What is Music Anyway?

A note is a number

Sound is air pressure wiggling back and forth. What you perceive as pitch is simply how many wiggles arrive per second — the frequency, measured in hertz (Hz). Slide the frequency below and listen: nothing about a "note" exists yet, just a continuous dial.

Now the first miracle. Play 220, then 440, then 880. They're obviously different pitches — yet they sound like the same note. Doubling a frequency produces this "sameness," and every musical culture on Earth discovered it independently. We call the doubling an octave, and it's the anchor everything else hangs on.

Why some ratios sound good: the harmonic series

Why does ×2 sound "the same"? Because real instruments never produce one frequency. A string of length L also vibrates in halves, thirds, quarters… simultaneously. So a string tuned to 110 Hz actually emits 110, 220, 330, 440, 550… — every whole-number multiple. These are the harmonics, and their recipe is what makes a violin sound different from a flute.

Here's the payoff: play two notes whose frequencies form a small whole-number ratio, and their harmonic ladders overlap. A 2:1 pair shares every other rung — that's why octaves fuse. A 3:2 pair shares every third rung. The fewer shared rungs, the rougher it sounds. Consonance isn't taste; it's arithmetic your ear performs.

How an interval is actually calculated

An interval is the distance between two pitches — but crucially, it's a distance you multiply, not add. A fifth above 200 Hz is 300 Hz (+100); a fifth above 400 Hz is 600 Hz (+200). Same interval, different hertz gap, same ratio. So pitch space is logarithmic, and modern tuning defines it in one line:

Each semitone multiplies frequency by 2^(1/12) ≈ 1.05946. Twelve of them multiply to exactly 2 — one octave. For finer measurement each semitone is split into 100 cents: cents = 1200·log₂(f₂/f₁). Try it — pick a root on the keyboard, then an interval:

Notice the last two cells above: equal temperament's ratios are irrational numbers that land eerily close to the pure whole-number ratios of the harmonic series. That closeness is not luck — it's the entire reason for the number 12, coming up next. The full ledger:

The devil's interval: tension → resolution

Look at the worst row in that table: n = 6, the tritone — ratio √2, the octave split exactly in half, with no small-number ratio to justify it. Medieval theorists called it diabolus in musica. But its instability is fuel: those two notes are desperate to move, and they can collapse inward or spring outward — the same pair resolving into two different keys a tritone apart. That double identity is the engine inside every dominant seventh chord, and the basis of jazz's tritone substitution.

In the G7 → C cadence, listen for the tritone hidden inside the first chord (B and F) doing the pulling: B slides up a half-step to C, F slides down a half-step to E. That "the song has to land now" feeling is this one interval demanding resolution — four centuries of Western harmony run on it.

Twelve, seven, and C — the booth version

The derivations live on the theory path. Here is what survives contact with a Friday night, then straight on to the wheel.

Why twelve? Stack fifths and watch

After the octave (2:1), the strongest consonance is the fifth (3:2). So here's the natural experiment every ancient tuning theorist ran: start on a note, keep multiplying by 3/2 (folding back into one octave whenever you overshoot ×2), and see if you ever return home. If you do after N steps, you've carved the octave into N notes where fifths work everywhere.

Twelve steps later you land almost exactly where you began: (3/2)¹² = 129.75, while 7 octaves is 2⁷ = 128. The overshoot — about 1.4%, or 23.5 cents — is the famous Pythagorean comma. No stack of pure fifths ever closes perfectly (powers of 3 can't equal powers of 2), but 12 comes astonishingly close. Equal temperament's fix: shave each fifth by 2 cents so the circle closes exactly.

Could other numbers work? Mathematically this is asking for good rational approximations of log₂(3/2) = 0.58496…, and the continued-fraction convergents are 1/2, 3/5, 7/12, 31/53… Twelve is the sweet spot: the smallest division that nails the fifth and lands a usable major third, with few enough notes to fit under ten fingers.

