yarns-loom

Introduction

Loom is an alternate firmware for the Yarns synthesizer module ↗. This firmware is aimed at making Yarns more powerful and user-friendly.

This manual explains how Loom is different from the original firmware for Yarns. For documentation of the original firmware, see the Yarns manual ↗ and Yarns firmware changelog ↗.

Contents

• Panel interface
  • Menus and commands
  • Panel controls
  • Display and LEDs
• MIDI controller input
  • Control Change messages
  • Play mode
  • New layouts
  • Note processing
  • Hold function
• Clocking
  • Clock settings
  • Master clock controls
  • Cueing synced events from master song position
• Sequencer
  • Sequencer controls
  • Sequencer input response
  • Step sequencer
  • Loop sequencer
• Arpeggiator
  • Arpeggiator basics
  • Arpeggiator rhythm settings
  • Sequencer-programmed arpeggiator
• Note voicing
  • Polyphonic voice allocation
  • Legato and portamento
• Voice modulation
  • Envelope
  • Low-frequency oscillator
• Voice oscillator
  • Oscillator audio mode
  • Oscillator timbre
  • Oscillator shape

Panel interface

Menus and commands

Submenus for settings

1. ▽S (SETUP MENU): configuration, MIDI input/output
2. ▽O (OSCILLATOR MENU): audio mode and timbre for the voice oscillator
3. ▽A (AMPLITUDE MENU): voice envelope and tremolo LFO

Part swap command

• *P PART SWAP SETTINGS in main menu
• The selected part (selected by turning encoder) swaps its settings with the active part
• Allows storing an alternate configuration in an inactive part

Save/load commands

• Display blinks command name when picking a preset to save/load
• Display splashes the result after executing a save/load
• Hold encoder to exit preset selection

Panel controls

Active part control

• Replaces the PART setting from the original firmware
• Display blinks the active part number and its play mode
• Hold TAP button to switch the active part
• The active part is used for part-specific functions of the panel interface:
  • Editing part settings
  • Recording a sequence
  • Activating the hold function
  • Swapping part settings
• Only used in multi-part layouts

Tap tempo

• If a single tap is received without follow-up, the tempo is set to use EXTERNAL clocking
• After setting tap tempo, display splashes the result

Display and LEDs

Full display of integer setting values

When display shows an integer setting value that has three characters, blink a prefix character over the left displayed digit:

• Three-digit unsigned integer 127 blinks 1 over 27
• Two-digit signed integer -42 blinks - over 42
• Two-digit labeled integer T23 blinks T over 23

Other changes to display and LEDs

• Display has 64 brightness levels (4 in original firmware)
• Channel LEDs have 64 brightness levels (16 in original firmware)
• When displaying sequencer note pitch, display prints flat notes as a lower-case letter next to the octave number
  • Original firmware behavior: display prints flat notes as an upper-case letter next to b, with no octave number
• Improved clock-sync of display fade for the TE (TEMPO) setting

MIDI controller input

Control Change messages

CC mode

New global setting CC (CONTROL CHANGE MODE) sets how a CC’s value is interpreted:

1. OFF: CCs are ignored
2. ABSOLUTE: the CC’s value becomes the new setting value
   • CC’s value is down-scaled to match the setting’s range
   • For use with traditional potentiometers
   • Original firmware behavior
3. RD RELATIVE DIRECT: the CC’s value is added to (or subtracted from) the setting’s current value
   • For use with endless encoders
   • Uses the “twos complement” standard for translating the CC’s value into an increment
     • MIDI value 1 => setting + 1 (increment)
     • MIDI value 127 => setting - 1 (decrement)
   • Settings will increase or decrease by one value for each click of the encoder
   • Depending on how many values the setting has, the encoder may take anywhere from 2 to 128 clicks to scan the range of setting values
   • Send 0 to display current value without change
4. RS RELATIVE SCALED: the CC’s value is added to (or subtracted from) the value of a “virtual potentiometer”
   • Similar to RELATIVE DIRECT, but always takes 128 encoder clicks to scan the range of setting values, no matter how many setting values there are
   • Gives all encoders the same travel distance from minimum to maximum

