56 projects
The pump of house is a compressor on the pad keyed from the kick, and the release knob is the sound. Measured off the pad's own level: a kick every 469 ms pushes it down 16 dB, and with a 150 ms release it's more than 3 dB down for 82% of every beat, back in 378 ms. At 50 ms that's 51% and 220 ms, a dent. At 400 ms it never gets back, a quieter pad. The worklet compressor rendered offline agrees. Twelfth of twelve.
The M1 organ bass under half of house is a Hammond in software: drawbars that hold, a percussion harmonic that comes in 17 dB above them and decays in 150 ms, and a click of 31 dB of high frequency in the first five milliseconds. All three read off the sound, the percussion to within 10 ms of its setting. It fires only on a note played alone, so a legato phrase gets it once, and that's the feel.
A 909 snare is two tones at 180 and 330 Hz and a burst of highpassed noise, and the tones decay faster than the noise, so every hit stops being a drum and becomes a hiss at some moment. Measured off the sound, that moment is 84 ms at half snappy, 203 at a quarter, and never arrives at full because the noise leads from the first millisecond. The two decays predict 91 and 202. Play it, or let it play the two and four.
A clap is four bursts of noise 10 ms apart and a tail. A drum machine gates one noise source, so every burst is fresh. A sampler layers one burst on itself, so every burst is the same noise, and a sound that repeats every 10 ms wears a comb of peaks every 100 Hz. Measured: the copied clap repeats exactly and its comb is 23 dB deep, the fresh one never comes closer than a third of the way and has no comb at all. Same envelope, same ear, different spectrum.
A 909 hat is six square waves at 205, 254, 370, 304, 523 and 800 Hz, frequencies picked to share no period. Measured by sliding the sound against itself, the closest it ever comes to repeating is 49% different. Snap the same six onto a 50 Hz grid and it repeats exactly every 20 ms, and a tone appears where the metal was. Play it, tune it, or let it play house hats.
A 303 slide is a lag on the pitch in semitones, so it takes the same time whatever the interval. Measured off the sound, a semitone and an octave both get 63% of the way in 59.8 ms and 95% in 180 ms, and an octave covers twelve times the ground in that time rather than taking twelve times as long. The accent opens the filter 1.4 octaves and adds 3.5 dB, and leaving the level out makes an accented note measure quieter than a plain one.
Pulse width does not filter, it deletes. A rectangle of width d has harmonic h at four over pi h times the sine of pi h d, so harmonic h is exactly silent whenever d is a multiple of one over h. Measured on a real rectangle against that formula, every harmonic lands within a thousandth of a dB, and sweeping the width takes each harmonic to silence exactly floor(h/2) times between nothing and a half. Play it, or let it play rave stabs.
Seven sawtooths spread across a detune, and the seven lines around every harmonic get h times further apart as you go up. Measured on A3 at 40 cents, the fundamental swells 53 dB and the beating dies at harmonic 36, where the voices are 183 Hz apart, against 43 predicted from where the clusters touch. At 80 cents it dies at 18 and at 10 cents it never does. That's why a supersaw beats at the bottom and hisses at the top.
Two sines, one modulating the other, and the spectrum is a Bessel function. Measured against the arithmetic at every ratio and index the page offers, the worst sideband is 0.18 dB out. At an index of 2.405 the first zero of J0 takes the note out of its own sound, 80 dB down while the sound stays loud, and a decaying index passes through that null 62 ms into every note, predicted at 61. At a ratio of 1 or 2 the folded sidebands land back on the note and the null only half happens.
A wobble is a saw through a resonant lowpass swept by a tempo locked LFO, and it doesn't sweep, it steps. The filter peak can only land on a harmonic of the note, so on a 55 Hz saw it climbs a staircase of 32 harmonics from the third to the 45th, a fourth wide at the bottom and 0.39 of a semitone at the top, going round 4.6 times a second on an eighth note at 140. Sharp resonance lifts each step 18 dB, soft lifts it 6. Play it, or let it play a bassline.
