TT-REESER — User Manual

JSFX instrument for REAPER // Version 0.9.1 // TimTechlor

Two detuned saws are all a Reese needs — TT-REESER adds everything that makes it move: unison and drift, a vowel formant filter, wobble LFOs, phaser and a bass-safe stereo image. One instrument for Reese, neuro, DnB, jungle, speed garage and dark techno basses. No samples.

01The concept

A Reese bass is two sawtooth oscillators, a few cents apart, through a low-pass: the slow beating between them is the sound. Modern neuro, DnB, speed garage and dark techno basses grow from that seed — more voices, drifting pitch, a talking (formant) filter, moving phasers, echoes and space, and a stereo image that stays mono where the sub lives. TT-REESER puts all of that in one JSFX instrument:

TT-REESER front panel — oscillators A and B, filter and formant section with frequency response and vowel pad, two LFOs, envelopes, oscilloscope and effects row
FRONTThe TT-REESER front (preset Noisia Talk). Left: OSC A, OSC B, SUB/NOISE/DRIFT. Middle: filter and formant with the calculated frequency response and the vowel pad, voice and modulation amounts. Right: LFO 1, LFO 2, voice LEDs. Bottom: amp and filter envelope editors, oscilloscope and the effects row.

02Sound examples

21 short phrases, each played with one factory preset as it ships — no edits, no extra effects, no mastering beyond a common loudness. The number in the chips is the preset number in the FACTORY list; the tempo is the one the delay and LFO sync follow. Filter by style, click a row to play it, click or drag the waveform to jump (the faint lines mark the bars). With AUTO on, the next example follows.

TT-REESER // Sound examplesReady
Widescreen Reese

Five voices per oscillator, 1/8 ping-pong delay, short hall

174 BPMPreset 56DnB
0:00
0:00/0:06
Audio file not found. The examples are online at timtechlor.de/docs.

03Quick start

  1. Insert TT-REESER on a track (REAPER 7.x). It is an instrument: it has no audio input and generates sound from MIDI.
  2. Play notes around C1-C3. The INIT program is a single saw through an open filter.
  3. Pick a program with the preset bar in the header (arrows, or click the name for the list) — for example Classic Reese, Wobble 1/8 or Noisia Talk.
  4. Turn CUTOFF, drag the dot in the vowel pad (left-right = vowel, up-down = formant shift), pull the LFO amount sliders.
  5. Automate anything: the first 57 parameters are host-automatable (section 13); everything is stored with the project and in REAPER presets.
Drag a knob vertically, Shift = fine, mouse wheel = step, double-click / right-click / Ctrl+click = default value. Combos: arrows step, click on the middle cycles, right-click goes back. Hover a control for a tooltip with the name and value.

04Oscillators, unison, sub

Both oscillators share the same controls. WAVE: SAW, SQUARE (pulse width PW 5-95 %, also the LFO PWM target), TRI, SINE and REESE-SAW (a saw with a 5 % symmetry shift: the falling edge becomes a short ramp). OCT -2...+2, SEMI -12...+12 and FINE +-100 ct tune each oscillator; the classic Reese is A and B on SAW with B FINE at +5...+15 ct. The beat frequency is f * (2^(cent/1200) - 1): 7 ct at 55 Hz beats at 0.2228 Hz.

Unison, detune, spread

UNISON 1-7 voices per oscillator spread over +-DETUNE cents (button DIST: LIN evenly, EXP clustered towards the middle). SPREAD pans the voices across the stereo field (oscillator B mirrored against A). Each unison voice is scaled by 1/sqrt(n). PH chooses the start phase: FREE (random, analog) or RESET (all voices start at PHASE degrees — identical, punchy attacks).

At most 14 oscillators run in MONO and 8 per voice in POLY (4 x 8 = 32). If A + B exceed the limit, both unison counts are scaled down proportionally; the panel titles show the effective voice count (voices 4).

