Emulator II V

How it works

Sampling and playback

A sample is a digital recording: the sample rate (here about 27.7 kHz) determines the maximum reproducible frequencies (about half, 13–14 kHz), the number of bits the precision of each measurement. To play different notes the sample is read faster or slower: each octave up doubles the reading speed (and shortens the sound), each octave down halves it.

Companding

The Emulator II stored only 8 bits per sample, but with companding (non-linear compression-expansion, similar to the μ-law of telephony, applied to a differential DPCM encoding): the quantization steps are small for weak signals and large for strong ones. This way the noise stays low in quiet parts and the apparent dynamic range approaches that of a converter with many more bits, with a slight audible compression and coloration.

Variable sample rate

The system read samples at a variable speed for each voice (clock dividers): transposition requires no digital interpolation, and the resulting aliasing has a specific character, part of the E-mu sound.

The analog path

After the converter, each voice passes through an analog resonant 24 dB/octave low-pass filter with its own ADSR envelope and a VCA with its own ADSR: the sample is shaped as in a subtractive synth (filter opening with velocity, decays, swells). An LFO adds vibrato or filter and volume modulation.

Voices and maps

A preset is made of several voices (up to 8): each contains a sample assigned to a keyboard zone with root note, loop and its own filter and VCA parameters. Overlapping or placing zones side by side creates splits, layers and multisamples.

Character and tone

8-bit sampling with digital companding that sounds cleaner and more dynamic than linear 8-bit, at about 27.7 kHz (usable bandwidth up to about 13 kHz), followed by 4-pole analog low-pass filters and analog VCAs per voice. The result is warm, round and cinematic, with a slight grain and natural compression: strings and choirs become soft and wide, flutes and ethnic sounds intimate, drums full-bodied. Compared with CMI V it is less raw and more musical, with a real resonant analog filter; compared with synthesis (DX7 V, Synclavier V) it offers sampled realism but less timbral transformation. Limits: limited high-frequency bandwidth (no air above 13–15 kHz in vintage mode), aliasing when transposing high, and a strongly '80s aesthetic that needs to be dosed in modern productions.

Architecture

Original E-mu Emulator II (1984)

• Successor to the Emulator (1981, linear 8-bit, a few hundred units); launch price about $7,995 ($9,995 for the version with more memory), about a third of a Fairlight Series II.
• 8-bit sampling with digital companding (DPCM with μ-255 law), 27.7 kHz sample rate, 512 kB of memory (expandable), 5.25'' floppy and later an optional hard disk.
• 8 voices of polyphony; 8-voice architecture (samples with keyboard zone) per preset.
• Per voice: analog 4-pole 24 dB/oct resonant low-pass filter (SSM2045 chip) with ADSR envelope, analog VCA with ADSR, LFO for vibrato and modulation; velocity and real-time controls; internal sequencer and separate outputs.
• Famous libraries: Marcato Strings, Shakuhachi (used by Peter Gabriel in Sledgehammer and by Enigma), choirs, and some samples resampled from Fairlight and Synclavier.

Emulator II V (Arturia, 2020)

• Rebuild of the engine with 8 voices per preset, each with sample browser, Start/End, Transpose ±24, Detune, Gain, playback directions (forward, backward, forward+backward, backward+forward with balance), Trig Delay (also synced), loop with crossfade and loop in release.
• Import of WAV, AIFF and FLAC without the memory limits of the original.
• Emulated 24 dB/oct filter (65 Hz–20 kHz) with ENV, KBD, LFO, VEL, Q Vel, velocity on envelope attack; LFO with Sync and Variation (random rate per note); VCA with Alternate > Pan and Pan.
• Polyphony per voice from mono to 32 notes; vintage DAC Mod (emulation of the original converter) or modern, per voice.
• Voice link with offsets relative to the main panel controls; graphic map of keyboard and velocity zones with crossfades and root note.
• Arpeggiator (6 modes, 1–5 octaves, sync); Controls matrix with 5 fixed sources (Velocity, Keyboard, Aftertouch, Pitch wheel, Mod wheel) and many destinations that can be enabled per voice.
• One insert effect per voice (Chorus, Phaser, Distortion, Bitcrusher, Reverb, Delay, 3-band EQ, Multimode Filter) and 3 master slots in series or parallel with 11 effects (Reverb, Delay, Chorus, Flanger, Phaser, Overdrive, Compressor, BitCrusher, Multi Filter, Param EQ, Stereo Pan); undo/redo.

The controls, one by one

Main Panel — Master Control

Volume

Output volume.

Tune

Tuning 400–480 Hz (center 440 Hz).

Transpose

±24 semitones in semitone steps.

Voicelink

Links the voices to the panel controls: moving a control moves all the linked voices (with optional per-voice offsets).

