A qubit is the basic unit of quantum information. Unlike a classical bit, which is either 0 or 1, a qubit can be in a superposition - a combination of 0 and 1 at the same time - described by two amplitudes. When you measure it, the superposition collapses and you get a single 0 or 1 with a probability set by those amplitudes. Qubits can also be entangled, linking their outcomes.

Why do quantum chips need to be so cold?

Many leading quantum chips use superconducting qubits, which must be cooled to around 10-15 millikelvin - colder than outer space - using a dilution refrigerator. At those temperatures the circuits become superconducting and thermal noise is suppressed enough for fragile quantum states to survive. Other qubit technologies, such as trapped ions or photonics, use different setups and are not all cooled the same way.

What can quantum computers actually do?

Quantum computers are not faster general-purpose computers. They offer advantages on specific problems: factoring large numbers (Shor's algorithm, which threatens current encryption), unstructured search (Grover's algorithm, a quadratic speedup), and especially simulating quantum systems like molecules and materials for chemistry and drug discovery. For everyday tasks - email, browsing, most software - a classical computer is and will remain better.

What is decoherence and why does it matter?

Decoherence is the loss of a qubit's fragile quantum state due to interaction with its environment - noise, heat, stray fields. It happens very quickly, often in microseconds, and is the central challenge of quantum computing. To fight it, quantum error correction encodes one reliable logical qubit using many noisy physical qubits, which is why useful fault-tolerant machines may need thousands or millions of physical qubits.

How a Quantum Chip Works — Qubits, Superposition & the Honest Truth

Q: Why are quantum computers powerful?

With n qubits the system holds 2^n complex amplitudes simultaneously, so the state space grows exponentially - 300 qubits would have more amplitudes than there are atoms in the observable universe. A quantum algorithm uses interference to make the amplitudes of wrong answers cancel and right answers reinforce, so the correct result is likely when measured. This gives exponential or quadratic speedups for specific problems, not for everything.

The core idea

The qubit — not just 0 or 1

A classical bit is a switch: 0 or 1. A quantum bit, or qubit, can be in a superposition — a blend of 0 and 1 at the same time, described by two numbers (amplitudes) that say "how much" 0 and "how much" 1.

The catch — and this is where the magic and the limits both live: when you measure a qubit, you don't see the blend. It collapses to a single 0 or 1, with a probability set by those amplitudes. So you can't just "read out" all that hidden information. The whole art of quantum computing is arranging things so the answer you want is the one most likely to appear.

Figure 1 — A bit is one of two values; a qubit's state is an arrow that can point anywhere on a sphere — a continuous blend of |0⟩ and |1⟩ until measured.

The three superpowers

Superposition, entanglement, interference

Superposition — a qubit holds 0 and 1 at once, so n qubits represent many combinations simultaneously.
Entanglement — two qubits can be linked so their outcomes are correlated no matter the distance; measuring one instantly tells you about the other. This lets qubits act as one connected system, not independent bits.
Interference — like waves, amplitudes can add up or cancel out. A good quantum algorithm steers interference so wrong answers cancel and the right answer reinforces — so it dominates when you measure.

💡 The honest version of "tries all answers at once"

You'll hear "a quantum computer tries every possibility at once." Half-true. It holds all possibilities in superposition — but you only get to measure once, yielding a single answer. The cleverness is interference: shaping the amplitudes so the useful answer is the likely one. Without that, you'd just get a random result.

Why it scales

n qubits = 2ⁿ amplitudes

Add qubits and the state space grows exponentially. 1 qubit needs 2 amplitudes; 2 qubits need 4; 3 need 8; n qubits need 2ⁿ. This explodes fast:

Figure 2 — The state space doubles with every qubit. This exponential scaling is the source of quantum's potential power.

That's the promise. But remember Figure 1's catch: you can't read all 2ⁿ amplitudes — measurement gives you one n-bit answer. Exponential power only helps when an algorithm can use interference to concentrate it.

