Audio

Machines

Sequencing model

Two kinds of command matter to a sequencer. A latched command is written once and the hardware holds or evolves it on its own; a driven command must be refreshed every tick by the CPU or the player routine. The tables below use A for automatic (latched) and D for driven, with a dash where the machine has no such feature. Most register chips latch their tone, volume and envelope; the 1-bit outputs, and any effect a chip has no hardware for, must be driven. This split decides whether the editor emits a single event or a per-tick automation lane.

Channel capabilities

Each capability below is a setting a channel can expose. The section gives what it is, then a table with a column for each of the capability's parameters and a row for each machine that supports it, showing that machine's range. A machine that lacks the capability is left out of the table. The per-machine hardware detail (registers, bit widths, addresses) lives on each platform's own audio page.

Pitch

Sets the fundamental frequency of the voice; the one control every tonal channel has.

Volume

Sets channel loudness, for dynamics and mix balance.

Waveform

Chooses the tone shape of the voice; where more than one shape or a drawable shape is offered.

Duty (pulse width)

The high-to-low ratio of a pulse wave; narrow duty is thin and nasal, 50% is hollow.

Amplitude envelope

Shapes a note's volume over time, from a hard pluck to a slow swell.

Hardware pitch sweep

The chip slides the pitch on its own once armed; used for zaps, whoops and rising alarms.

Noise

A pseudo-random source for drums, cymbals, wind and explosions.

Ring modulation

Multiplies two oscillators for clangorous, bell-like, inharmonic tones.

Hard sync

One oscillator resets another for an aggressive, tearing sync lead.

Filter

Shapes the harmonics for muffling, wah and squelchy resonant sweeps.

Wavetable

Plays a short user-drawn cyclic waveform, for any static timbre you draw.

Sample / PCM playback

Replays recorded audio as the voice; the native sample voices, distinct from forced playback.

FM synthesis

One operator modulates another's frequency for rich, evolving, metallic tones.

Panning

Places the voice in the stereo field for width and movement.

Pitch modulation

A voice frequency-modulates the next for gritty, FM-flavoured sample textures.

Echo and reverb

A delayed, filtered feedback for space and ambience.

Player effects (software)

These have no dedicated hardware: the player produces them by updating a channel every tick, so they are available on any driven channel (the volume-based ones need a volume control). They are the portable effect core the sequencer can offer everywhere.

Command layers

A sequencer can lean on a small core that every target shares, then add capability-gated commands per machine. The core is what makes one editor drive them all; the extensions are what make each machine sound like itself.

MIDI and MOD sit on top of this model. MIDI is the core in protocol form: the sequence sends events and controller streams, and the target synth realises the envelopes. A MOD maps a note-and-sample trigger, a one-shot, against an effect column driven every tick, which is the same latched-versus-driven split the machines show, expressed in one portable file.