Added CC types

• CC support for all new settings: see Loom CC Implementation Chart ↗
• Recording controls
  • Start recording
  • Stop recording
  • Erase recorded sequence

Macro CCs that control combinations of settings

• Recording state and erase: off, on, triggered erase, immediate erase
• Sequencer mode and play mode: step sequencer, step arpeggiator, manual, loop arpeggiator, loop sequencer

Other CC improvements

• When CC is received, display splashes the result (value, setting abbreviation, and receiving part)
• Bug fix: bipolar settings can receive a negative value via CC

Play mode

How play mode works

• Part setting PM (PLAY MODE) sets how each part generates MIDI and CV/gate output
• This explicit setting replaces the implicit switching in the original firmware (sequencer enabled if recorded sequence exists; arpeggiator enabled if ARP RANGE > 0)
• Allows changing mode without altering the recorded sequence or arpeggiator range

Play mode options

1. MANUAL: output is generated by MIDI input only
2. SEQUENCER: output is generated by the sequencer
   • Sequencer output can be augmented by MIDI input, depending on the sequencer input response
3. ARPEGGIATOR: output is generated by the arpeggiator
   • Rhythm is controlled by the arpeggiator pattern setting

New layouts

• 2+2 3-part layout: 2-voice polyphonic part + two monophonic parts
• 2+1 2-part layout: 2-voice polyphonic part + monophonic part with aux CV
• *2 3-part layout: 4-voice paraphonic part + monophonic part with aux CV + monophonic part without aux CV
  • Paraphonic part has 4x oscillators
  • Oscillator mode cannot be turned off for the paraphonic part
  • Output channels:
    1. CV: Part 1’s 4x oscillators mixed to 1 audio output, Gate: Part 4’s gate
    2. Part 2, monophonic CV/gate
    3. Part 2, modulation configurable via 3>
    4. Part 3, monophonic CV/gate
• 3M 3-part layout: 3 monophonic parts, plus clock on gate 4 and bar/reset on CV 4
• *1 2-part layout: 4-voice paraphonic part + monophonic part with aux CV
  • Output channels:
    1. CV: Part 1’s 4x oscillators mixed to 1 audio output, Gate: Part 1’s gate
    2. Part 2, monophonic CV/gate
    3. Part 1’s aux CV, configurable via CV
    4. Part 2’s aux CV, configurable via CV

Note processing

Input octave transpose

• Part setting IT (INPUT TRANSPOSE OCTAVES) applies octave transposition to notes received by the part
• Effectively an octave switch for the controller

Filtering by velocity

• Part settings V> (VELOCITY MIN) and V< (VELOCITY MAX) (inclusive) prevent the part from receiving notes with velocity outside the specified range
• Output velocity range is scaled up to compensate for the restricted range imposed by input filtering
• Velocity filtering settings are hidden in 4V layout

Fix for playing on part A while recording part B

• Bug fix: If playing part A while part B is recording, any MIDI notes ignored by the recording part (due to channel, velocity, etc) are still eligible to be received by other parts

Hold function

How the hold function works

• The hold function is a generalization of the original firmware’s panel-controlled latch and pedal-controlled sustain
• Hold function is per-part, as opposed to original firmware’s global latch/sustain
• The hold function is controllable by both pedal CC and panel interface
  • Pedal: send MIDI CC 64 to a part channel to activate or deactivate the part’s hold function
  • Panel: hold REC button (while not recording a sequence) to manually toggle whether the active part’s hold function is active

Hold mode

New part setting HM (HOLD PEDAL MODE) sets how the part responds to MIDI key-releases and the state of the hold function:

