IDM / Aphex Chaos

Genre: IDM · Tempo: 172 BPM · Time signature: 4/4 · Length: 32 steps · Style: glitchy, probability-driven IDM programming in the Aphex Twin/Rephlex tradition

Open this pattern in the interactive Patchpad (with playback)

The pattern

Instrument	Steps (1 = start of the bar)	Velocity	Timing and chance
kick	1, 6, 11, 14, 17, 22, 27, 30	55-120	step 6: late by about 17% of a step, plays 65% of the time; step 11: early by about 25% of a step, plays 70% of the time; step 14: late by about 29% of a step, plays 40% of the time; step 22: early by about 17% of a step, plays 60% of the time; step 27: late by about 21% of a step, plays 75% of the time; step 30: early by about 25% of a step, plays 35% of the time
perc1	2, 12, 19, 28	48-55	step 2: late by about 12% of a step, plays 60% of the time; step 12: early by about 8% of a step, plays 55% of the time; step 19: early by about 17% of a step, plays 50% of the time; step 28: late by about 12% of a step, plays 60% of the time
hat-closed	3, 5, 7, 10, 13, 15, 18, 21, 23, 26, 29, 31	30-65	step 3: early by about 8% of a step, plays 70% of the time; step 5: late by about 4% of a step, plays 55% of the time; step 7: early by about 8% of a step, plays 60% of the time; step 10: late by about 8% of a step, plays 50% of the time; step 13: late by about 4% of a step, plays 90% of the time; step 15: early by about 12% of a step, plays 40% of the time; step 18: early by about 8% of a step, plays 55% of the time; step 21: late by about 8% of a step, plays 60% of the time; step 23: early by about 4% of a step, plays 65% of the time; step 26: late by about 8% of a step, plays 45% of the time; step 29: early by about 4% of a step, plays 85% of the time; step 31: late by about 17% of a step, plays 35% of the time
snare	4, 9, 20, 25	45-105	step 4: late by about 21% of a step, plays 35% of the time; step 9: plays 80% of the time; step 20: late by about 25% of a step, plays 30% of the time; step 25: plays 85% of the time
fx	8, 16, 24, 32	60-75	step 8: early by about 12% of a step, plays 50% of the time; step 16: late by about 25% of a step, plays 45% of the time; step 24: late by about 12% of a step, plays 55% of the time; step 32: early by about 21% of a step, plays 40% of the time

Digitakt II

Tracks

• T1: kick
• T2: perc1
• T3: hat-closed
• T4: snare
• T5: fx

Settings

• Every step's TRIG PARAMETERS microTiming and PROB value must be entered exactly as written in the steps array (e.g. step 11 kick: microTiming -6, PROB 70%; step 14 kick: microTiming +7, PROB 40%) — this pattern leans entirely on aggressive per-step micro-timing (up to ±7 units, near the hardware's ±23 limit) and probability, enter every value by hand rather than copying a template and nudging it
• Step 8 and step 24 (fx, T5): SRC page SAMP p-locked to a different sample slot than steps 16/32's fx hits, so the two fx hit-types sound like different events rather than one repeated one-shot
• Kick (T1): 'strong' hits (steps 1, 17, PROB 100%) left at base SRC page TUNE; scattered ghost hits (steps 6, 11, 14, 22, 27, 30, PROB 35-75%) p-locked TUNE down around -6 to -10 so weak hits read as debris rather than a second kick voice
• Re-randomize a handful of TRIG PARAMETERS PROB values by hand each time this pattern is looped live (stay within roughly ±15 of each step's written value) for a pattern that visibly mutates rather than repeats exactly

Digitone II

Tracks

• T1: kick (ALGO 8, FM click-into-boom)
• T2: perc1 (ALGO 1 plain carrier)
• T3: hat-closed (high-FDBK FM noise)
• T4: snare (moderate-FDBK FM noise)
• T5: fx (ALGO 5/6, contrasting per hit-type)

Settings

• Every step's TRIG PARAMETERS microTiming and PROB value must be entered exactly as written in the steps array (e.g. step 11 kick: microTiming -6, PROB 70%; step 14 kick: microTiming +7, PROB 40%) — this pattern leans entirely on aggressive per-step micro-timing (up to ±7 units, near the hardware's ±23 limit) and probability, enter every value by hand rather than copying a template and nudging it
• Kick (T1): 'strong' hits (steps 1, 17, PROB 100%) left at base TRIG page NOTE / SYN page 1 RATIO C 0.30; scattered ghost hits (steps 6, 11, 14, 22, 27, 30, PROB 35-75%) p-locked TRIG page NOTE down (roughly -7 to -10 semitones) so weak hits read as debris rather than a second kick voice. Base sound: SYN page 1 ALGO 8, SYN page 2 Operator A DEC 15, LEV 105
• Step 8 and step 24 (fx, T5): SYN page 1 ALGO 5, RATIO A 2.00 p-locked; steps 16 and 32's fx hits instead use ALGO 6, RATIO A 5.00 — two contrasting FM algorithms so the pattern's two fx hit-types sound like different events, not one repeated one-shot
• Hat-closed (T3) base sound: SYN page 1 FDBK 90, FLTR page 1 TYPE toward HP (~85), AMP page DEC 20
• Snare (T4) base sound: SYN page 1 FDBK 85, AMP page DEC 45
• Re-randomize a handful of TRIG PARAMETERS PROB values by hand each time this pattern is looped live (stay within roughly ±15 of each step's written value) for a pattern that visibly mutates rather than repeats exactly

Syntakt

Tracks

• T9: kick (BD FM)
• T10: snare (SD FM)
• T11: perc1 (UT NOISE GEN)
• T12: hat-closed (CH METALLIC)
• T1: fx (SY RAW)

Settings

• Every step's TRIG PARAMETERS microTiming and PROB must be entered exactly as written (e.g. step 11 kick: microTiming -6, PROB 70%; step 14 kick: microTiming +7, PROB 40%) — hand-enter each value individually, this is the one pattern in the library where that precision matters most
• Kick (T9, BD FM): 'strong' hits (steps 1, 17) left at base SYN page TUNE; scattered ghost hits (steps 6, 11, 14, 22, 27, 30) p-locked TUNE down so they read as debris rather than a second kick voice
• Step 8 and step 24 (fx, T1, SY RAW): WAV1/WAV2 p-locked to a different waveform pair than steps 16/32's fx hits, so the two fx hit-types read as distinct glitch events
• Re-randomize a handful of TRIG PARAMETERS PROB values by hand each time this pattern loops live (stay within roughly ±15 of each step's written value) for a pattern that visibly mutates rather than repeats exactly

Notes

Deliberately irregular: no repeating subdivision, heavy use of probability (30-90%) so hits drop unpredictably, and micro-timing pushed to ±7 (near the hardware's ±23 µtime limit) so nothing sits cleanly on the grid. Two bars are written with related-but-mutated placement rather than an exact repeat, in the spirit of Aphex Twin/Rephlex-style IDM programming. An original pattern, not a transcription.

Patchpad is a free reference of drum patterns and synth patches written for real hardware: Digitakt II, Digitone II, Syntakt, Novation Summit, Arturia MiniFreak, Dreadbox Typhon and ASM Hydrasynth Desktop.