Topic · Analysis

The Riemann Hypothesis

Every non-trivial zero of ζ lies on a single vertical line.

Riemann's 1859 memoir analytically continued a humble sum of reciprocals to the whole complex plane, then noticed that the places where the result vanishes seem to all share one real part: ½. A proof would lock down the distribution of prime numbers tighter than any bound we have. A counter-example would topple a century and a half of consequences. None has been found.

Discover the story → Open the Explorer

First encounter

A function. A line. A century-old silence.

The conjecture

Take the function ζ(s) = 1 + 1/2ˢ + 1/3ˢ + 1/4ˢ + …, extend it (analytically) to all complex s, and ask where it equals zero. Some zeros are trivial — the negative even integers. All the others, says Riemann, lie on the vertical line Re(s) = ½. Trillions of them have been checked. Not one has missed.

A concrete example

The first non-trivial zero sits at s = ½ + 14.13… · i. The next at ½ + 21.02… · i, then ½ + 25.01… · i, marching upward forever. Each one is a tiny act of resonance between the integers — the heights are not random, they are the spectrum of how the primes deviate from average.

Why it matters

Hundreds of theorems are written in the conditional form 'assuming RH, then…'. A proof would activate them all simultaneously: tighter prime gaps, sharper error terms in number theory, hardness guarantees in cryptography, and a deeper bridge between analysis and arithmetic. It is Hilbert's eighth problem and one of the seven Clay Millennium Prizes.

Below: walk the function ζ along the critical line and watch it loop through the origin at every zero. The Explorer scales the same picture up.

Section 01 · Euler's product

ζ has the primes baked into its DNA

In 1737, almost a century before Riemann, Euler noticed that the infinite sum ζ(s) = Σ 1/nˢ can be rewritten as an infinite product over primes: ζ(s) = ∏ₚ 1 / (1 − p⁻ˢ). The equivalence is one of the most quietly stunning identities in mathematics — the additive structure of the integers and the multiplicative structure of the primes are literally the same object, viewed from two sides. Once you know that, the question 'where does ζ vanish?' is no longer a question about a function; it is a question about the primes themselves.

Section 02 · Analytic continuation

Extending a sum that diverges

The series Σ 1/nˢ only converges when Re(s) > 1. Riemann's 1859 paper performs a single, audacious move: it shows that ζ has a unique extension to the entire complex plane (apart from a single pole at s = 1), and that this extension satisfies the symmetric functional equation ζ(s) = 2ˢ πˢ⁻¹ sin(πs/2) Γ(1−s) ζ(1−s). The new ζ is no longer defined by the sum — it is defined by being the only well-behaved function that agrees with the sum where the sum makes sense. Everything afterwards is geometry on this extended object.

Section 03 · The critical strip

Trivial zeros outside, the interesting ones inside

The functional equation immediately reveals the 'trivial' zeros at s = −2, −4, −6, … — the sine factor forces them. Everything else has to live inside the critical strip 0 < Re(s) < 1, the only place where the function still has room to oscillate. Riemann's hypothesis is the claim that this whole strip is, in fact, deceptive: the non-trivial zeros do not spread out across it, they all collapse onto a single vertical seam down the middle, Re(s) = ½. The strip is a stage; the seam is where the action is.

The critical strip 0 < Re(s) < 1. The dashed verticals mark its borders; the warm line in the middle is Re(s) = ½. Each bead is one of the first eight non-trivial zero heights — 14.13, 21.02, 25.01, 30.42, 32.93, 37.59, 40.92, 43.33.

Section 04 · The conjecture

Re(s) = ½. Always.

Hardy proved in 1914 that there are infinitely many zeros on the critical line. Selberg later showed a positive fraction of all zeros sit there; that fraction has since been pushed above 41% (Conrey, 1989) and edged upward every decade. The hypothesis itself is the maximal statement: not 'most' zeros, not 'all but finitely many' — all of them, with no exception, on Re(s) = ½. It is the sharpest conceivable form of a probabilistic intuition about primes — that they are as uniformly distributed as a deterministic sequence can be.

ζ(½ + it) traced in the complex planezeros crossed: 5

t_max: 35.0t ∈ [0, 35.0]

Each time the curve loops through the origin, ζ has hit zero — and the imaginary part t at that moment is the height of a non-trivial zero. Drag the slider up: the first loop arrives around t ≈ 14.13, the second around 21.02, the third around 25.01.

Section 05 · What a proof would buy us

The downstream theorems are already written

The prime counting function π(x) is approximated by the logarithmic integral Li(x). RH is equivalent to the error |π(x) − Li(x)| being bounded by √x log x — the sharpest bound possible, and tighter than anything currently known. Hundreds of further statements are RH-equivalent: bounds on the Möbius function, on character sums, on growth of arithmetic functions. Cryptography quietly depends on these for its asymptotic security claims. A proof would not change a single computation, but it would settle every conditional theorem in one breath.

Section 06 · Numerical evidence and quantum chaos

The zeros line up — and they look like a physics spectrum

Computer verifications have placed the first ten trillion non-trivial zeros exactly on the critical line (Gourdon, 2004 and successors). And then, in 1972, Hugh Montgomery and Freeman Dyson noticed something stranger than any number-theory result: the statistics of the spacings between zeta zeros match precisely the spacings between eigenvalues of large random Hermitian matrices — the same matrices physicists use to model the energy levels of heavy atomic nuclei. The primes, in some sense yet to be understood, are humming at the resonant frequency of a quantum system.

See the engine

Open the Explorer.

The Explorer draws ζ along the critical line at full size, and lets you reconstruct the prime counting staircase from the zeros themselves — one Riemann correction at a time. The clean curves you build are the clearest physical statement of what the hypothesis is really claiming.

→ Open the Explorer

Where you meet it

How and where this technique lives in the world today.

Prime number theorem

The error term in the count of primes below N is controlled by the zeros of ζ; the Riemann hypothesis is equivalent to the sharpest possible bound on how irregularly the primes are spaced.

Cryptography

RSA, elliptic-curve crypto and integer-factorisation hardness all depend on assumptions about prime distribution; RH-equivalent statements feed into best-known bounds on cryptographic security.

Quantum chaos

The statistics of zeta-zero spacings match the eigenvalue statistics of random Hermitian matrices — the same matrices that model heavy-nucleus energy levels. The Montgomery-Dyson conjecture (1972) is one of the most surprising links in mathematics.

Computer-verified numerics

The first 10^13 non-trivial zeros have been computed to lie on the critical line (Xavier Gourdon, 2004 and later). No counterexample to RH has ever been found in any verification campaign.

Pop science & open-problem prestige

Riemann is the most famous unsolved problem outside Fermat — a Clay Millennium Prize ($1M), recurring fictional appearances (A Beautiful Mind, The Music of the Primes), and a steady stream of disproofs that don't survive peer review.