Main Panel — Filter

Frequency

Cutoff of the 24 dB/oct low-pass, 65 Hz–20 kHz. Strings and choirs: 3–6 kHz; basses: 400–1200 Hz.

Resonance

Emphasis at the cutoff; raising it attenuates the rest of the spectrum. 10–30% for color, beyond for wah effects.

ENV

Amount (positive or negative) of the filter envelope on the cutoff.

KBD

Key tracking: the cutoff rises with high notes, keeping the brightness constant.

LFO

LFO modulation of the cutoff (cyclic wah).

VEL

Filter opening with velocity: playing hard makes the sample brighter.

A, D, S, R (Filter Envelope)

Attack, decay, sustain and release of the filter envelope: decay 200–600 ms for plucks and stabs, slow attack for swells.

Main Panel — LFO

Rate

LFO speed (5–6 Hz for natural vibrato).

Delay

Delay before the LFO kicks in: vibrato that enters after the attack.

Vibrato

Amount of LFO on the pitch.

Main Panel — VCA

LFO

LFO modulation of the volume (tremolo).

VEL

Sensitivity of the volume to velocity.

A, D, S, R (Loudness Envelope)

Volume envelope: attack 0 for percussion, 200–600 ms for strings and choirs, release 300 ms–2 s.

Main Panel — Arpeggiator

On

Turns the arpeggiator on.

Rate

0.01–50 Hz free-running or from 1 to 1/64 synced.

Sync

Free or locked to the tempo.

Pattern

As Played, Up, Down, Up-down Inclusive, Up-down Exclusive, Random.

Octave

Range of 1–5 octaves.

Main Panel — FX and wheels

FX 1–3 type and amount

Choice of the master effect for the three slots and their Dry/Wet.

Pitchbend Wheel / Mod Wheel

Bend (range settable 0–24 in the Controls section) and assignable modulation.

Advanced — Offset Assign and Voice Selector

Offset Assign

With linked voices, drag the ring around a knob to define a single voice's offset relative to the panel value.

Voices 1–8: Solo, Mute, Copy/Paste, Volume, Link

Selection and management of the preset's eight voices.

Advanced — Edit: Voice (sample)

Sample Browser

Factory and user banks; click to audition, double-click to load. Accepts WAV, AIFF, FLAC.

Start / End

Played region (0–100%): remove silences and unwanted attacks.

Transpose

±24 semitones for the sample.

Detune

Fine ±50 cents: two voices with the same sample at ±5–8 cents = ensemble.

Gain

Sample level from −80 to +24 dB.

Play direction

Forward, Backward, Forward+Backward, Backward+Forward.

Play direction balance

Mix between forward and backward phases in the combined modes.

Trig delay (bypass, time, sync)

Delays the whole voice by 0–2000 ms or 1/32–1 bar: internal echoes, flams, layers that enter later.

Loop on/off, Loop start/end

Loop within the sample, limited by Start/End.

Loop fade

Loop crossfade 0–1: eliminates clicks on imperfect loops (needs distance between Start and Loop start).

Loop in release

The loop continues during the envelope release (natural long tails).

Advanced — Edit: Filter and LFO

F Vel

Velocity on the filter frequency.

Cutoff, Resonance, LFO amount, KBD, ENV

As on the main panel, for the individual voice.

Q Vel

Velocity on resonance: hard hits are more nasal.

Vel (attack)

Velocity on the filter envelope attack: 50–100% = original behavior, 0–50% = high velocity shortens the attack.

Attack, Decay, Sustain, Release

Voice filter envelope.

LFO Rate, Sync, Delay, Vibrato

Voice LFO with tempo sync.

Variation

Slightly different, random rate for each note: ensemble vibrato, more natural on strings and choirs.

Advanced — Edit: VCA, Voice FX, Output

Alternate > Pan

Each new note goes alternately left and right (bipolar to choose the starting side).

Pan, Level

Stereo position and level of the voice.

LFO amount, Vel, Attack, Decay, Sustain, Release

Voice VCA modulation and envelope.

Voice FX

Per-voice insert: Chorus (Freq, Depth), Phaser (LFO Rate, Amount), Distortion (Tone, Drive), Bitcrusher (Bit Depth, Downsampling), Reverb (Damping, Decay), Delay (tempo sync, Feedback), 3-band EQ, Multimode Filter (mode, cutoff, resonance), with Dry/Wet.

Polyphony

Notes per voice from 1 (mono) to 32.

DAC Mod

Vintage (emulation of the converter of the era, with grain) or Modern (clean), per voice.

Advanced — Assign: map and controls

Zone map

Rectangles of all the voices on the keyboard with root note.

KEY (Key Assignment)

Moves the zone, adjusts lower and upper limits, creates crossfades between zones by dragging the white bar.

VEL (Velocity Assignment)

Limits the voice to a velocity range (e.g. 30–100) with crossfade: dynamic layers.