1. OFF: hold function has no effect
2. SUSTAIN: sustains key-releases while hold is active, and stops sustained notes on hold deactivation
   • Matches original firmware behavior when pressing the pedal to sustain
3. SOSTENUTO: while hold is active, sustains key-releases only on keys that were pressed before hold activation; stops sustained notes on hold deactivation
4. LATCH: sustains key-releases while hold is active; stops sustained notes on key-press regardless of hold state
   • Matches original firmware behavior when holding START/STOP button to latch
5. MOMENTARY LATCH: like LATCH, but stops sustained notes on hold deactivation, instead of on key-press
6. CLUTCH: while hold is active, sustains key-releases only on keys that were pressed before hold activation (like SOSTENUTO); while hold is inactive, stops sustained notes on key-press (like LATCH)
   • Notes triggered while hold is active are not sustained, and do not cause sustained notes to be stopped, which allows temporarily augmenting a sustained chord
7. FILTER: while hold is active, ignores key-presses and sustains key-releases; while hold is inactive, stops sustained notes on key-press
   • By setting opposite hold polarity on two parts, allows using the hold function to play one part while latching another

Hold polarity

New part setting HP (HOLD PEDAL POLARITY) inverts the part’s active/inactive hold state:

• Allows compatibility with any manufacturer’s pedals
• Allows reversing a hold mode’s semantics
• More information on negative (default) and positive pedal polarity ↗

Display of keys affected by the hold function

Display blinks 1 tick mark for each pressed/sustained key in the active part (max 6 keys):

1. Bottom-half tick: key is manually pressed, and will stop when released
2. Full-height tick: key is manually pressed, and will be sustained when released
3. Steady top-half tick: key is sustained, and will continue after the next key-press
4. Flickering top-half tick: key is sustained, and will be stopped by the next key-press

Clocking

Clock settings

Master clock offset

New global setting C+ (CLOCK OFFSET) allows fine-tuning the master clock phase by ±63 ticks:

• Applied in real time if the clock is running
• Designed to aid in multitrack recording
• Arithmetically, offset is applied after I/ (INPUT CLK DIV)
• If offset is negative, the sequencer will not play any notes until tick 0 is reached

Settings for synced events

• Clock divisions/multiplications are expressed as a ratio of the synced event tempo to the master clock tempo
  • E.g. 2/1 means that the synced event runs at twice the master clock tempo
• Part sequencer: C/ (CLOCK RATIO) sets the base tempo of the part’s sequencer, relative to the master clock tempo
  • Sequence phase may be offset by step offset or loop phase offset
  • Sequencer phase is also affected by number of steps or loop length
• Output clock: O/ (OUTPUT CLOCK RATIO) sets the tempo of the clock gate output, relative to the master clock tempo
• LFO sync: LFO rate sets the part’s base LFO tempo, relative to the master clock tempo
  • Individual LFOs may be dephased or detuned by LFO spread
• 32 clock ratios available:
  • 20 slower than master clock:
    • 1/8, 1/7, 1/6, 1/5, 2/9, 1/4, 2/7, 1/3, 3/8, 2/5, 3/7, 4/9, 1/2, 4/7, 3/5, 2/3, 3/4, 4/5, 6/7, 8/9
  • 12 equal or faster:
    • 1/1, 8/7, 6/5, 4/3, 3/2, 8/5, 2/1, 8/3, 3/1, 4/1, 6/1, 8/1

Master clock controls

Start/stop master clock

• Start master clock
  • Press START/STOP button while not running, or send MIDI Start/Continue, or send MIDI Note On
    • NB: starting via MIDI Note On requires MS (CLOCK MANUAL START) be disabled, i.e. note-based starts are not blocked. Original firmware terminology: button or MIDI Start/Continue is “manual” start, MIDI Note On is “automatic” start
    • Bug fix: a manual start supersedes an automatic clock start, preventing the clock from stopping after all notes are released
  • Display splashes |>
  • Starts from master song position
  • Bug fix: recording part responds to MIDI Start
• Stop master clock
  • Press START/STOP button while running, or send MIDI Stop
  • Display splashes ||
  • Preserves master song position at the time of stopping
  • Bug fix: stopping the clock no longer stops manually held keys, though it still stops notes generated by the sequencer/arpeggiator

Set master song position

• Reset master song position
  • Hold START/STOP button, or start via MIDI Start
  • Display splashes []
  • Updates song position to 0
• Cue to an arbitrary master song position
  • Send MIDI Song Position Pointer with the desired song position
  • Display splashes << if moving earlier, >> if moving later, or [] if moving to 0
  • Note: this was tested with a Tascam Model 12 as a MIDI clock source. If this doesn’t work with your MIDI clock source, let me know!

