The story of quantum computing

Nature isn't classical, dammit, and if you want to make a simulation of nature, you'd better make it quantum mechanical.

Chapter 1 · The quantum puzzle (1900–1935)

1. 1900· Max Planck

   Energy comes in packets

   To explain black-body radiation Planck assumed energy is exchanged in quanta E = hν. He thought it was a mathematical trick. It was the birth certificate of quantum physics.

2. 1925–26· Heisenberg, Schrödinger, Born, Dirac

   Matrices and waves

   Heisenberg built mechanics from matrices; Schrödinger from a wave equation. Born showed |ψ|² is a probability, Dirac unified the two pictures and invented the bra–ket notation |ψ⟩ you see everywhere in this app. Every gate matrix here is Heisenberg's idea; every statevector is Schrödinger's.

3. 1935· Einstein, Podolsky, Rosen · Schrödinger

   Entanglement named

   EPR argued that correlated particles made quantum theory incomplete; Schrödinger coined the word 'entanglement' (Verschränkung) in reply. The Bell state in the hero animation is exactly their thought experiment.

Chapter 2 · From physics to computation (1964–1985)

1. 1964· John Bell

   Bell's theorem

   Bell proved entanglement cannot be explained by hidden local variables, turning a philosophical argument into a measurable inequality. Experiments by Clauser (1972) and Aspect (1982) confirmed quantum mechanics; the three shared the 2022 Nobel Prize.

2. 1980–82· Paul Benioff · Yuri Manin · Richard Feynman

   The idea of a quantum computer

   Benioff showed a Turing machine could run on quantum mechanics. Manin and Feynman asked the sharper question: classical computers choke on simulating quantum systems (2ⁿ amplitudes), so why not compute with quantum systems themselves? The ideal_statevector simulator here hits exactly that 2ⁿ wall at 24 qubits.

3. 1985· David Deutsch

   The universal quantum computer

   Deutsch defined the quantum Turing machine and gave the first algorithm where a quantum computer beats a classical one, later generalized with Jozsa (1992). You can run Deutsch–Jozsa from the examples gallery.

Chapter 3 · The algorithms that changed everything (1994–1997)

1. 1994· Peter Shor

   Factoring in polynomial time

   Shor's algorithm factors integers exponentially faster than any known classical method, threatening RSA encryption. Its heart is the quantum Fourier transform (QFT) in the examples gallery. Governments started funding the field the next year.

2. 1995· Shor · Steane · Cirac & Zoller · Barenco et al.

   Error correction and the first blueprint

   Shor and Steane invented quantum error-correcting codes, answering the objection that noise makes quantum computers impossible. Cirac and Zoller proposed the trapped-ion CNOT; Barenco et al. proved CNOT plus single-qubit gates are universal: the reason the IDE's palette is enough.

3. 1996· Lov Grover

   Quadratic search speed-up

   Grover's algorithm finds a marked item among N in about √N steps. The 2-qubit Grover example finds |11⟩ in a single iteration with probability 1.

4. 1997–98· Kitaev · Aharonov & Ben-Or · Chuang, Gershenfeld, Cory

   Threshold theorem and the first qubits

   The threshold theorem showed that if gate errors are below a constant, arbitrarily long computations are possible. The same year the first 2-qubit NMR experiments ran Deutsch's algorithm on real molecules.

Chapter 4 · Building the machines (1999–2019)

1. 1999· Nakamura, Pashkin, Tsai (NEC)

   The first superconducting qubit

   A Cooper-pair box showed coherent oscillations for a few nanoseconds. Today's IBM transmons descend from it, with T1 and T2 a hundred thousand times longer.

2. 2000· David DiVincenzo

   Five criteria

   Well-defined qubits, initialization, long coherence, a universal gate set, and measurement. Every hardware tab in this IDE reports against them: qubit count, T1/T2, basis gates, readout error.

3. 2007· Yale (Koch, Schoelkopf, Girvin, Devoret)

   The transmon

   A charge-insensitive superconducting qubit that made coherence times jump. It is the qubit inside every IBM Eagle and Heron processor you can select in the Hardware tab.

4. 2016–17· IBM

   Quantum computing on the cloud · Qiskit

   IBM put a 5-qubit processor online for anyone to use, then released Qiskit, the open-source SDK this whole application is built on.

5. 2018· John Preskill

   The NISQ era

   Noisy Intermediate-Scale Quantum: 50–1000 qubits without full error correction. Variational algorithms (VQE, QAOA in the examples) and error mitigation and circuit cutting (this IDE's Phase 7) are the tools of this era.

6. 2019· Google (Arute et al.)

   Quantum advantage claimed

   Sycamore, 53 qubits, sampled random circuits in 200 seconds; Google estimated 10 000 years classically. Classical teams later shrank that gap, but the race was on.

Chapter 5 · Scale and correction (2021–today)

1. 2021–23· IBM

   Eagle (127) and Heron (133)

   Eagle crossed 100 qubits with a heavy-hex lattice; Heron introduced tunable couplers and CZ gates with lower crosstalk. Their fake twins are the fake_* backends you simulate against.

2. 2023· Harvard/QuEra · IBM

   Logical qubits and 'utility'

   Neutral-atom arrays ran 48 logical qubits with error-corrected operations; IBM showed a 127-qubit Ising simulation beyond brute-force classical methods, using error mitigation.

3. 2024· Google

   Below threshold

   Willow demonstrated that a larger surface code (distance 7) had a lower logical error rate than a smaller one: the first time error correction clearly scaled the right way, 30 years after Shor's proposal.

4. 2025–26· Qiskit 2 · IBM Heron r2/r3 · the whole field

   Where you are now

   156-qubit Herons, fractional RZZ gates, Qiskit 2 with the C-based transpiler, and open circuit-cutting tools. QubitLab sits on top of all of this: every job you submit from the IDE travels this entire history in a few seconds.