The reese is two saws a few cents apart, and every harmonic beats at its own multiple of the detune. At 15 cents on A1 the fundamentals drift round each other every 2.09 s and the eighth harmonic beats 3.83 times a second, both read off the sound. The lowpass decides the fastest beat you hear: 5 Hz closed, 12 Hz open, and at 30 cents open it passes 15 Hz and the growl becomes grit. Play it, or let it play a bar of drum and bass.
The 808 is one sine with a pitch envelope. It starts at 147 Hz and lands on 49 in 155 ms, and the tail takes 2.4 s to fall 60 dB. A phone speaker plays 1% of it, and 84% of that is the first 50 ms. Clipping it hard buys the phone under 3 dB. Playing it an octave up buys 10. Play it on the keys, or let it play itself.
The lo-fi knobs, measured. Eight bits puts the noise floor 48.9 dB down, a tenth off the textbook figure, and turns a note at -40 dB into 27% distortion. Four bits gives 25 dB and that quiet note comes out as nothing at all. Dither brings it back as a tone in hiss for 4.9 dB of noise, and holding samples at 6 kHz grows a 1 kHz tone two images at 5 and 7 kHz, -14 and -17 dB, where the sinc says they should be.
Most systems play the bass in mono, and a bass widened for headphones pays for it there. A 10 ms delay on one side reads as out of phase and takes 16 dB off a 55 Hz bass in the sum. A detuned pair pumps in mono every 1.6 s, 54 dB deep. Sum and difference combs are as wide as either and lose nothing at all, measured.
A chord's shimmer is its tuning. In equal temperament the third of a C major is 13.7 cents sharp of 5:4, so the root's fifth harmonic and the third's fourth miss by 10.4 Hz and beat that fast, measured off the sound, swelling 15 dB. In just tuning the same pair sits dead still. A minor third beats faster, and a 250 ms stab still holds two and a half swells of it.
Distortion is a curve, and everything people say about warm and harsh is a claim about its shape. Measured at the same distortion, a soft clip's harmonics fall 39 dB an octave and a hard clip's fall 12. Even harmonics need an asymmetric curve, and on two tones they cost a note two octaves down that nobody played, at -29 dB.
Noise has a colour, and the colour is a number: white is flat, pink falls 3 dB an octave, brown falls 6, measured. A hi-hat is what noise is for, and the 808's is six square waves chosen to share no harmonic. How little pitch that leaves is a number too, 45% for the six and 16% for the finished hat.
Sidechain pumping, the bass ducking under every kick, is done for loudness. Measured at the same peak, it buys about half a loudness unit at best, costs 1.5 without lookahead, and a loudness meter shows 2.4 LU of a 15 dB duck. The last part of the series, and the one where the reason turns out to be the groove.
A pulse wave has one knob, its width, and the width decides which harmonics do not exist. At half it is a square and every even harmonic is gone. At a third, every third. At a quarter, every fourth. Measured, exactly, and then the naive digital square measured against the exact one.
A saw through a resonant lowpass with the cutoff swept by an LFO at the tempo. Sweep it in hertz and the wobble sounds open 80% of the time, because the ear hears octaves. And however you sweep it, the resonance cannot keep up, 3 dB short at half notes and 6 dB short at sixteenths.
Seven saws spread across thirty cents is the sound of every trance lead. Everyone argues about how many voices. Measured, the movement is about 2 dB whether you stack two or sixteen, and what more voices buy is that the shimmer never repeats, but only if the spacing is uneven. Evenly spaced, seven voices loop every 3.1 seconds.
A phone speaker hears one per cent of a clean 808, and 93% of that is the first 60 milliseconds. Saturation does not bring the energy back, it moves it later into the body of the note. Drive makes a sub last on a phone. It does not make it louder.
Two saws seven cents apart, into a lowpass, into saturation. The bass under most of jungle, built from the last four parts. Saturation flattens its swell from 6.4 dB to under 1, and cannot stop the sub vanishing 60 dB down every four and a half seconds. That is why there is a sine under every Reese.
Two saws seven cents apart is the Reese, the bass under most of jungle. Detune them and they beat, but a saw is a stack of harmonics and the tenth one beats ten times faster than the first. One knob, and every harmonic moves at its own speed.