Drift, sub, noise

Quality

QUALITY HIGH renders the oscillators at twice the sample rate with 4-point PolyBLEP corrections and a 48-tap FIR decimator: a saw at C6 (1046 Hz, 48 kHz) has its strongest non-harmonic component at -93 dBc. DRAFT runs at the base rate (still 4-point PolyBLEP, -43.5 dBc at C6) and saves roughly a quarter of the CPU (measured with unison 7 + 7) — an emergency switch for huge projects; low bass notes alias far less than notes near C6.

05Voices, glide, MIDI

MODEBehaviour
MONO-LEGATOOne voice. Overlapping notes are bound: the pitch glides to the new note and the envelopes are not restarted (no click: measured max step 0.004 FS). Releasing the newer note while the older is held returns to the older note (note stack, last-note priority). Detached notes retrigger.
POLY 4Four voices, round-robin allocation. A fifth note steals the oldest gated voice (voices already in release are taken first). Unison is reduced to 8 oscillators per voice.

GLIDE 0-2 s glides exponentially in pitch: after the GLIDE time 63 % of the pitch distance (in cents) is covered — measured 0.501 s for GLIDE 0.5 s. GLIDE LEGATO glides only between bound notes, GLIDE ALWAYS also from the previous note into a detached one. Glide is a MONO feature.

All channels are accepted, MIDI is consumed (no thru). Sustain pedal (CC64) and MPE are not implemented.

06Filter

A four-pole ladder in zero-delay-feedback form. The resonance loop is solved without delay (y4 = (G^4 x + B) / (1 + G^4 k)) and limited by the signal level (k_eff = k / (1 + 0.4 |y4|), the principle of the TT-303 filter) so a high RESO growls instead of exploding.

TypeOutputSlopeCUTOFF means
LP244th stage24 dB/oct-3 dB point at RESO 0, resonance peak at RESO 100 %
LP122nd stage12 dB/oct-3 dB point
BPstages 1-3 difference12 dB/oct each sidecentre (geometric mean of the -3 dB points)
HP4th-order high-pass24 dB/oct-3 dB point

CUTOFF 20 Hz-18 kHz (log). KEYTRK makes the cutoff follow the note (100 % = one octave per octave, reference C4). ENV AMT +-100 % moves the cutoff by up to +-6 octaves with the filter ADSR; VEL>CUT, LFOs and mod wheel add to it. RESO 0-100 % feeds back up to k = 4 (self-oscillation); passband level is partly compensated (sqrt(1+k)).

Measured LP24: -3 dB point 200.0 Hz for CUTOFF 200 Hz and 1008 Hz for 1000 Hz, response within 0.8 dB of the analytic model, slope 23.0-23.5 dB/oct (the 4-pole cascade only approaches 24 dB/oct asymptotically; close to the cutoff it is 14-21 dB/oct). LP12: 12.3 dB/oct.

07Formant filter

The formant filter is three parallel resonant band-passes at the first three formants of a vowel — the "talking bass". The vowel table (male voice):

VowelF1 / F2 / F3 (Hz)Bandwidth (Hz)Q (F/BW)Gain (dB)
A730 / 1090 / 244080 / 90 / 1209.1 / 12.1 / 20.30 / -6 / -12
E530 / 1840 / 248060 / 100 / 1208.8 / 18.4 / 20.70 / -8 / -14
I270 / 2290 / 301060 / 100 / 1204.5 / 22.9 / 25.10 / -14 / -18
O570 / 840 / 241070 / 80 / 1008.1 / 10.5 / 24.10 / -6 / -14
U300 / 870 / 224040 / 80 / 1007.5 / 10.9 / 22.40 / -8 / -16

VOWEL 0-4 walks the table A-E-I-O-U linearly (frequencies, Q and gain are interpolated: VOWEL 0.5 = 630 / 1465 / 2460 Hz, measured 632 / 1463 / 2478). F SHIFT moves all three formants +-1 octave, F Q scales the Q by +-1.5 octaves around the table (50 % = the table values, narrow enough to articulate), F MIX blends the formant sound with the signal. FORMANT position:

At F MIX 0 the stage is skipped completely (bit-identical output, no cost). The vowel pad shows and edits VOWEL (horizontal) and F SHIFT (vertical); right-click resets both. The frequency response display is calculated from the current filter and formant settings (without modulation). VOWEL is modulatable by the filter envelope (ENV>VOW), velocity (VEL>VOW), mod wheel (MOD>VOW) and both LFOs.

