Quantum Computing — Brief ☧
Deep version → | Related: NVIDIA/AMD → | Neurosymbolic →
"The hearing ear, and the seeing eye, the LORD hath made even both of them."
— Proverbs 20:12 (KJV)
Q: A classical computer stores information as bits — each one is
either 0 or 1, like a light switch that is either off or on. What if a
switch could be both off and on at the same time?
A: That is a qubit — the basic unit of quantum computing. Thanks
to a property called superposition, a qubit exists as a blend of 0
and 1 simultaneously:
a|0> + b|1>, whereaandbare probabilityamplitudes. Only when you measure the qubit does it collapse to a
definite 0 or 1 — with probability |a|^2 for 0 and |b|^2 for 1.
Q: How is that useful? If it collapses when you look at it, how do
you compute with it?
A: The power comes from manipulating qubits before measuring. With
N qubits in superposition, you represent 2^N states simultaneously.
Quantum algorithms carefully arrange
interference — amplifying paths that lead to the right answer and
canceling paths that lead to wrong ones — so that when you finally
measure, the correct answer is overwhelmingly likely.
Q: Can you give a concrete example?
A: Grover's search: you have a list of N items and want to find
the one that satisfies some condition. A classical computer needs up to
N checks (linear time). Grover's
algorithm finds it in roughly sqrt(N) checks — a quadratic speedup. For
a million items, that is 1,000 checks instead of 1,000,000.
Q: "For now we see through a glass, darkly; but then face to face"
(1 Corinthians 13:12). Is a qubit in superposition like seeing through
that dark glass?
A: A beautiful parallel. In superposition, you glimpse all
possibilities at once — "through a glass, darkly." Measurement collapses
the superposition to a single answer — you see "face to face," but only
one face. The art of quantum computing is arranging things so the face
you see is the one you need.
Key Concepts
| Concept | Classical | Quantum | Analogy |
|---|---|---|---|
| Bit / Qubit | 0 or 1 | 0 and 1 (superposition) | Light switch / dimmer |
| Gate | AND, OR, NOT | X, H, CNOT | Logic operation / rotating the qubit's state |
| Measurement | Read the bit | Collapse superposition | Looking and getting a definite answer |
| Entanglement | Independent bits | Correlated qubits | Two coins that always land the same way |
Classical bit: |0> or |1> (one state)
Qubit: a|0> + b|1> (both states, amplitudes a, b)
where |a|^2 + |b|^2 = 1
Measurement: -> |0> with probability |a|^2
-> |1> with probability |b|^2The concepts in this table map cleanly between the classical and quantum worlds, and understanding these mappings helps demystify quantum computing. A qubit is to a bit as a dimmer is to a light switch -- it adds a continuum of possibilities between the extremes. Gates are to qubits as logic operations are to bits -- they transform the state. And entanglement is perhaps the most exotic concept: two qubits that are correlated in a way that has no classical analog, like two coins that always land the same way no matter how far apart they are flipped.
Does quantum computing make classical computing obsolete? No. Quantum
excels at specific problems — search, factoring, optimization — but for
most everyday tasks, classical computers (including GPUs and FPGAs) are
faster, cheaper, and more practical. The time complexity
advantage is problem-specific, not universal. Today's quantum computers have roughly 1,000 noisy physical qubits, far from the millions needed for practical advantage on real-world problems. This is likely to change in the coming decades, but for now, classical hardware remains the practical choice.
Connection to our project: Our FPGA explores all constraint assignments
through bitmask parallelism — a
classical form of superposition. A 64-bit domain represents 64 possible
values simultaneously, and a bitwise AND operation tests all 64 values against a constraint in a single clock cycle. Grover's search offers a quadratic speedup over
exhaustive search (sqrt(N) instead of N queries); our FPGA offers a
constant-time 70,000x speedup
per intersection that applies immediately, with no error correction needed and no exotic cooling requirements. For practical problem sizes (under a million values), our classical bit-parallel approach
often wins. The honest comparison is this: quantum computing promises asymptotically better scaling for enormous problems, while our FPGA delivers reliable, deterministic speedups for the problem sizes that matter today.
Soli Deo Gloria