The price of the compromise — hear it

Twelve's weak spot is the major third: 400¢ vs the pure 5:4 at 386¢ — 14 cents sharp. When the harmonics misalign by a few hertz they interfere, producing a slow shimmer called beating. Compare:

Every major third you've ever heard on a piano beats like this. We collectively agreed the trade was worth it: slightly restless thirds in exchange for the freedom to play in all 12 keys and modulate between them — the deal that unlocked Bach's Well-Tempered Clavier and everything after.

Why only seven letters?

Twelve notes exist, but melodies rarely use all of them at once. Take the same fifth-stacking machine and stop at seven notes — the previous convergent of that continued fraction. Start from F:

Sorted into one octave, those seven fifths-related notes form a scale with a beautiful property: only two step sizes — whole steps (W) and half steps (H) — in the pattern W·W·H·W·W·W·H. Maximally even, no two half-steps adjacent, and every note reachable by fifths. This is the diatonic scale, and it's the seven-note core that medieval musicians named with letters A through G.

The five leftover notes (the black keys) arrived later, as alterations of the seven — which is why they never got their own letters and are spelled as sharps and flats: C♯ is "C, raised."

So why does everything start on C?

Pure historical accident — the one piece of this that isn't math. Around 500 CE, Boethius labeled the notes alphabetically starting from the lowest note in use: A. Perfectly logical. But medieval music was built on modes, and over the following thousand years, European taste drifted toward one particular pattern — the mode that happens to begin on the third letter.

By the 1600s that pattern (the Ionian mode) had become so dominant we renamed it "the major scale." The letters were never re-dealt. So the white keys got named for an A-based world, and we now live in a C-based one. Hear what changed taste — same seven white keys, different starting note:

The Camelot wheel: the circle of fifths, rebranded

Everything above compresses into one DJ tool. Take the 12 major keys and arrange them so each clockwise step goes up a perfect fifth — the same ×3/2 move from section 04 — then number the slots like a clock. That's the outer B ring. Beside each major key, on the inner A ring with the same number, sits its relative minor: the scale from section 06 that uses the exact same seven notes but starts three semitones lower. That's the entire invention. Neighbouring slots share 6 of their 7 notes, so blending between them can't clash hard — Mixed In Key just replaced "up a fifth" with "+1" so nobody has to think about it at 2 a.m.

The scale formula behind each slot

Every B slot is a major scale, every A slot a natural minor — the two step-recipes from sections 05–06, applied to a different root:

Converting any key to its wheel number is pure modular arithmetic, because each semitone of root movement equals 7 steps around a circle of fifths:

Mixing rules — check any two keys

Route finder: bridge any two keys

Need to travel from a warm-up key to a distant climax key? A breadth-first search over the safe moves (+1, −1, relative swap, +2, +7) finds the shortest smooth pathway — every hop is a clean mix:

Setlist lab: grade a whole set

Sequence real tracks and every consecutive transition gets analyzed automatically — the wheel's rules applied to your actual crate, with a harmonic score and a booth tip per hop. The list persists in your browser.

Why +2 is only a mild risk while +7 is a leap: the wheel measures distance in fifths, and two scales share 7 − (fifth-distance) notes. +2 is two fifths away (5 shared); +7 is five fifths the short way round (2 shared) — but its root moves just one semitone, which is exactly what makes it feel like a gear change rather than a key change.

Rhythm is slow pitch

Tempo and pitch are the same phenomenon at different speeds. BPM is just frequency measured per minute instead of per second — divide by 60 and a tempo is a hertz value. Your brain draws a border around 16–20 Hz: below it, you count events (rhythm); above it, the events fuse into tone (pitch). Prove it to your own ears:

The DJ bridge: varispeed

Change playback speed and pitch and tempo move together, locked by semitones = 12·log₂(rate). The magic constant every turntablist knows by feel: +5.9% ≈ +1 semitone — which on the Camelot wheel means +7 slots. Nudge the fader and watch your harmonic-mix key silently drift; keylock is the algorithm that severs this natural bond.