Pink-noise excitation, peaks vs. the table (+-4 % required): A 733 / 1108 / 2477 Hz, E 534 / 1847 / 2522, I 277 / 2281 / 3030, O 565 / 851 / 2442, U 296 / 878 / 2262 (worst case 2.6 %). The parallel sum of three bands shifts overlapping peaks a little towards each other; that is physics, not error.

08Modulation

LFO 1 / LFO 2

Shapes SINE, TRI, SAW, SQUARE, S&H. RATE 0.05-30 Hz free, or SYNC to the host tempo with these divisions:

DivisionBeats (quarters)Rate
1/320.12516 Hz at 120 BPM
1/16T0.166712 Hz at 120 BPM
1/160.258 Hz at 120 BPM
1/8T0.33336 Hz at 120 BPM
1/80.54 Hz at 120 BPM
1/4T0.66673 Hz at 120 BPM
1/8.0.752.667 Hz at 120 BPM
1/412 Hz at 120 BPM
1/2T1.3331.5 Hz at 120 BPM
1/4.1.51.333 Hz at 120 BPM
1/221 Hz at 120 BPM
1/2.30.6667 Hz at 120 BPM
1/140.5 Hz at 120 BPM
2/180.25 Hz at 120 BPM
4/1160.125 Hz at 120 BPM
8/1320.0625 Hz at 120 BPM

PHASE shifts the start. RETRIG KEY restarts the LFO with every new (non-bound) note; FREE runs on — with SYNC it is locked to the timeline position, drift-free. Each LFO has three destinations with a bipolar amount:

Destination100 % amount
PITCH+-12 semitones (vibrato, sirens)
DETUNE+-50 ct added to both detune values (beating that breathes)
CUTOFF+-4 octaves
RESO+-50 %
VOWEL+-2 table steps
PHASER+-2 octaves on the phaser sweep centre
PWM+-45 % pulse width
AMPtremolo, full depth at 100 % (gain 0...1)
DRIVE+-50 % drive amount
LFO sync 1/4 at 120 BPM = 2.0000 Hz (1/8 = 4.0000 Hz, 90 BPM = 1.5000 Hz), free 3.000 Hz, shape correlation 0.99+, key-retrigger reproducible to 0.7 % of the envelope.

Envelopes

Amp and filter ADSR: 0.5 ms-5 s, exponential curves. Attack aims at 1.3 so it reaches full level after exactly the displayed time; decay and release reach 1 % of their target in the displayed time. Drag the nodes in the editor (attack node = A, decay node = D and S, release node = R); right-click a node resets it. A new attack starts from the current level, so fast retriggers never click.

09Effects & bass mono

Fixed order: voices - drive - phaser - chorus - bass mono / width - delay - reverb - volume - soft limiter. The FX row of the front has two pages (buttons FX 1 / FX 2 in the title bar of the left panel): FX 1 = drive, phaser, chorus; FX 2 = delay, reverb. MASTER is visible on both.

Drive

Types SOFT (tanh), TUBE (asymmetric), FUZZ (exponential) and FOLD (sine wavefolder, input limited to +-4 so it can never blow up). DRIVE 0-40 dB pre-gain, TONE low-passes the wet signal (800 Hz-18 kHz), MIX blends parallel. The shaper runs 2x oversampled and each type is level-normalised. DRIVE 0 (and MIX 0) is an exact bypass: the output is bit-identical to a chain without the stage. The curve display shows the transfer function.

Phaser

4, 6 or 8 first-order all-pass stages per channel, sweep centre CENTER 100-5000 Hz, DEPTH in octaves (100 % = +-2 oct), FEEDBACK +-95 %, free rate or tempo sync, STEREO offsets the right channel's LFO up to 180 degrees. The output is in + MIX * 0.5 * (allpass - in): at MIX 100 % you get the classic notch pattern (8 stages = 4 notches). Measured with 8 stages, FEEDBACK 0.5: notch depth 16-19 dB, notch positions within tolerance of the analytic model at CENTER 2 kHz (412 / 1360 / 3049 Hz) and 200 Hz (124 / 288 / 1030 Hz), the sweep follows CENTER x 2^(+-DEPTH) (ratio 10.6-11 for 200 Hz - 2 kHz).