Cueing synced events from master song position

Synced events have deterministic clocking

• All synced events have sync settings that control the event’s relationship to the master clock’s tempo and song position
  • E.g.: given that an LFO has sync ratio 1/2 and phase offset 0, and the master song position is the 11th beat, the LFO knows that it should be halfway through its 6th cycle
• Past clock state is ignored, preventing temporary setting changes from causing permanent phase drift
• Allows synced clocks to maintain a consistent response to a given song position and sync ratio

How synced events respond to master song position

• If you change the song position (or a sync setting), synced events change phase accordingly
• Sequencers rewind or fast-forward to a recalculated position
• Synced LFOs slew to a recalculated phase
  • NB: free-running LFOs ignore song position, i.e. if part’s base LFO rate is free-running and/or spread is detuning LFOs
• Arpeggiator uses the held arp chord (if any) to fast-forward to the arp chord position corresponding to the new song position

Sequencer

Sequencer controls

• To enable: set desired active part, then set play mode to SEQUENCER, and set SM (SEQ MODE) to STEP or LOOP
• Hold REC button to clear sequence
• Hold TAP button to toggle triggered-erase mode, which will clear the sequence as soon as a new note is recorded
• First press of REC button switches the display to show the pitch (or RS/TI) instead of the step number. Press REC button a second time to exit recording

Sequencer input response

Part setting SI (SEQ INPUT RESPONSE) sets the response of the part’s sequencer to MIDI input (when not recording):

1. OFF: ignores MIDI input
2. TRANSPOSE: original firmware behavior
3. REPLACE: sequencer controls note timing, but MIDI input overrides note pitch
4. DIRECT: MIDI input is directly voiced, allowing accompaniment of a sequence

Step sequencer

Step swing

• Global setting SW (SWING) is compatible with all clock ratios
  • Swing was hardcoded for ratio 4:1 (f.k.a. sixteenth notes) in original firmware
• Swing can be applied to either even or odd steps
  • Turning counter-clockwise: odd steps (1, 3, 5…) are swung by the selected amount
  • Turning clockwise: even steps (2, 4, 6…) are swung (original firmware behavior)

Step selection interface

• Display brightens while the selected step is being played
• When using encoder to scroll through steps, wraps around if the end is reached

Step slide interface

• While recording, hold START/STOP button to toggle slide on the selected step
• If the selected step has slide, the display will show a fade effect
• When a rest/tie step is recorded, slide is removed from that step
  • If a real note is later overdubbed into this step, slide must be re-added manually

Step offset

• Part setting SO (STEP OFFSET) replaces ER (EUCLIDEAN ROTATE) from the original firmware
• Allows rotating both step sequences and euclidean patterns
• Allows starting the step sequencer on any step
  • If using arpeggiator, arp state is pre-advanced to match
• If updated while the sequencer is running, affects the next step played

Other step sequencer changes

• Euclidean rhythms affect the step sequencer as well as the arpeggiator
  • Sequencer always advances, but euclidean rhythm will make some steps emit a rest instead of a note
• Max sequence length reduced from 64 to 30 steps, to free up space in the preset storage

Loop sequencer

How the loop sequencer works

• Real-time recording captures the start and end of notes as you play them
• Chords and overlapping notes are played back according to the part’s polyphony settings
• Start/end times are recorded at 13-bit resolution (1/8192 of the loop length)
• Loop length is set by L- (LOOP LENGTH) in quarter notes, combined with the part’s clock ratio
• Holds 30 notes max – past this limit, overwrites oldest note

Recording a new loop

• Set SM (SEQ MODE) to LOOP
• Press REC button to begin recording
• Send MIDI notes to the part to record them into the loop
• Display brightness fades to show loop progress (if not playing a note), or the progress of the currently playing note
• Channel LEDs show the quarter-phase of the loop

