Atoms

The y-component of the ray direction provides a natural UV for sky gradient and cloud placement

There are exactly 17 distinct symmetry groups for periodically tiling the plane

There is no fixed boundary between point and plane — the distinction is relative to scale and context, judged by feeling

Three phase-offset cosine waves generate smooth procedural color palettes

Threshold-armed sampling records incoming audio only when it crosses a set level

tick advances a beat counter per live_loop and returns the current ring element; look reads without advancing

Tidal chop, striate, cut, and loopAt turn long samples into granular and looping textures

Tidal controls global tempo with setcps, and individual pattern speed with fast and slow functions

Tidal Cycles treats music as infinite cycles of patterns in a terse mini-notation manipulated live

Tidal effect parameters are control patterns, combined with sound patterns using the # operator

Tidal filters cutoff, hcutoff, and djf shape a sound's frequency content

Tidal functions can be partially applied (curried) to produce new functions that accept remaining arguments

Tidal mini-notation uses *, /, !, and <> to speed up, slow down, repeat, and rotate steps within a cycle

Tidal mini-notation uses | to randomly pick between subsequences, and ? to randomly drop events

Tidal oscillators with range modulate parameters as LFOs, smoothed via control busses

Tidal sometimes, degradeBy, irand, and rand add controlled randomness to patterns

Tidal transformations chain with the . operator to combine into one function

Tidal users without functional programming background can make music with it, showing DSL usability decouples from host language difficulty

Tidal's ? and degrade functions randomly remove pattern events to introduce controlled probabilistic variation

Tidal's continuous oscillator patterns (sine, saw, tri, square) modulate control values smoothly over time

Tidal's mini-notation parses polymetric rhythms from strings using square and curly bracket operators

Tidal's n selects a sample by index while note pitches a sample up or down

Tidal's rand emits a continuous stream of random floats 0-1 and irand n emits random integers 0 to n-1, both deterministic in time

Tidal's randomness is seeded by cycle count, so resetCycles restarts the same random sequence

Tidal's tempo is set in cycles per second (cps), not BPM; converting requires choosing beats per cycle

TidalCycles `chop N` divides each sample event into N equal, individually addressable slices

TidalCycles `while` applies a function only when a binary pattern is true, gating transformations per cycle

TidalCycles can pick one of several patterns at random each cycle to add variation

TidalCycles euclid k n places k onsets Euclideanly across n steps, the function form of the (k,n) mini-notation

TidalCycles is designed exclusively for live coding algorithmic patterns, with a mini-notation for rhythms and an extensive combinator library for pattern manipulation

TidalCycles is oriented around cycles rather than beats, so mixing fractional steps yields cross-rhythms easily

TidalCycles parenthesis notation (k,n) generates Euclidean rhythms directly in mini-notation

TidalCycles step builds a pattern from a step-sequencer string, mapping x to a hit and digits to sample indices

TidalCycles' pattern model originates from Indian tabla rhythm analysis via Bernard Bel's Bol Processor syntax

Tides accepts V/Oct pitch CV for musically-calibrated frequency tracking alongside exponential FM

Tides automatically adapts its behavior at audio range to keep waveshapes musical and prevent aliasing

Tides has four output modes that determine the relationship between its four simultaneous outputs

Tides SMOOTHNESS applies a low-pass filter (CCW) or wavefolder (CW) to shape waveform texture

Tides uses FREQUENCY for overall envelope speed and SLOPE for attack-to-decay ratio, not separate attack/decay knobs

Tides' CLOCK input locks its frequency to an external signal multiplied by the FREQUENCY ratio

Tighten timing to the groove of a chosen reference instrument, and don't neglect vocal timing

Tile and zoom crop and reposition a layer's texture without altering the source or the layer shape

Timbre is a multi-dimensional percept that requires a feature list rather than a single number to describe

Timbre is a perceptual quality; spectrum is its physical correlate—they are related but not equivalent

Time resolution and frequency resolution in windowed analysis are inversely constrained

Time-stretching and pitch-shifting are decoupled operations requiring phase vocoder or WSOLA rather than sample rate change

Timing correction should be referenced to the groove instrument and tightened track by track in order of rhythmic importance

Timing delays to the song's tempo makes them pulse with the music and become nearly imperceptible

Timing in animation is controlled by the number of frames between poses: more frames = slower, fewer = faster

To make software react to music playing on a computer you must route system audio back in as an input via a platform-specific loopback device

Todd Edwards pioneered treating individual vocal syllables as instruments by reversing, pitch-shifting, and chopping them rhythmically

Todd Terry brought a hip-hop sampling sensibility to house music and pioneered disco looping later adopted by Daft Punk

TorchScript (.ts) files are the portable, runtime-independent format for deploying PyTorch models without Python

TouchDesigner organises every node into six operator families named by data type: TOPs, CHOPs, SOPs, MATs, DATs, COMPs

Touching a battery-powered AM radio's exposed circuit board with damp fingers turns the radio into a synthesizer by adding your body as a variable resistor

Touchstrips offer space-saving continuous control with unipolar, bipolar, stepped, and pressure-sensitive modes

Tracker music originated in Amiga game culture and was shaped and popularised by the demoscene

Trance pads are built from detuned supersaw oscillators with 7–9 voices that filter-sweep open before drops

Trance tracks use sparse intros and outros designed for DJ blending

Transfer learning reuses a pre-trained network as a feature extractor to train a classifier from few examples