Understanding Quantum Computing
For the great doesn’t happen through impulse alone, and is a succession of little things that are brought together. —Vincent van Gogh, Letter to Theo van Gogh, 1882
Before We Begin
This article gives a simple, practical overview of quantum computing. I’m not an expert, and you’ll find all references at the end. If you want more depth in the maths and physics, try Quantum Computing for Dummies (whurley & Floyd Earl Smith) for more mathematical detail, and Quantum Computing (Andrew Glasner) for a clearer, easier read.
1. Introduction
Picture flipping a coin in the air. As it spins, you can’t say it’s heads or tails — it’s in a kind of ‘both/and’ limbo until it lands. That everyday image is a handy way to grasp a key idea in quantum computing: superposition. While a classical bit is either 0 or 1, a quantum bit (qubit) can occupy a range of possibilities at once until it is measured, at which point its state ‘lands’ as a definite 0 or 1.
The uncertainty you see while the coin is spinning is like the uncertainty we capture and use in quantum computing. We put many processing elements — qubits — into a state of uncertainty. Then we program the qubits, run the program, and capture the results — just like when the coin lands.
Quantum computing is different from the fixed 0s and 1s, bits and bytes, used in today’s devices. Quantum computing is based on quantum mechanics, a branch of physics that can be hard to comprehend. But the way in which quantum computing deals effectively with large degrees of uncertainty feels like the way we make many of the decisions we encounter in daily life.
A traditional computer (also called a conventional or classical computer) is based on manipulating the simplest unit of information, called a bit. A bit can represent one of two possible states, which have been given the names 0 and 1. A quantum computer follows the same general idea and manipulates the simplest unit of quantum information, called a qubit (pronounced CUE-bit).
Quantum computing is complementary to classical computing, the kind of computing we use today, not a replacement for it. By working with uncertainty, we can take on some of the biggest, most complex problems that humanity faces in a new and powerful way. Quantum computing will solve problems for which today’s computing falls short — problems in areas such as modeling the climate, drug discovery, financial optimization, and whether or not it’s a good morning to launch a rocket.
Quantum computing is just getting started; many advanced quantum computers run only for a fraction of a second at a time. However, steady progress is being made. Even now, at this early stage, quantum computing is inspiring us to, as a sage once said, “think different” about the way we use existing computing capabilities.
2. Why Quantum Feels “Strange”
Quantum computers have a sense of strangeness about them, almost a mystical aura. (The 2022 movie, Dr. Strange in the Multiverse of Madness, captures some of the feeling that people have about quantum mechanics in general.) Why is this?
There are two main reasons. The first reason is people’s fundamental misunderstanding of the nature of matter, which quantum mechanics explains. The second is the incredible power that quantum computing, when mature, is expected to deliver to humanity.
How does quantum mechanics change people’s view of the world? The world we live in, where rocks fall down and rockets go up, seems to be dominated by solid matter, with energy as a force that acts on matter at various times. Yet matter can simply be seen as congealed energy.
Most of the mass of the protons and neutrons inside the nucleus of an atom, for instance, is simply a bookkeeper’s description of the tremendously powerful energetic fields that keep these particles in place. One of the most important kinds of particles in quantum computing, photons, have no mass at all; they are made up of pure energy.
And it was Einstein himself who told us that matter and energy are equivalent, with his famous equation, E=mc2. To translate: The energy contained in solid matter equals its mass times the speed of light squared.
The speed of light is a very large number — 300,000 km/second, or 186,000 miles/second. Squaring the speed of light yields a far larger number. Plug this very large number into Einstein’s famous equation and you'll see that there is a lot of energy in even small amounts of matter, as demonstrated by nuclear power plants and nuclear weapons.
The point is that, in quantum mechanics, matter is relatively unimportant; particles act more as bundles of energy. And quantum computing takes advantage of the exotic properties of these particles — ionized atoms, photons, superconducting metals, and other matter that demonstrates quantum mechanical behavior.
The second reason that quantum computers get such a strong emotional reaction is the tremendous power of quantum computing. The best of today’s early-stage quantum computers are not much more powerful, if at all, than a mainstream supercomputer. But future quantum computers are expected to deliver tremendous speedups.