The physical chip

What a quantum chip actually is

A qubit needs a real physical system with two quantum states you can control. Several technologies compete:

Qubit type	How	Used by
Superconducting	tiny superconducting circuits (transmons); the most common today	Google, IBM, others
Trapped ions	individual atoms held by fields, manipulated with lasers	IonQ, Quantinuum
Photonic	qubits encoded in particles of light	PsiQuantum, Xanadu
Neutral atoms	atoms held in optical tweezers	QuEra, Pasqal
Spin / topological	electron spins / exotic states (early research)	Intel, Microsoft

The famous "chandelier" you've seen photos of is not the quantum chip — it's the dilution refrigerator around it. The actual superconducting chip is a small square at the very bottom.

Why so cold

Colder than outer space

Superconducting quantum chips are cooled to about 10–15 millikelvin — a hundredth of a degree above absolute zero, colder than deep space. Two reasons: the circuits become superconducting (zero resistance) at those temperatures, and thermal noise is suppressed so the fragile quantum states aren't instantly destroyed. The dilution fridge's gold "chandelier" is the multi-stage cooling and wiring that gets there. (Other qubit types — trapped ions, photonics — use different setups and aren't all cooled this way.)

Operating it

Quantum gates

You program a quantum computer with quantum gates — operations that rotate or entangle qubits, applied as precisely shaped microwave or laser pulses. Examples: the X gate flips a qubit (like NOT), the Hadamard (H) gate creates an equal superposition, and the CNOT gate entangles two qubits. A sequence of gates is a quantum circuit — the quantum analogue of the logic circuits in our Logic Gate Simulator, but operating on amplitudes, not fixed 0/1 values.

The hard wall

Decoherence & error correction

Quantum states are extraordinarily fragile. Any stray heat, vibration or field nudges a qubit and its delicate superposition leaks away — this is decoherence, and it often happens in microseconds. It's the single biggest obstacle in the field.

The defence is quantum error correction: spread one reliable logical qubit across many noisy physical qubits so errors can be detected and fixed. The ratio is brutal — a single robust logical qubit may need hundreds to thousands of physical qubits. That's why headline "qubit counts" can be misleading, and why genuinely useful, fault-tolerant machines may require millions of physical qubits.

⚠ The NISQ reality

Today's machines are NISQ — Noisy Intermediate-Scale Quantum: tens to a few hundred physical qubits, too noisy for full error correction. They're real and improving, but they are not yet the world-changing fault-tolerant computers of the headlines.

The honest part

What quantum can — and can't — do

The biggest myth: "a quantum computer is just a much faster computer." It isn't. For everyday tasks — email, browsing, spreadsheets, most software — a classical computer is and will stay better. Quantum helps only on specific problem structures:

Factoring large numbers — Shor's algorithm could break today's RSA encryption (driving "post-quantum cryptography").
Unstructured search — Grover's algorithm gives a quadratic (√N) speedup.
Simulating quantum systems — molecules, materials, chemistry. This is the most natural and likely most valuable use: a quantum machine simulating quantum physics, for drug discovery, catalysts and batteries.
Certain optimisation — promising but the real-world advantage is still unproven.

	Classical	Quantum
Unit	bit (0 or 1)	qubit (superposition)
Best at	everyday, general computing	specific: factoring, search, simulation
Errors	extremely rare	frequent (decoherence)
Status	mature, everywhere	early (NISQ), improving

✅ The whole guide in three sentences

A qubit holds 0 and 1 in superposition, and n qubits hold 2ⁿ amplitudes — but measurement gives one answer, so algorithms use interference to make the right one likely. Superconducting chips run near absolute zero to protect these fragile states, which still decohere in microseconds, so error correction needs many physical qubits per logical one. Quantum isn't a faster everyday computer — it's a specialised tool for factoring, search and especially simulating nature itself.

Quantum computing is advancing fast but is genuinely early and heavily hyped. Treat bold "quantum will do X next year" claims with healthy skepticism — including this page; verify the latest before relying on it.

Reference

FAQ

What is a qubit?

The basic unit of quantum information — unlike a 0/1 bit, it can be in a superposition of 0 and 1, and collapses to one value when measured.

Why are quantum computers powerful?

n qubits hold 2ⁿ amplitudes; interference concentrates probability on the right answer. This helps specific problems, not all computing.

Why so cold?

Superconducting qubits need ~10–15 mK so circuits superconduct and thermal noise doesn't destroy the fragile states.

What can they actually do?

Factoring (Shor), search (Grover), and simulating molecules/materials. Not faster email or browsing.

How a Quantum Chip Works — and What It Can't Do