Editing a recorded loop

• Play more notes to overdub
• Press START/STOP button to erase the oldest note, or TAP button to erase the newest note
• Hold REC button to erase the loop

Loop phase offset

• Scroll the encoder to change the loop phase offset
• Turning counter-clockwise: shifts notes later
• Turning clockwise: shifts notes earlier by 1/128

Arpeggiator

Arpeggiator basics

• To enable: set desired active part, then set play mode to ARPEGGIATOR
• NP (NOTE PRIORITY) determines how held keys are ordered in the arp chord
• AD (ARP DIRECTION) sets the algorithm used to traverse the ordered arp chord
• Combine these to create traditional and novel arp behaviors:
  • “Up”: set NOTE PRIORITY to LOW and ARP DIRECTION to LINEAR
  • “Up-down”: set NOTE PRIORITY to LOW and ARP DIRECTION to BOUNCE
  • “Played order”: set NOTE PRIORITY to FIRST and ARP DIRECTION to LINEAR

Arpeggiator rhythm settings

Selecting an arpeggiator rhythm basis

• AP (ARP PATTERN) sets whether the arp rhythm is pattern-based or sequencer-based
• Turning counter-clockwise: sequencer-based arp rhythm
  • S1-S8 will reset the arp state after every 1-8 plays of the entire sequence (step or loop)
    • Reset helps generate more predictable arp output
  • S0 never resets the arp state
• Turning clockwise: rhythm is selected from 23 hardcoded rhythms (original firmware behavior)

Using sequencer-based arpeggiator rhythm

• Like with pattern-based rhythms, the arp chord is set by holding keys on the MIDI controller
• Unlike pattern-based rhythms, the sequencer-based arp rhythm moves through the arp chord only when the loop/step sequencer encounters a new note, instead of advancing on every clock pulse
• A sequence (either loop or step) must exist to produce arpeggiator output
• If a sequencer step is a rest/tie, the arpeggiator will emit that rest/tie
• The velocity of an arpeggiator output note is calculated by multiplying the velocities of the sequencer note and the held key

Sequencer-programmed arpeggiator

Using sequencer notes to program the arpeggiator

• The JUMP and GRID arp directions can interpret the sequencer pitch as a movement instruction
• The arpeggiator always has some active position within the ordered arp chord, e.g. “the 3rd key of the chord”
• Changes in the active position (“movement”) are determined by the pitch of notes emitted by the sequencer
  • If arp pattern is not sequencer-based, the sequencer pitch data is replaced by the position in the 16-step pattern cycle
• Sequencer pitch is interpreted based on its:
  • Key color (is the key black or white?)
  • Shown octave number (with C as the first note of the octave)
  • Pitch ordinal within octave and color, e.g.
    • When the sequencer pitch is the 2nd white note of octave 5, the pitch ordinal is 2
    • When the sequencer pitch is the 4th black note of octave 2, the pitch ordinal is 4
• The octave and color are used for different purposes by JUMP vs GRID (see below)
• The pitch ordinal sets the minimum arp chord size for which the arpeggiator will emit a note
  • For a sequencer pitch with pitch ordinal N, the arpeggiator emits a note only if there are N or more keys in the arp chord, e.g.:
    • When the sequencer pitch is the 3rd white key of its octave, a note is emitted only if there are 3+ keys in the arp chord
    • When the sequencer pitch is the 1st black key of its octave, a note is emitted only if there are 1+ keys in the arp chord
  • Allows dynamic control of the arpeggiator’s rhythmic pattern by varying the size of the arp chord

JUMP direction

• Uses a combination of relative and absolute movement through the ordered arp chord
• Both colors advance the active position in the chord by octave-many places, wrapping around to the beginning of the chord
• White steps emit a note from the active position in the chord, e.g.:
  • When the sequencer pitch is the 5th white note of octave 2, and the active position is 1 out of the chord’s 6 notes, the active position is first incremented by 2 to become 3, and then the 3rd note of the chord is emitted
• Black steps ignore the active position, instead treating the pitch ordinal as an absolute position in the chord, e.g.:
  • When the sequencer pitch is the 3rd black note of octave 5, the emitted note is the 3rd note of the chord, while the active position is incremented by 5