Over the next decade or two, we expect quantum computers to become hundreds, thousands, even millions of times faster than today’s computers for the problems at which they excel. People can’t really predict, nor even imagine, what it’s going to be like to have that kind of computing power available for some of the most important challenges facing humanity. That future is very exciting, yes. But it’s also a bit, as Einstein described quantum mechanics, “spooky.”
3. Core Quantum Concepts
3.1 Qubits
Qubits are the quantum computing version of bits — the 0s and 1s at the core of classical computing. They have quantum mechanical properties. Qubits are where all the magic happens in quantum computing.
3.2 Superposition
While bits are limited to 0 or 1, a qubit can hold an undefined value that is neither 0 nor 1 until the qubit is measured. The capability to hold multiple values at once is called superposition.
3.3 Entanglement
In classical computing, bits are carefully separated from each other so that the value of one does not affect others. But qubits can be entangled with each other. When changes to one particle cause instantaneous changes to another, and when measuring a value for one particle tells you the corresponding value for another, the particles are entangled.
3.4 Tunneling
A quantum mechanical particle can instantaneously move from one place to another, even if there’s a barrier in between. (Quantum computing uses this capability to bypass barriers to the best possible solution.) This behavior is referred to as tunneling.
3.5 Coherence
Coherence is the lifetime over which qubits maintain their quantum behavior. Heat, vibration, electromagnetic noise, and imperfect controls all cause decoherence and errors. Much of quantum engineering is about preserving coherence long enough to compute, and correcting errors that occur.
These five terms are at the heart of the promise of quantum computing and are involved in many of the challenges that make quantum computing difficult to fully implement
Classical computing describes the computers we use every day, which includes not only laptop and desktop computers but also smartphones, web servers, supercomputers, and many other kinds of devices. The term classical computing is used because classical computers use classical mechanics, the cause-and-effect rules of the road that we see and use in our daily lives, for information processing. Quantum computing uses quantum mechanics — which is very different, very interesting, and very powerful indeed — for information processing.
Thanks to superposition and entanglement, a quantum computer can process a vast number of calculations simultaneously. A classical computer works with ones and zeros, a quantum computer will have the advantage of using ones, zeros and “super-positions” of ones and zeros. Certain difficult tasks that have long been thought impossible for classical computers will be achieved quickly and efficiently by a quantum computer.
4. Why Quantum Computing Is Powerful
In simplified terms, a register of n classical bits describes exactly one of 2ⁿ possible states at a time. A register of n entangled qubits can evolve as a superposition over many of those states simultaneously. When algorithms exploit this structure (via interference), they can achieve dramatic speedups for specific problems.
Important caveat: quantum speedups are not automatic. They appear when problems have structure that quantum algorithms can exploit, and when hardware can preserve coherence and manage errors well enough to realize those algorithms.
5. State of Quantum Computing Today
We are in the NISQ era (Noisy Intermediate-Scale Quantum): devices with tens to low thousands of qubits, significant noise, and short coherence times. Superconducting and trapped-ion systems dominate deployments today; photonic and neutral-atom platforms are rapidly advancing. Roadmaps (e.g., IBM’s) highlight steady gains in qubit counts, quality, and modular architectures, along with milestones like ‘quantum utility’ — demonstrations of useful results not feasible on comparable classical resources.[Image placeholder: IBM quantum processor suspended from cryostat; superconducting qubits near absolute zero]
IBM quantum processor suspended from cryostat; superconducting qubits near absolute zero - Source: Flickr/Lars Plougmann
6. What Quantum Computing Will Do
Quantum’s promise is to accelerate or enable solutions in domains where classical
Recommended by LinkedIn
Quantum computers won’t replace your laptop. Instead, expect hybrid workflows: classical systems orchestrate data and pre/post-processing, while quantum accelerators tackle specific subproblems.