Chorus

Two modulated delay taps per channel (15 / 19 ms centre, up to +-4 ms), right LFO 90 degrees behind: RATE, DEPTH, MIX. MIX 0 is an exact bypass.

Delay

Stereo delay after the bass mono stage, so a mono bass is echoed in mono. SYNC switches between a free time (TIME 1-1500 ms) and a tempo division (DIV: 1/32, 1/16T, 1/16, 1/8T, 1/8, 1/4T, 1/8., 1/4, 1/2T, 1/4., 1/2, 1/2.; host tempo, 120 BPM when the host reports none; times above 2.5 s are limited). PING-PONG feeds the mono sum into the left line and bounces the repeats left - right - left. FEEDBACK 0-95 %. Inside the feedback loop each repeat passes a LOW CUT (20-400 Hz, one pole; keeps the bass out of the repeats), a HIGH CUT (1-16 kHz, one pole) and the SATURATION (0-100 %, tanh, unity gain for small signals; at 0 % a transparent soft clip). The loop can never build up: the feedback path is always limited, so 95 % feedback with a steady signal settles at a fixed level. DUCK lowers the delay signal while the dry bass plays (envelope follower, 5 ms attack / 200 ms release; at 100 % the repeats are silent above roughly -16 dBFS dry level) so the bassline stays clear. MIX adds the delay signal to the dry signal (0-100 %). Changing TIME or the tempo moves the read pointer smoothly (at most 0.3 samples per sample, tape-like pitch sweep, no clicks). MIX 0 is an exact bypass; the buffers are cleared once when the delay is switched off.

Reverb

A feedback delay network (FDN): the input passes a low cut and the pre-delay, then feeds eight delay lines (31-88 ms, scaled by SIZE and by the sample rate). Each line has its own damping low-pass and a gain set from DECAY so that the level falls by 60 dB after DECAY seconds; the lines are coupled by an orthogonal Householder matrix (no energy is lost in the mixing) and a very slow modulation (about 0.35 ms) breaks up metallic ringing. Even lines make up the left, odd lines the right output. SIZE 0-100 % scales the line lengths (0.4 - 1.6 x), DECAY 0.2-10 s (this is the low-frequency decay time; with DAMPING above 0 the highs die sooner, and measured RT60 comes out about 10-15 % below DECAY even at DAMPING 0), PRE-DELAY 0-200 ms, DAMPING (highs decay faster; 0 = none, 100 % = 1.2 kHz), LOW CUT 20-600 Hz (default 150 Hz) on the reverb input so the bass does not turn into mud, WIDTH 0-100 % (0 = mono), MIX adds the reverb to the dry signal. The input level is compensated so that the total reverb energy does not depend on DECAY. MIX 0 is an exact bypass (buffers cleared once).

Tails. Delay and reverb keep running when the voices are silent: the repeats and the reverb tail ring out after note-off and after the transport stops. Starting the transport from stop (and a sample-rate change) clears them. QUALITY DRAFT makes the reverb cheaper: 4 instead of 8 lines and no modulation; the delay is unchanged (it is cheap).

Bass mono and width

BASS MONO keeps everything below MONO BELOW (20-300 Hz, default 120 Hz) in the centre. The signal is split into mid M = (L+R)/2 and side S = (L-R)/2; only S passes a 4th-order Butterworth high-pass; then L = M + S*WIDTH, R = M - S*WIDTH. The mid signal is never touched. WIDTH 0-200 % scales the side signal (0 = mono, exactly L = R). Measured: S at 0.34 x the corner frequency (41 Hz vs 120 Hz) is -36.7 dB, at 0.69 x -12.9 dB, above 2.7 x the corner 0.00 dB (the specification asked for -20 dB at one third of the corner, a 2nd-order high-pass would give only -19.3 dB).

