Every note your keyboard sends to a synthesizer, every chord your DAW plays back, is slightly wrong. Not by accident. By mathematical necessity.
That wrongness has a name: the Pythagorean comma. It is 23.46 cents of accumulated error, generated by a collision between two of the most natural intervals in Western music. Understanding it is the first step toward understanding why your major thirds beat, why string quartets drift sharp in sustained chords, and what microtuning tools actually fix.
The Problem: Stacking Perfect Fifths
A perfect fifth is the interval between two notes with a frequency ratio of exactly 3:2. It is arguably the most consonant interval after the octave, and it is the interval by which most traditional music theory builds scales. The Pythagorean scale is constructed by stacking 12 perfect fifths: C, G, D, A, E, B, F#, C#, G#, D#, A#, F, and back to C.
If you start on C=261.63 Hz and stack 12 perfect fifths, you arrive at a note 1,200 cents above the starting octave. Except you don't. You arrive at:
(3/2)^12 = 531441 / 524288 ≈ 129.746Seven octaves is exactly:
2^7 = 128The ratio between these two values is 129.746 / 128 = 1.01364, which is 23.46 cents sharper than seven octaves. That 23.46-cent gap is the Pythagorean comma. It represents the fact that the "C" you reach after 12 perfect fifths is not the same "C" you started from. They are almost the same pitch, but not quite.
This is not a problem of measurement precision. It is a fundamental property of the integers involved. Perfect fifths (3:2) and octaves (2:1) are built from incompatible prime factors: the former introduces factors of 3, the latter only factors of 2. No power of 3/2 will ever equal any power of 2 exactly. The comma is not rounding error. It is proof that pure intervals cannot close a chromatic circle.
Three Centuries of Solutions
Before equal temperament existed, keyboard makers had to decide what to do with this gap. Their choices led to radically different-sounding instruments.
Pythagorean tuning keeps all 12 fifths pure at 3:2 and simply stacks the comma onto one interval, usually the fifth between G# and D#. That interval, the so-called wolf fifth, measures roughly 678 cents instead of 702. It sounds genuinely terrible, like two slightly mismatched pitches fighting each other. Pythagorean tuning gives you perfect fifths everywhere except at the wolf, and major thirds that are consistently 21.5 cents sharp.
Quarter-comma meantone (circa 1523) takes a different approach. It prioritizes pure major thirds at the 5:4 ratio (386.31 cents), accepting worse fifths in exchange. To get pure major thirds, each fifth is narrowed by 1/4 of the syntonic comma (the gap between a Pythagorean major third and a just major third, equal to 21.51 cents). Each meantone fifth is therefore about 5.4 cents flat. This is tolerable in common keys, but the wolf fifth re-appears, now dramatically wider, and the keys farthest from C are effectively unplayable.
Well temperament, described by Andreas Werckmeister in his 1687 book Musikalische Temperatur, distributes the comma unevenly. Not all fifths are the same size. Keys close to C major get nearly pure fifths and consonant thirds. Keys with many sharps or flats get slightly stretched thirds and a brighter, more tense character. Every key is playable, but each sounds different. Werckmeister's tuning is now associated with J. S. Bach's Well-Tempered Clavier, though whether Bach used Werckmeister III specifically or some other well temperament remains debated among historically informed performance scholars.
The Modern Settlement: 12-TET
Equal temperament solves the problem by distributing the Pythagorean comma evenly across all 12 fifths. Each fifth is narrowed by exactly 1/12 of 23.46 cents, which is 1.955 cents. The result: every fifth measures exactly 700 cents (two semitones of 100 cents each), and every semitone is exactly the twelfth root of 2, an irrational number.
This guarantees that every key sounds identical, that transposition is completely lossless. The cost is that no interval except the octave is pure.
The trade-off for the major third is severe. A just major third is 5:4 = 386.31 cents. An equal-tempered major third is 400 cents, exactly four semitones. The difference is 13.69 cents sharp. That is far more audible than the 1.96-cent error in the fifths. A sustained equal-tempered major third between two sine waves beats at a rate proportional to that 13.7-cent deviation. At middle C, it produces roughly 7 beats per second on the major third alone.
This beating is not subtle. Organs tuned to equal temperament and played with sustained chords sound noticeably rough compared to instruments in meantone or just intonation. Piano music sounds fine largely because piano string attack masks the beating and the inharmonicity of piano strings interacts with the tempered thirds in ways that happen to soften the roughness.
Why Unfixed-Pitch Instruments Escape
String players, brass players, and singers do not have fixed-pitch constraints. Given a sustained chord, a well-trained string quartet will drift subtly toward just intonation: the violist narrows the major third without consciously deciding to do so, responding to the absence of beating. This is why string quartet recordings of sustained chords often measure slightly flatter or slightly sharper than the written 12-TET pitch, and why they sound more consonant than a keyboard playing the same notes.
The interaction between fixed-pitch and unfixed-pitch instruments creates audible tuning conflicts in performance. A sustained keyboard chord in C major, held pure at 12-TET, will beat with a string quartet that has locked onto just intonation. Neither player is "wrong," they are operating in different tuning systems.
Microtuning Tools for Synthesis
For sound designers working with synthesis, escaping equal temperament is now straightforward.
The Scala file format (.scl) is the de facto standard for describing alternative tuning systems. Created by Manuel Op de Coul, the format is a plain text file: one note per line, each specified as a cent value or a frequency ratio, with the interval of repetition as the final entry. Scala ships with a library of over 5,000 historical, theoretical, and experimental tunings, and the .scl format is supported by nearly every synthesis environment that claims microtuning support.
MTS-ESP (MIDI Tuning Standard - Extended Precision), developed by ODDSound, provides a real-time tuning protocol that lets a single master plugin simultaneously control the tuning of any connected plugins across a DAW session. Supported by synthesizers from U-He, Arturia, Xfer, and others, it allows you to load a just intonation scale or a historical temperament once and have every instrument retune instantly. No manual per-note pitch bend. No per-synth configuration.
Practical applications include:
- Designing drones or ambient pads where the beating of equal-tempered thirds is replaced by pure 5:4 intervals that sustain without roughness
- Historical keyboard emulation where Werckmeister III's key color variation is actually audible
- Non-Western scales (Arabic maqam, Indian srutis, gamelan pelog/slendro) that require intervals outside 12-TET
What This Changes
Equal temperament is not a universal truth. It is a specific solution to a specific problem, chosen for its property of equal transposition, accepted at the cost of slightly rough thirds. That trade was worth it for the past 300 years of Western keyboard music. For synthesis, where you control the oscillator ratios directly, it is an arbitrary constraint.
The Pythagorean comma is 23.46 cents of accumulated arithmetic. It does not go away. Every tuning system moves it somewhere, divides it, or hides it. Understanding where it lands in 12-TET, and how it landed elsewhere historically, gives you precise control over a dimension of sound that most production environments treat as a fixed default.