What quantum computers can and can't do

The myth to unlearn

The most common description of quantum computing — 'it tries every answer in parallel' — is wrong in an instructive way. Yes, n qubits can hold a superposition over all 2ⁿ candidates. But a measurement returns one random candidate, which is no better than guessing. If parallel-universes-brute-force were the mechanism, Grover's search would be instant instead of √n.

The real mechanism is the one you've practiced across these lessons: choreograph interference so wrong answers cancel and right answers reinforce. That choreography is hard to find, which is why fast quantum algorithms exist only for problems with special structure — not for everything.

Where quantum wins

Factoring and codebreaking: Shor's algorithm (1994) factors big numbers exponentially faster than any known classical method by converting factoring into finding a repeating pattern — a period — which interference detects the way an ear picks out a musical pitch. Since RSA encryption's security is the difficulty of factoring, large quantum computers would break much of today's public-key cryptography; 'post-quantum' replacements are being deployed now, well ahead of the hardware.

Search and optimization: Grover's algorithm speeds up any brute-force search, but only quadratically — a million-item haystack takes ~a thousand steps, not one. Useful, not miraculous, and it doubles as a reason to double symmetric key lengths.

Simulating nature: the application Feynman proposed first may matter most. Molecules and materials are quantum systems; simulating them classically costs exponential memory (remember: 2ⁿ amplitudes). A quantum computer pays no such penalty for speaking nature's native language — chemistry, drug design and materials science are the field's most anticipated payoffs.

An honest scorecard

Quantum computers will not replace laptops: for email, spreadsheets and video games they offer nothing. They don't 'solve NP-complete problems in polynomial time' as far as anyone knows — the strongest evidence points the other way. And today's hardware remains noisy and small; the fault-tolerant machines the famous algorithms assume are still being engineered.

The honest summary: a quantum computer is not a faster computer, it's a different one — a machine that computes with amplitudes instead of probabilities. For problems whose structure lets interference do the work — periods, correlations, quantum systems themselves — that difference is exponential. Everything you've built in this simulator, from the two-path cancellation of HH to Grover's amplitude flow, is that difference in miniature.