Master

VOLUME -30...+6 dB, then a soft limiter (linear up to 0.7 FS, tanh knee, hard ceiling 1.0) with NaN/Inf protection. The oscilloscope shows the mono sum, triggered on a rising zero crossing.

10Preset bar, user bank & randomize

Header of TT-REESER with the preset bar: previous, preset name with drop-down, next, SAVE and a menu button, left of the OPTIONS button, and the open preset list with FACTORY and USER sections
PRESETSThe preset bar sits in the header, directly left of OPTIONS: < name > SAVE .... The name opens the list with the FACTORY and USER sections.

57 factory programs plus INIT (single saw, open filter) form the FACTORY list (INIT is its first entry). All are loudness-matched (about -14 dBFS RMS on A1) so you can browse without touching the fader. A program sets every parameter — it does not stack on the previous one. Loading INIT resets everything, also when INIT is already selected.

FamilyPrograms
Classic ReeseClassic Reese, Kemistry Reese, Storm Reese, Wide Reese, Reese Bite, Pulse Reese
Neuro / GrowlNeuro Growl, Noisia Talk, Growl Saw, Metal Neuro, Robot Voice, Neuro Sweep
Speed Garage / 2-Step2-Step Wobble, Garage Stab Bass, Speed Garage Sub, UKG Reese, Bounce Bass
Vowel / TalkingVowel Bass A-E, Talking I-O, U Bass Hum, Formant Wobble, Vocal Growl, Mod Wheel Vowel
Dark Techno ReeseDark Reese, Rumble Reese, Industrial Grit, Hypno Reese, Warehouse Bass
Sub-ReeseSub Reese, Deep Sub Layer, Sine Weight, Low Beat
Wobble 1/4 + 1/8Wobble 1/4, Wobble 1/8, Wobble 1/8 Tri, Wobble 1/8 Dotted, Half-Time Wobble, Wobble Growl
Detroit / FoghornFoghorn, Detroit Reese, Motor City
Stab (poly)Reese Stab, Poly Stab, Neuro Stab
Pad-ReeseReese Pad, Wide Drone, Choir Pad
FX programs (delay / reverb, v0.7.0)Dub Reese, Neuro Hall, Speed Garage Echo, Cathedral Pad Bass, Tape Sub Echo, Dark Techno Room, Jungle Spring Stab, Vowel Cave, Widescreen Reese
House organHouse Organ

New: House Organ, a 90s house organ bass. A triangle carries the body, a sine a fifth two octaves above it rings like a bell and fades through the filter envelope while the body sustains; both oscillators restart in phase, which gives the short key click. Play it around C2.

Preset 57 // New presetReady
House Organ

Triangle plus a decaying sine bell: 90s house organ bass

122 BPMPreset 57House
0:00
0:00/0:08
Audio file not found. The examples are online at timtechlor.de/docs.

The bar loads and saves the whole sound of the instance in one step: all 123 sound parameters, including the ones that have no slider (LFO shapes and destinations, delay and reverb, quality, and so on) and the output volume. Not part of a preset: the colour scheme (OPTIONS), the FX page (FX 1 / FX 2), the randomize locks and the program number that the hidden PRESET parameter carries for automation.

ControlFunction
< / >Previous / next preset: first INIT and the 57 factory programs, then the used user slots; wraps around and loads immediately.
Name fieldOpens the list: FACTORY (58 entries) and USER (32 slots, empty ones show - empty -). Mouse wheel scrolls, click loads. A * behind the name means the sound differs from the loaded preset.
SAVESaves the current sound into the selected user slot; with a factory preset (or none) selected, into the first free slot. A name box opens (up to 24 characters, default User 07): type, Enter or OK saves, CANCEL or a click outside cancels.
...RENAME, COPY (into the next free slot) and DELETE (a second click confirms) for the selected user preset.

Where the user bank lives. Plug-ins cannot write files, so the 32 user slots are part of the plug-in state: they are saved with the project, with FX-chain and track-template copies, and in every REAPER preset of this plug-in (the + button in the FX window toolbar). To take your bank to another project or computer, save a REAPER preset once (+ → Save preset) and load it there: the whole bank comes with it. Projects saved with earlier versions (0.6 to 0.8) load unchanged with an empty bank.