GRID direction

• Simulates an X-Y coordinate system
• The ordered arp chord is mapped onto the grid in linear fashion, repeated as necessary to fill the grid
• Octave sets the size of the grid: 4th octave => 4x4 grid (minimum 1x1)
• White keys advance by 1 along the X-axis, moving left-to-right and wrapping back to the left
• Black keys advance by 1 along the Y-axis, moving top-to-bottom and wrapping back to the top

Note voicing

Polyphonic voice allocation

Polyphonic voicing options

New and improved values for VO (VOICING) setting:

• sM STEAL LOWEST PRIORITY RELEASE MUTE
  • Steal from the lowest-priority existing note IFF the incoming note has higher priority
  • Does not reassign voices to unvoiced notes on release
  • F.k.a. POLY in original firmware
• PRIORITY ORDER
  • Voice 1 always receives the note that has priority 1, voice 2 the note with priority 2, etc.
  • F.k.a. SORTED in original firmware
• UR UNISON RELEASE REASSIGN
  • F.k.a. U1 in original firmware
• UM UNISON RELEASE MUTE
  • F.k.a. U2 in original firmware
• SM STEAL HIGHEST PRIORITY RELEASE MUTE
  • Steal from the highest-priority existing note IFF the incoming note has higher priority
  • Does not reassign voices to unvoiced notes on release
  • F.k.a. STEAL MOST RECENT in original firmware
• sR STEAL LOWEST PRIORITY RELEASE REASSIGN
  • Steal from the lowest-priority existing note IFF the incoming note has higher priority
  • Reassigns voices to unvoiced notes on release
• SR STEAL HIGHEST PRIORITY RELEASE REASSIGN
  • Steal from the highest-priority existing note IFF the incoming note has higher priority
  • Reassigns voices to unvoiced notes on release

Note priority changes

• Added new FIRST (oldest) setting to NP (NOTE PRIORITY)
• Polyphonic voicing respects note priority where applicable
• Arpeggiator respects note priority

Other polyphony changes

• Notes that steal a voice are considered legato
• Bug fix: UNISON allocates notes without gaps
• Bug fix: prevent unneeded reassignment/retrigger in PRIORITY ORDER/UNISON during a partial chord change
• PRIORITY ORDER and UNISON RELEASE REASSIGN reassign voices on NoteOff if there are unvoiced notes
• Allow monophonic parts to use all voicing modes

Legato and portamento

Legato settings

• Replaced original LG (LEGATO) setting (three values) with two on/off part settings
• LG (LEGATO RETRIGGER): are notes retriggered when played legato?
• PL (PORTAMENTO LEGATO ONLY): is portamento applied on all notes, or only on notes played legato?
• Enables a new behavior: notes played legato are retriggered + portamento is applied only on notes played legato

Portamento setting

• PO (PORTAMENTO) setting remapped and extended from 51 to 64 values per side
• Turning counter-clockwise: increases constant-time portamento from T1 to T63
• Turning clockwise: increases constant-rate portamento from R1 to R63

Voice modulation

Envelope

Envelope shape

• Configured per-part in ▽A (AMPLITUDE MENU)
• Each voice’s envelope is gated by the voice receiving note-on and note-off events
• Baseline envelope shape is set per part. When a voice in the part receives a note-on, the note velocity modulates this shape, yielding a unique envelope shape for that note
• Part settings for the base ADSR of each voice in the part:
  • AI (ATTACK INIT), DI (DECAY INIT), SI (SUSTAIN INIT), RI (RELEASE INIT)
• Part settings for the bipolar modulation of each voice’s ADSR by that voice’s note velocity:
  • AM (ATTACK MOD VEL), DM (DECAY MOD VEL), SM (SUSTAIN MOD VEL), RM (RELEASE MOD VEL)
• Part setting PV (PEAK VEL MOD) sets the velocity-sensitivity of the peak level for each voice’s envelope
  • Peak level: height of the instantaneous point where attack ends and decay begins
  • Zero: peak level is always maximum (unity, i.e. equal to the maximum sustain level)
  • Turning clockwise (positive): peak level is increasingly damped by low note velocity
  • Turning counter-clockwise (negative): peak level is increasingly damped by high note velocity
• All curves are exponential
• Min/max stage times: ~0.089 ms (4 samples = 4/45000 of a second) to 10 seconds

