01 · AUDIO ENGINE

The 12-bit Engine

DMC60's audio engine reproduces the character of early sampling drum machines: the aliasing, quantization noise, and filter warmth of that era.

Sample Rate Conversion

Every sample loaded into DMC60 is converted to the native 26.04 kHz sample rate using Lagrange interpolation. A ~13 kHz anti-alias filter is applied during conversion, but the resulting aliasing artifacts are intentional and are part of the engine's character.

Internal rate26.04 kHz (Vintage); 48 / 96 kHz tiers available
Host rates44.1, 48, 88.2, 96, 176.4, 192 kHz
InterpolationLagrange
Anti-alias~13 kHz lowpass

Sample Fidelity & Playback Engine

Since 1.0.25 the engine is a choice. In Settings > Playback & Sampling, Sampling sets the resolution new imports and recordings are stored at, and Engine sets what every voice is played through. Each sample remembers its own resolution, so vintage and hi-res material mix freely in one kit, and switching either setting never repitches what is already loaded.

Sampling: Vintage12-bit / 26.04 kHz: the DMC60 sound (default)
Sampling: Hi-Fi24-bit / 48 kHz, 8th-order anti-alias filter
Sampling: Studio32-bit float / 96 kHz, no quantization
Engine: VintageHi-res voices pass through a 26.04 kHz hold, 12-bit quantizer and per-voice SSM2044 stage (default)
Engine: Hi-FiEvery voice plays at its native resolution

Material captured in Vintage is already the engine's output and passes through untouched in both engine modes, so existing kits and projects sound exactly as before. The Engine switch is live: use it to A/B a kit imported at Studio clean or crunchy. Drum synth and keygroup voices keep their own per-sound Vintage / Clean switches.

12-Bit Quantization

The signal path uses 12-bit quantization (−2048 to +2047), matching early hardware samplers where quiet sources break up and loud sources punch. Quantization is applied after sample rate conversion, as on the original hardware.

Bit depth12-bit
Value range−2048 to +2047
CharacterQuantization noise on quiet signals

SSM2044 Filter

Each voice runs through an SSM2044-style ladder filter: a 4-pole, 24 dB/octave lowpass with smooth resonance and self-oscillation capability.

Type4-pole ladder (24 dB/oct)
CutoffNormalized 0.0–1.0
ResonanceTo self-oscillation
Per-voiceYes

Voice Allocation

DMC60 provides 8-voice polyphony with intelligent voice stealing. When all voices are in use, the oldest active voice is stolen for the new note. Mute groups (choke groups) allow hi-hat behavior where triggering one sample silences others in the same group.

Polyphony8 voices
Voice stealingOldest-priority
Mute groupsPer-pad assignment
Choke behaviorSame-group retrigger silences

Per-Voice Processing

Each voice has its own processing chain applied in real-time:

Pitch/Tune±14 semitones
ADSR EnvelopeAttack, Decay, Sustain, Release
FilterSSM2044-style cutoff
DriveSaturation amount
PanStereo position (−1.0 to +1.0)
OutputMaster or individual (1–8)

Trigger Modes

Each pad can be set to one of two trigger modes:

  • One-Shot: The sample plays to completion regardless of note release. Standard behavior for drums and percussion.
  • Gate: The sample sustains while the key is held and triggers the release envelope on key-up. Useful for sustained sounds or expressive playing.