Polyrhythms are intervals

Since intervals are frequency ratios, tempo ratios map onto them exactly. Three beats against two is literally a perfect fifth slowed down ~6 octaves — and the lurching feel of 7-against-5 is the tritone's roughness, experienced as rhythm. Consonance and groove are one theory:

Time signatures: how pulses get grouped

A time signature is two numbers: the top counts how many pulses per bar, the bottom says which note value carries the pulse — and note values (whole, half, quarter, eighth, sixteenth) halve each time, so the denominators are powers of 2. Note durations are literally octaves in time. The top number is where meter gets interesting: the same six pulses accented 3+3 (6/8) or 2+2+2 (3/4) are different meters — grouping, not speed, is the information:

The final bridge: Euclidean rhythms are scales

Here's where rhythm theory and everything from sections 04–05 collapse into one construction. Take k hits and spread them over n time slots as evenly as possible. This single recipe generates the world's great rhythms — the Cuban tresillo is E(3,8), the cinquillo E(5,8), the bossa nova E(5,16). And when the gaps come out in exactly two sizes, you get the rhythmic version of the diatonic property. Now set it to E(7,12): the hit pattern's gaps read 2·2·2·1·2·2·1 — the diatonic scale as a drum pattern. Same necklace, one domain in time, one in pitch:

Press E(7,12), play it as a rhythm, then play it as a scale. That's the deepest fact on this page: "spread things maximally evenly around a cycle" is one theorem, and melody and groove are two of its outputs. E(5,12) gives the pentatonic — the black keys — the same way.

The synthesizer: this whole page with knobs on it

A synth is nothing mystical — it's sections 01–02 made adjustable. Subtractive synthesis, the architecture behind most classic synths, runs one philosophy: start with a wave that's too rich in harmonics, then carve away. The signal chain is three blocks, and every knob on every synth ever made belongs to one of them:

The four primitive waveforms

Each basic shape is a fixed harmonic recipe — compare these spectra with the harmonic mixer in section 02. Brightness is just harmonic count:

Build a patch: the working mini-synth

Everything below is live. Pick a preset, play the keys, then break it: sweep the cutoff, crank the resonance, stretch the attack. Every famous synth sound is some setting of exactly these controls.

Look at the LFO rate range: 0.3–18 Hz. That's the rhythm zone from section 08 — below the 20 Hz border, an oscillator stops being a tone and becomes a gesture. Vibrato, tremolo, and dubstep wobble are all just "notes" too slow to hear as pitch, played onto another parameter. One continuum, three uses.

Sound engineering: decibels & distortion

Loudness perception is logarithmic too — the same Weber-Fechner law that gave pitch its cents gives amplitude its decibels: dB = 20·log₁₀(A₂/A₁). Equal ratios sound like equal steps, so halving amplitude is always the same perceived drop (−6 dB), whether you're loud or quiet. Listen — each button halves the amplitude, and the steps sound even:

And the cardinal sin of gain staging — clipping — is secretly a return to section 02. Push a signal past the ceiling and the peaks get flattened; a flattened sine starts looking like a square, which means the distortion literally manufactures odd harmonics that were never played. Overdrive isn't "noise" — it's involuntary additive synthesis. That's why it can sound musical (guitar amps) or ruin a mix (accidental clipping):

What "brass," "pad," and "keys" actually mean

Open any preset browser and the names come from two different taxonomies, silently mixed. Some are acoustic families — brass, woodwind, strings, organ, keys — named after the physical instrument being imitated. Others are roles — pad, lead, pluck, stab, bass — that describe a job in the arrangement, not an instrument: anything with a slow attack and a long release becomes "a pad." Every one of them is a point in the same three-axis space you built in section 09:

The Bell card is the section's hidden lesson: its partials sit at 1 · 2.76 · 5.40 · 8.93 — not whole numbers. Everything else on this page fused into a clean pitch because its overtones were integer multiples (section 02); bells break that rule, which is why their pitch sounds ambiguous, why they shimmer, and why tuning percussion is an art. Harmonicity was never guaranteed — strings and air columns just happen to enforce it.

Once you read patch names this way, production browsing changes: "warm analog brass pad" decodes to saw spectrum, bloomed filter, pad envelope, detune motion — four settings, not a mystery. And the family names are honest physics: what makes a real trumpet a trumpet is exactly what makes the synth patch one — the recipe, not the metal.

The whole system in eight lines