Typing names. REAPER keeps Space for the transport, so type Shift+Space for a blank. Depending on your REAPER settings Esc closes the whole FX window; use CANCEL instead. (The + menu of the FX window also has Send all keyboard input to plug-in.)

RANDOMIZE (second header row, left of the lock keys) re-rolls four groups within musical ranges — OSC (waves, levels, unison, detune, drift, sub), FLT (cutoff, resonance, envelope, formant), MOD (LFO shapes, rates, destinations, amounts), FX (drive, phaser, chorus; delay and reverb are left alone). The four LOCK buttons exclude a group; amp envelope, voice mode, glide and master stay untouched. A randomized sound can be kept with SAVE.

11Front panel map

AreaContents
HeaderPreset bar (< name > SAVE ...), OPTIONS (COLOR), RANDOMIZE + LOCK buttons, MODE (MONO-LEGATO / POLY 4), QUALITY (HIGH / DRAFT), output meter
OSC A / OSC BWAVE, PH (FREE / RESET), DIST (LIN / EXP), LEVEL, PW, PHASE, OCT, SEMI, FINE, UNISON, DETUNE, SPREAD
SUB / NOISE / DRIFTSUB, NOISE, DRIFT, VEL SENS, sub waveform, LP, routing, noise colour
FILTER / FORMANTFilter type, formant position, CUTOFF, RESO, KEYTRK, ENV AMT, VEL>CUT, F MIX, VOWEL, F SHIFT, F Q, response display, vowel pad
VOICE / MODULATIONGLIDE, glide mode, PB RANGE, ENV>VOW, VEL>VOW, MOD>CUT / VOW / DRV, MOD SOURCE
LFO 1 / LFO 2Shape, SYNC, RETRIG, RATE, DIV, PHASE, preview, three destination + amount rows
EnvelopesAmp and filter ADSR editors (drag nodes), oscilloscope
FX row, page FX 1DRIVE (type, drive, tone, mix, curve), PHASER, CHORUS, MASTER (bass mono, width, volume)
FX row, page FX 2DELAY (SYNC, DIV, PING-PONG, TIME, FEEDBACK, LOW CUT, HIGH CUT, SATURATE, DUCK, MIX), REVERB (SIZE, DECAY, PRE-DLY, DAMPING, LOW CUT, WIDTH, MIX), MASTER. Switch pages with the FX 1 / FX 2 buttons in the title bar of the left panel
TT-REESER effects page FX 2: delay with sync, ping-pong, time, feedback, low cut, high cut, saturation, duck and mix; reverb with size, decay, pre-delay, damping, low cut, width and mix
FX 2Delay and reverb page. The FX 1 / FX 2 buttons sit in the title bar of the left panel; MASTER stays on the right.
TT-REESER while editing: the vowel pad dragged with tooltip, cutoff raised, the preset bar with a changed-state star
EDITINGInteraction state. The vowel pad is being dragged (tooltip VOWEL 1.00), CUTOFF was pulled up and the response display follows immediately; the preset bar shows Noisia Talk with a star after the first change.

12Options / Color

The OPTIONS button sits in the header, right of the preset bar and above the randomize lock buttons. It opens a small box with COLOR: ten colour schemes for the whole front — ACID, CYAN, ORANGE, MAGENTA, YELLOW, RED, BLUE, VIOLET, MINT and ICE. Each swatch shows the accent colour (top) and the second colour (bottom). A click on a swatch switches the front at once; the box stays open until you click the button again or click anywhere outside the box.

TT-REESER with the OPTIONS box open: ten colour swatches ACID, CYAN, ORANGE, MAGENTA, YELLOW, RED, BLUE, VIOLET, MINT and ICE, MAGENTA selected
OPTIONSThe OPTIONS box with the ten colour schemes. The factory colour of TT-REESER is MAGENTA.

