and more about creating a system that produces musically convincing melodic variation.
The 266t Source of Uncertainty can be a powerful melodic tool when paired with a voice and, ideally, a quantizer.
Most effective melodic workflows:
If you want, I can also turn this into:
1. a set of concrete patch recipes for techno/ambient/Buchla-style music, or
2. a signal-flow diagram showing exactly how to patch the 266t into a complete melodic voice.