How the envelope adapts to interruptions

Envelope adjusts to notes that begin/end while a stage or another note is in progress:

1. Problem: release/attack is farther from target than expected
   • Cause: note ends while decay is falling toward sustain; new note reverses the polarity of the timbre envelope
   • Solution: steepen stage curve to cover more distance in the same time
2. Problem: attack/release is closer to target than expected
   • Cause: note begins during release; note ends while attack is rising toward sustain
   • Solution: shorten stage duration in proportion to remaining distance, maintaining nominal curve shape
3. Problem: attack starts above the peak level
   • Cause: after an early release from a high peak level, new note begins with a low peak level
   • Solution: skip to decay
4. Problem: new note begins with an updated sustain level, without a release from the previous note
   • Cause: legato play
   • Solution: use a decay stage to transition to updated sustain level

Modulation destinations for envelope output

• Aux CV output: ENVELOPE (itself modulated by tremolo LFO)
• Oscillator gain, when oscillator audio mode is ENVELOPED
• Oscillator timbre, when oscillator audio mode is not OFF
  • Timbre modulation by envelope has both a velocity-agnostic and a velocity-dependent component that are added together:
    • Part setting TE (TIMBRE ENV MOD) sets the velocity-agnostic bipolar modulation depth of each voice’s timbre by its envelope
    • Part setting TV (TIMBRE VEL MOD) sets the velocity-dependent bipolar modulation of each voice’s timbre by its envelope
  • Combine these for effects like: negative timbre envelope on low velocity and positive timbre envelope on high velocity, or vice versa

Low-frequency oscillator

LFO speed and sync

• Part setting LF (LFO RATE) sets the part’s base LFO speed, and whether it’s synced or free-running
• Turning counter-clockwise: LFO is synced to master clock, with sync ratio increasing from 1/8 to 8/1
• Turning clockwise: LFO is free-running, with frequency increasing from 0.125 Hz to 16 Hz
  • NB: free-running LFOs ignore master song position
• F.k.a. VIBRATO SPEED in original firmware

LFO spread: dephase or detune

• Each voice within a part has three LFOs: vibrato, tremolo, and timbre
• Within a part, related LFOs can have a phase or frequency offset from each other
• LT (LFO SPREAD TYPES): for each voice in the part, dephase/detune between the voice’s LFO destinations (vibrato, tremolo, timbre)
• LV (LFO SPREAD VOICES): dephase/detune LFOs between the part’s voices
  • Only available in polyphonic/paraphonic layouts
• Turning counter-clockwise: dephase LFOs
  • Each LFO’s phase is progressively more offset, ranging from 0° to 360°
  • Ideal for quadrature and three-phase modulation
  • When dephasing, the LFOs have the same frequency but different phases
• Turning clockwise: detune LFOs
  • Each LFO’s frequency is a multiple of the last, ranging from 1x to 2x
  • Good for chorus/supersaw effects
  • NB: detuned LFOs are free-running and ignore master song position
• In polyphonic/paraphonic layouts, spread can simultaneously apply across both types and voices
  • E.g. a 4-voice paraphonic part can have a distinct phase or frequency for each of its 12 LFOs (3 per voice)
  • Spread is additive, e.g. if LFO SPREAD VOICES is set to dephase and LFO SPREAD TYPES is set to detune, all LFOs will be dephased by voice and then additionally detuned by type

