Ambi Encoder VOT

s3g Ambi Encoder VOT is a 64-voice vector-wavetable CLAP instrument with direct first-through-seventh-order ACN/SN3D output. VOT stands for Vector Oscillator Transform.

Sixteen waveforms form a 4 x 4 timbre field. Voices move through that field under manual control, spatial motion, or an editable vector score, then occupy independent azimuth, elevation, and distance positions in the encoded sound field.

Ambi Encoder VOT waveform field, vector score, tuning relationships, and ambisonic voice space — U/V controls timbre, score time controls traversal, and AED controls spatial position.

Workflow

Insert s3g Ambi Encoder VOT on a 64-channel REAPER track and select the ambisonic ORDER.
Follow it with an ambisonic decoder.
Choose FREE, MIDI, or BOTH for the voice source.
Select the active VOICES, wave set, and pitch relationship.
Use FIELD for spatial motion, VECTOR for the wavetable surface, and SCORE to compose a timbre path.

FREE generates sustained voices internally. MIDI creates voices from incoming notes. BOTH combines both behaviors.

Control Model

The instrument keeps timbre and space independent:

U is horizontal position in the 4 x 4 waveform field.
V is vertical position in the waveform field.
TIME places a node along the vector score.
AZ, EL, and DIST place a voice in ambisonic space.

U and V are normalized from 0 to 1. They are timbre coordinates, not physical axes. The waveform at any position is a smooth Gaussian-weighted blend of nearby tables.

Views

View	Purpose
FIELD	Shows voice positions, nearest-neighbor relationships, camera views, and AED motion.
VECTOR	Shows the 4 x 4 waveform bank, the current U/V route, and the interpolated waveform.
SCORE	Edits U and V as parallel timeline lanes and shows the resulting route through the waveform field.

Ambi Encoder VOT VECTOR view showing the populated four-by-four waveform field, current U and V position, and interpolated waveform — VECTOR view: the 4 × 4 table field, current U/V cell blend, and resulting interpolated waveform.

Vector Score

The score contains 2 to 16 nodes. Select a node in SCORE view, then edit its time, U/V position, and interpolation curve. Use the header - and + buttons to change the node count, or RESET to restore the default score.

Ambi Encoder VOT SCORE view showing parallel U and V breakpoint lanes, loop region, and point route — SCORE view with the parallel U/V lanes, editable nodes, and shaded sustain-loop region.

Control	Meaning
MODE	`OFF` disables score traversal. `ONE` plays once, `LOOP` repeats the sustain region, and `PING` alternates direction.
DURATION	Sets the complete score duration.
DEPTH	Sets how strongly the score owns U/V. At 100%, the score supplies the full timbre path; lower values retain more spatial-scene and `LINK` influence.
LOOP A	Selects the first node in the sustain loop.
LOOP B	Selects the last node in the sustain loop.
TIME	Places the selected node within the score duration.
U / V	Places the selected node in the waveform field.
CURVE	Chooses `LINEAR`, `SMOOTH`, `EXP`, or `HOLD` interpolation into the next node.

On trigger, score playback begins at the first node and advances into the sustain loop. On release, it moves from the current position toward the final score node. The score changes timbre only; the AED motion engine can continue moving voices through space.

Tuning

Control	Meaning
BASE	Sets the root MIDI note for free-running voices and the scale lattice.
TUNE	Applies a global tuning offset in cents.
SCALE	Selects from the shared 101-scale catalog used by Ambi Vox and Wave Terrain, including core Western tonal and modal scales, paired pentatonic/blues/bebop families, regional and global selections, and modern symmetric collections.
SPREAD	Scales intervals around `BASE`. 0% collapses voices to the root, 100% uses normal scale spacing, and 200% doubles the intervals.
DEV	Adds deterministic per-voice pitch deviation, up to approximately +/-100 cents.
HARM	Pulls upper scale degrees toward the harmonic series around `BASE`.
SUB	Pulls lower scale degrees toward the subharmonic series around `BASE`.

Synthesis

WAVE selects SINE, CLASSIC, DIGITAL, FORMANT, or USER.
LOAD imports a 16-table user bank from a WAV file. Purpose-built 4096-sample atlases map directly to the grid; other files are divided into sixteen equal regions. Multichannel files are averaged to mono.
SCAN and its rate move voices around their current vector position.
MORPH controls interpolation among neighboring waveform tables.
ATTACK, DECAY, SUSTAIN, and RELEASE shape note amplitude.
OUT trims the encoded output.

Wavetable Library

The included library contains twenty-eight 4 x 4 atlases. Neighboring cells are phase-aligned for smooth Gaussian interpolation, while pitch-dependent table bands reduce aliasing in higher registers. Most banks are RMS-matched; selected dynamic banks intentionally change level as the vector moves. A loaded USER atlas is saved with the REAPER project.

Foundation, Body / Bass, Pulse / Reed, Formant, Organ, Glass / Metal, Fold / Phase, Digital, and String cover tonal, harmonic, resonant, and early-digital materials.
Rungler turns shift-register feedback and tap families into an energy-rising digital field.
Barberpole wraps octave-weighted spectra across the grid edges for looping ascent and descent illusions.
Harmonic Mirage moves between chord-like missing fundamentals and compact harmonic clusters.
Wave Terrain samples rotated paths through a deformed mathematical surface, with a quiet center and louder perimeter.
Codec Ghost treats prediction residuals, stale blocks, and packet substitution as cyclic oscillator material. Codec Conceal and Cascade extend it through lost-block reconstruction and repeated transcode generations. Codec Residual is joined by Delta, Comb, and Feedback variations that expose finite-difference order, delayed predictor memory, and bounded recursive error as pitched material.
Cellular converts evolving cellular-automata generations into activity-shaped waveforms.
Glitch / Address moves through cyclic repeats, reversals, skips, and reordered waveform addresses.
Glitch / PCM explores held samples, reduced amplitude words, bit patterns, and deterministic data-like cycles.
Glitch / Fracture introduces polarity blocks, dropouts, hard edges, and broken waveform continuity.
Vocal / Black Metal, Throat, Choir, and Animal provide formant, closure, strain, and body fields for vocal-source vector synthesis.

Rungler and Cellular span approximately 11 dB across U. The Codec family spans approximately 9 to 12 dB; Residual Delta, Comb, and Feedback span approximately 9.4, 9.9, and 12 dB respectively. Wave Terrain follows a radial energy basin. These level changes are part of the instrument behavior rather than automatic gain compensation. The exact U/V and energy mapping is listed in the library manifest and preview graphics. The generated files and source format are available in the VOT wavetable directory.

Spatial Motion

The AED engine provides MANUAL, ORBIT, FLOW, PATH, and PULSE scenes. Motion can run freely or follow host transport through CLOCK and DIV.

AMOUNT, motion SPREAD, COHERE, CHAOS, and SMOOTH shape the group motion. AZIMUTH, ELEV, and DISTANCE position its center. NEAR and NBR define nearest-neighbor relationships. In free mode, those relationships can determine when voices become active; LINK lets the spatial state influence U/V while DEPTH decides how much control remains with the vector score.

Order and Routing

The output bus remains 64 channels wide. The selected order activates the corresponding ACN channels and zeroes the remainder.

Order	Active channels
1OA	4
2OA	9
3OA	16
4OA	25
5OA	36
6OA	49
7OA	64

Use a 64-channel track when changing order during a session or working at 7OA.

Automation and State

The synthesis, tuning, motion, and high-level score controls are available to REAPER automation. Individual score nodes and the loaded USER atlas remain internal composition layers and are saved with plugin state.