The colour is stored per plug-in instance (in the project and in the FX chain), so several instances can look different. The factory colour is MAGENTA. It is a display setting only: it does not belong to a preset, is not touched by RANDOMIZE, and does not change the sound. The choice is kept as the hidden slider 65 (UI Color), so it is also saved with REAPER's own FX presets.

13Automation & state

REAPER sliders are limited to 64, so the 57 most useful parameters are real (hidden) sliders — automatable, learnable, stored in REAPER presets. All 124 parameters live in the plug-in's own model and are saved with the project (@serialize, versioned, 32-bit floats; the user preset bank is appended behind them — a save/load round trip returns exactly the same values, tested for all 123 parameters). Slider order is fixed:

#ParameterRange
1CUTOFF20 ... 18000 Hz (default 18000)
2RESO0 ... 100% (default 0)
3VOWEL0 ... 4 (default 0)
4F MIX0 ... 100% (default 0)
5F SHIFT-1 ... 1 oct (default 0)
6F Q0 ... 100% (default 50)
7ENV AMT-100 ... 100% (default 0)
8DRIVE0 ... 100% (default 0)
9VOLUME-30 ... 6 dB (default -3)
10MODEMONO-LEGATO / POLY 4
11GLIDE0 ... 2 s (default 0)
12A LEVEL0 ... 100% (default 100)
13B LEVEL0 ... 100% (default 0)
14B FINE-100 ... 100 ct (default 0)
15A UNISON1 ... 7 (default 1)
16B UNISON1 ... 7 (default 1)
17A DETUNE0 ... 100 ct (default 0)
18B DETUNE0 ... 100 ct (default 0)
19A SPREAD0 ... 100% (default 50)
20B SPREAD0 ... 100% (default 50)
21SUB LEVEL0 ... 100% (default 0)
22NOISE LEVEL0 ... 100% (default 0)
23DRIFT0 ... 100% (default 0)
24AMP ATTACK0.0005 ... 5 s (default 0.0020)
25AMP DECAY0.0005 ... 5 s (default 0.3)
26AMP SUSTAIN0 ... 100% (default 100)
27AMP RELEASE0.0005 ... 5 s (default 0.15)
28FLT ATTACK0.0005 ... 5 s (default 0.0020)
29FLT DECAY0.0005 ... 5 s (default 0.3)
30FLT SUSTAIN0 ... 100% (default 50)
31FLT RELEASE0.0005 ... 5 s (default 0.2)
32LFO1 RATE0.05 ... 30 Hz (default 2)
33LFO1 AMT1-100 ... 100% (default 0)
34LFO2 RATE0.05 ... 30 Hz (default 0.5)
35LFO2 AMT1-100 ... 100% (default 0)
36PHASER MIX0 ... 100% (default 0)
37PHASER RATE0.05 ... 10 Hz (default 0.4)
38PHASER DEPTH0 ... 100% (default 60)
39PHASER FB-95 ... 95% (default 30)
40CHORUS MIX0 ... 100% (default 0)
41MONO BELOW20 ... 300 Hz (default 120)
42WIDTH0 ... 200% (default 100)
43FILTER MODELP12 / LP24 / BP / HP
44KEYTRACK0 ... 100% (default 0)
45FORMANT POSPRE-LP / POST-LP / PARALLEL
46A WAVESAW / SQUARE / TRI / SINE / REESE-SAW
47B WAVESAW / SQUARE / TRI / SINE / REESE-SAW
48A OCT-2 ... 2 (default 0)
49B OCT-2 ... 2 (default 0)
50A SEMI-12 ... 12 st (default 0)
51B SEMI-12 ... 12 st (default 0)
52A FINE-100 ... 100 ct (default 0)
53DRIVE TYPESOFT / TUBE / FUZZ / FOLD
54DRIVE MIX0 ... 100% (default 100)
55DRIVE TONE0 ... 100% (default 100)
56PHASER CENTER100 ... 5000 Hz (default 800)
57PRESET0 ... 63 (default 0)

The remaining parameters (waveform shapes for the sub, PW / PHASE / phase mode / distribution, LFO shape, sync, division, retrigger, destinations 1-3 and amounts 2-3, velocity/wheel routings, phaser stages/sync/stereo, chorus rate/depth, bass mono switch, quality, PB range, glide mode, and all delay and reverb parameters) are editable on the front and saved with the project but cannot be automated.