Root note

Note at which the sample plays at its original pitch (C-2–G8).

Controls (matrix)

Fixed sources Velocity, Keyboard, Aftertouch, Pitch wheel, Mod wheel to: VCA level and attack, Filter frequency and resonance, Pitch, LFO Rate, LFO → VCA, LFO → filter, Vibrato, Dry/Wet of the Voice FX, Dry/Wet and parameter of the Master FX; bipolar amount and buttons 1–8 to choose the voices involved. Pitch wheel with range 0–24 semitones.

Advanced — Effects (3 master slots)

Serial / Parallel

Effects in a chain or in parallel (summed).

Reverb

Reverb with pre-delay, decay, damping and mix.

Delay

Free-running or synced echo with feedback and filters.

Chorus

Chorus to widen strings and choirs.

Flanger

Modulated comb filter.

Phaser

Phase sweep.

Overdrive

Saturation.

Compressor

Dynamics control.

BitCrusher

Bit and sample-rate reduction to accentuate the lo-fi.

Multi Filter

Master multimode filter.

Param Eq

Parametric EQ.

Stereo Pan

Autopan with Amount and Rate (free or sync).

Bottom bar

Undo, Redo, Undo History

Management of edits.

CPU meter / Panic

CPU load; click to silence hanging notes.

Sound design tips

Filter and envelopes bring the sample to life

• The sample is static: it is the filter envelope with velocity that makes it expressive. For strings and winds use VEL on the filter and a filter attack slightly slower than the VCA's, so the sound opens after the bow or the breath.
• Strings and choirs sit well with a cutoff between 3 and 6 kHz and KBD at 30–60%: they remove the harshness of the aliasing at the top while keeping presence.
• High resonance on full samples thins them a lot: for basses and drums don't exceed 20–30%.

Loops

• Choose loop points in the stable sustain and use Loop fade (0.2–0.5) to hide timbre jumps; for sounds with recorded vibrato, make the loop last at least one full vibrato cycle.
• Loop in release on for pads and choirs with long tails; off for instruments that should end naturally.

Transposition and multisamples

• Beyond ±7–9 semitones from the root the timbre changes a lot (shifted formants): for realistic sounds place several samples side by side with zones and crossfades; for the '80s aesthetic transpose freely.
• Transposing up increases aliasing: lower the cutoff or enable negative KBD in the high zones.

Frequency management

• The vintage bandwidth ends around 13 kHz: in a modern mix pair it with an element that has air (shaker, modern pad) or use DAC Modern on layers that need to shine.
• Sampled basses and drums: high-pass below 30–40 Hz and control the 150–300 Hz zone, where the 8-bit sample accumulates mud.
• Choirs and strings occupy 300 Hz–4 kHz like the voice: keep them in the background with EQ or a higher register.

Expression

• LFO with Delay 300–500 ms and Variation for natural vibratos on flutes and strings; Mod Wheel → Vibrato for live control.
• Alternate > Pan on percussion and plucks creates movement without effects.
• Velocity layers (VEL Assignment) with crossfade for dynamic drums and pianos.

Common mistakes

• Long VCA release with low Polyphony: the tails get cut off by new notes.
• Linked voices (Voicelink) when you want to edit only one voice: unlink it or use the offsets.
• Forgetting the correct root note after loading an external sample.
• Loop without crossfade on noisy material: periodic clicks.

Presets to recreate

In genres and records

• Synthpop and new wave: Depeche Mode use Emulator I, II and III in the studio and live, including on the choir and strings of Enjoy the Silence; Neil Tennant states that every instrumental sound of West End Girls by the Pet Shop Boys comes from an Emulator II, including the Marcato Strings from the library. Users also include New Order, Orchestral Manoeuvres in the Dark, ABC, Simple Minds and Yes.
• Art pop and singer-songwriter pop: the shakuhachi from the E-mu library, resampled and layered with the Fairlight, is the flute of Sledgehammer by Peter Gabriel; David Bowie, Paul McCartney, Brian Wilson and Stevie Nicks among the users.
• New age and world chill-out: Enigma use the same shakuhachi on MCMXC a.D.; Enya and Kitaro among the users.
• EBM and industrial: Front 242.
• Electronic music and film scores: Tangerine Dream, Jean-Michel Jarre, Vangelis; Brad Fiedel uses it in the Terminator score, Michael Kamen in Lethal Weapon and Highlander, John Carpenter in nearly all his '80s films.
• Jazz-funk and R&B: Herbie Hancock and Stevie Wonder (who receives the first Emulator I produced).
• Progressive rock and AOR: Tony Banks (Genesis), Mr. Mister, Brent Mydland with the Grateful Dead.
• Hip hop, boom bap and lo-fi: the grain of the E-mu samplers.
• Synthwave, vaporwave and chillwave.