A pure 55 Hz sine has one frequency in it, and it still clicks when the note starts. How much depends on where the wave was at that instant, by 19 dB. The click is the envelope, and the attack is a filter on it.
The cutoff dial on a synth is only telling the truth at one setting of the resonance dial. Turn the resonance to 2 and a cutoff reading 1,000 Hz cuts at 1,485. Measured off the filter, not the diagram.
A sawtooth is not a thing a synth has, it is a stack of sine waves each 1/n as loud as the first. Build it one sine at a time and listen to it turn into the sound every bass patch starts from.
Eight copies of the same system, started a billionth apart. They move as one and then they do not, and the moment they stop agreeing is a number. A thousand times more precision buys 2.6 more seconds.
How long a glass rings and how narrowly it listens are the same number. Measure it both ways and they agree, unless you sweep past it too fast, and then one of them is 64% wrong.
Wobble a note's pitch fast enough and it stops being a wobble and becomes a chord. Turn the depth to 2.405 and the note you started with is 132 dB down, while everything around it carries on.
A dropped ball never quite stops bouncing, and stops after 4.81 seconds. The rattle at the end is the tail of a series you can hear converging.
A sheet of magnets is either lined up or it is noise, except at one temperature, where it fills with patches of every size at once. Onsager worked that temperature out exactly in 1944 and the sheet finds it back.
Two pendulums joined by a weak spring. Start one and it stops dead a minute later with the other one swinging, then takes it all back. The clock is the difference between two frequencies a hundredth of a hertz apart.
Open a tap slowly and the drips are even. Open it more and they pair up, then go to four, then eight, and the doublings arrive faster each time at a rate that is the same in every system that does this.
Sound leaves one focus of an ellipse in every direction, bounces once, and every single copy arrives at the other focus having travelled exactly the same distance. Step five centimetres off it and half the effect is gone.
Two speakers playing the same note fill a room with fixed lines of silence. Pace out the gaps between them and you have measured the speed of sound without timing anything.
Air will not stay attached to the back of a wire. It peels off alternate sides, kicks the wire as it goes, and the rate it does that at barely changes across four decades of flow.
Drag a source about and its sound leaves from where it was, not where it is. Push it past the speed of sound and the waves pile onto a cone that nothing in the code draws.
Clap, hear the echo, click the distance. You will be within about five per cent in a minute, and completely lost outside a band three and a half doublings wide.
Two voices in different places are far easier to separate than two in the same place. The help has a size in decibels, and at this pitch almost none of it comes from your head being in the way.
A browser will hand you a 96 kHz audio context and let you synthesise a 35 kHz dog whistle without complaint. Three ceilings stand between that and a dog, and the code only clears one of them.
A shooting gallery for your ears. Switch off one cue at a time and the score tells you what that cue was worth, in degrees.
A bowed string will only speak inside a wedge of bow force and bow position. Two boundaries, two different power laws, and the gap between them closes near the bridge.
A clarinet with no oscillator in it. A tube and a flap of cane, and the note is what the loop between them settles on.
A loud sound hides quiet ones for a fifth of a second afterwards, and for a few milliseconds beforehand.
A buffer of noise and a two-sample average. That is a plucked string, and it cannot be tuned properly, for a reason you cannot fix.
Four algorithms doing the same job, with their comparison counts checked against the theory that predicts them.
High, low, high, rest. Either a galloping rhythm or two separate beeping lines, and near the boundary, both.
A loud tone does not make its quiet neighbour quieter. It makes it not exist. This is why MP3 works.
Add a sound to a copy of itself a millisecond late and specific frequencies vanish, at places you can calculate exactly.
Two tones, almost the same. Twenty questions later the page knows your pitch threshold to within about ten percent.
Play a chord of upper harmonics with no root, and your ear invents the root anyway.
Two steady tones, one per ear. You hear a pulse that exists in neither of them.
A scale that rises forever and never gets higher, drawn so you can see exactly how it cheats.
Sand on a vibrating plate draws the shape of a standing wave. Sweep for a resonance and watch it assemble.
One oscillator, one filter, one LFO, and the whole wobble-bass sound falls out.