Modulation destinations for LFO output

• Aux CV outputs: LFO, VIBRATO LFO (unattenuated and attenuated versions of vibrato LFO)
• Vibrato LFO: oscillator pitch, pitch CV
  • VB (VIBRATO AMOUNT) (in ▽S (SETUP MENU)): part setting that attenuates bipolar vibrato LFO’s mdulation of CV/oscillator pitch
    • Allows vibrato control via panel interface if your MIDI controller doesn’t have a modulation wheel
  • VS (VIBRATO SHAPE) (in ▽S (SETUP MENU)): shape of the vibrato LFO
• Tremolo LFO: oscillator gain, ENVELOPE output for aux CV
  • TR (TREMOLO DEPTH) (in ▽A (AMPLITUDE MENU)): part setting that attenuates the unipolar tremolo LFO’s reduction of gain for the oscillator and ENVELOPE aux CV
  • TS (TREMOLO SHAPE) (in ▽A (AMPLITUDE MENU)): shape of the tremolo LFO
• Timbre LFO: oscillator timbre
  • TL (TIMBRE LFO MOD) (in ▽O (OSCILLATOR MENU)): part setting that attenuates the bipolar timbre LFO’s modulation of oscillator timbre
  • LS (TIMBRE LFO SHAPE) (in ▽O (OSCILLATOR MENU)): shape of the timbre LFO
• Options for all LFO shapes: triangle, down saw, up saw, square

Voice oscillator

Oscillator audio mode

Part setting OM (OSCILLATOR MODE) in ▽O (OSCILLATOR MENU) sets whether the part outputs audio:

1. OFF: no audio output
2. DRONE: audio gain is modulated by tremolo LFO, but not by envelope
3. ENVELOPED: audio gain is modulated by both tremolo LFO and envelope

Oscillator timbre

• Each voice has a timbre parameter, which has different effects depending on oscillator shape
• Part setting TI (TIMBRE INITIAL) sets the base timbre for all voices in the part
• Each voice’s individual timbre can be modulated by that voice’s envelope and timbre LFO

Oscillator shape

• OS (OSCILLATOR SHAPE) part setting in ▽O (OSCILLATOR MENU)

*- Filtered noise

• Timbre: filter cutoff (resonance is set by note pitch)
• Shapes: low-pass, notch, band-pass, high-pass

┌┐CZ Phase distortion, resonant pulse

• Timbre: filter cutoff
• Shapes: low-pass, peaking, band-pass, high-pass

|⟍CZ Phase distortion, resonant saw

• Timbre: filter cutoff
• Shapes: low-pass, peaking, band-pass, high-pass

-◝ State-variable filter, low-pass

• Timbre: filter cutoff (resonance is constant)
• Shapes: pulse, saw

-W Pulse-width modulation

• Timbre: pulse width
• Shapes: pulse, saw

|⟍┌┐ Saw-pulse morph

• Timbre: morph from saw to pulse

-$ Hard sync

• Timbre: detunes the synced oscillator
• Shapes: sine, pulse, saw

-F Wavefolder

• Timbre: folding amount
• Shapes: sine, triangle

┴┴ Dirac comb

• Timbre: harmonic content

ST Compressed sine: hyperbolic tangent (tanh) function

• Timbre: compression amount

SX Exponential sine

• Timbre: exponentiation amount

FM Frequency modulation

• Timbre: modulation index
• Shapes: 26 preset modulator ratios
  • 11 harmonic ratios, ordered from most harmonic to least harmonic:
    • 1/1, 2/1, 3/1, 5/1, 7/1, 5/2, 7/2, 9/2, 7/3, 8/3, 9/4
  • 8 irrational numbers based on the inverse of Minkowski’s question-mark function ↗:
    • 1/?⁻¹(4/9), 1/?⁻¹(3/7), 1/?⁻¹(2/9), 1/?⁻¹(2/7), 1/?⁻¹(2/5), 1/?⁻¹(1/7), 1/?⁻¹(1/5), 1/?⁻¹(1/3)
  • 7 irrational divisions/multiples of pi:
    • π/4, π/3, π/2, π, π*2, π*3, π*3/2