Slider 58-64 (Q IDX, Q SETV, Q GO, Q VAL, Q CHK, Q ACK, Q NANCOUNT) are a diagnostic interface used by the automated tests to read and write any parameter and to checksum the state. They have no function for playing; leave them alone. Slider 65 (UI Color) is the colour scheme of the OPTIONS box (0-9, factory 3 = MAGENTA).

14Measured values

Every number below was measured in REAPER 7.79 (offline render, 48 kHz unless stated) against an independent Python oracle; the full matrix with pass/fail is in the product folder (PLAN.md, LOGBUCH.md).

TestTargetMeasured
Tuning A1 / C2 / E3+-1 ct0.00 ct
Reese beat, B +7 ct at 55 Hz0.2228 Hz +-5 %0.2228 Hz
Unison 3 x 20 ct-20 / 0 / +20 ct +-2-20.00 / 0.00 / +20.00 ct
Saw alias at C6, HIGH<= -60 dBc-93 dBc (44.1 kHz -95, 96 kHz -116)
Glide 63 % (pitch)= GLIDE time +-10 %0.501 s for 0.500 s
LP24 -3 dB point+-7 %+0.0 % (200 Hz), +0.8 % (1 kHz)
Formant peaks+-4 %worst 2.6 %
Drive 0 / Phaser MIX 0 / F MIX 0bit-exactdiff 0
Bass mono, S at 0.34 fc<= -20 dB-36.7 dB
Poly 4, fifth notesteals the oldestoldest gone (-66 dB), four others present
Programsfinite, peak <= 0 dBFS, tail < 10 s58/58
Save / loadexactidentical (all parameters); a v0.6.0 state loads with delay and reverb off
Delay echo position (free 250 ms, sync 1/8. @100 BPM, 1/4 @120 BPM, 1/8T @90 BPM, 1500 ms)+-0.7 samplesexact to 0.1 sample (48 / 44.1 / 96 kHz)
Delay ping-pong, channel separation<= -60 dBexact (repeat 1 left only, 2 right only, 3 left only)
Delay feedback filter (LOW CUT 400, HIGH CUT 1000)+-1 dB against the model0.00 dB
Delay feedback 95 %, 60 s steady signalfinite, no growthpeak <= 1.0, level ratio late/early 0.997-1.001
Reverb RT60 (DAMPING 0)DECAY +-25 %0.49 s / 1.74 s / 5.02 s for 0.5 / 2 / 6 s
Reverb pre-delay, sizePRE + 31.3 ms; 12.5 / 50.1 ms+100 ms exactly; 12.7 / 50.2 ms
Reverb low cut 600 Hz on a 55 Hz sine>= 15 dB quieter20 dB
Delay / reverb MIX 0identical to v0.6.00 LSB in every repeatable case (see LOGBUCH)
Sample rates44.1 / 96 kHztuning, filter, formant, LFO sync all pass
Measured as offline render time on an AMD Ryzen 9 7950X (48 kHz, one core), minus the same render with the plug-in idle (REAPER's own overhead, 2-3 %): one voice with unison 3 + 3, formant, drive, phaser, chorus, two LFOs, sub and noise at QUALITY HIGH takes about 7 % of a core (5.0 % without sub and noise), unison 7 + 7 about 8 %, four poly voices with 8 oscillators each and all effects about 22 %. QUALITY DRAFT saves about 30 % of the oscillator-heavy cases; effects at MIX 0 cost almost nothing. Delay and reverb (v0.7.0), measured the same way on the unison 3 + 3 case: the delay (ping-pong, saturation, ducking) adds about 3.3 % of a core, the reverb (HIGH, 8 lines) about 4.5 %, both together about 5.3 %; with QUALITY DRAFT the reverb runs 4 lines and both together cost about 3.5 %. The measurement is noisy (+-1 %: other programs were running); a slower CPU scales these numbers, so use DRAFT or fewer unison voices there.

15Troubleshooting