7. Types of Quantum Computing
7.1 Quantum-Inspired Computing
Classical computers are based on logic gates, which take in electrical currents and apply binary logic to them to produce either no outgoing current (a 0) or an easily measurable current (a 1) as a result. By breaking down any mathematical or logical problem to a low enough level and running it through logic gates, you can come up with an answer.
When insights from quantum computing are used to create new algorithms that run on classical computers, or when quantum computer simulators are run on classical computers, this is called quantum-inspired computing. Quantum-inspired computing has proven to be a productive approach to finding new solutions at this early point in the development of quantum computing.
7.2 Quantum Annealing
In classical computing, there’s an approach called simulated annealing. It’s analogous to annealing in metallurgy, in which a metal is heated to melt its internal structure, and then cooled to yield a softer, easier-to-work result. Simulated annealing is good for solving optimization problems, such as the least expensive route for visiting a number of cities. A similar process to this can be used to solve problems by putting a set of equations through a series of transformations until a result appears.
The quantum computing version of this approach is called quantum annealing. Quantum annealing doesn’t use a gate-based quantum computing approach, which is more advanced but harder to implement. Instead, quantum annealing uses qubits as a group to solve optimization problems, which is a less demanding use of qubits.
Quantum annealers allow for qubit errors and return results that are inexact but still useful in many cases. For instance, a quantum annealer might identify a very good way to route a fleet of delivery trucks, rather than delivering the one and only best possible routing. Only one substantial company, D-Wave, makes quantum computers that use the quantum annealing approach — and they have recently announced plans to make logic-gate quantum computers as well.
Quantum annealers were criticized at one point as not being actual quantum computers, but that is no longer the case. However, gate-based quantum computers get most of the attention, research effort, and investment.
Theory says that quantum annealing should eventually be replaced by a combination of better classical computing approaches — taking over its use cases from “below,” in terms of technical sophistication — and gate-based quantum computing, taking over its use cases from “above.” We’ll see.
7.3 Gate-Based (Universal) Quantum Computing
Most quantum computers use quantum circuitry to reproduce the logic-gate structure of classical computers. These computers are called, sensibly enough, gate-based quantum computers. Gate-based quantum computers use qubits in a specific way and are not tolerant of errors in the operation of the qubits; the quantum logic gates don’t work reliably if the qubits aren’t error-free. Today’s qubits are not reliably error-free, which compromises the utility of today’s gate-based quantum computers.
Gate-based quantum computing is so hard to achieve that we can’t yet be certain that it will ever reach its full potential. But if it does, it will probably render both quantum-inspired computing and quantum annealing obsolete. In the fullness of time, gate-based quantum computing should become the best way to handle any kind of quantum-related computing problem.
Four major types of qubits are used in this type of quantum computer, based on the actual physical object kept in a coherent state to perform quantum computing:
Each kind of qubit has its pluses and minuses. Different quantum computing companies are basing their efforts on different kinds of qubits.
Keeping qubits coherent long enough to run logic-gate operations through them is difficult, so logic-gate quantum computers tend to have relatively few qubits — a few hundred at the most and to run programs for only a fraction of a second before decohering.
8. Major Challenges on the Path to Scalable Quantum
8.1 Hardware
8.2 Software
8.3 People & Practice
9. IBM Quantum computing
IBM, a leader in quantum computing, has published a roadmap showing past and future increases in the number of qubits that power its current and upcoming quantum computers. A simplified version of the roadmap is shown in the image below. You can find a link to the current version of the roadmap at https://research.ibm.com/blog/ibm-quantum-roadmap-2025.
10. Blowing the past
Albert Einstein wears two hats in the history of quantum mechanics — and the two hats don’t fit comfortably on a single head.
One hat comes from Einstein’s discovery of relativity, published in 1905. Relativity says that speed in this universe depends on your motion relative to other observers, but that the speed of light — about 186,000 miles per second, or 300,000 kilometers per second — is always the same for all observers. This universal speed limit is called locality.
The other hat comes from Einstein’s discovery of the photon, also in 1905. (This discovery, not relativity, is the source of Einstein’s sole Nobel Prize.) The discovery of the photon is fundamental to quantum mechanics.
Einstein’s problem is that quantum mechanics later asserted that quantum particles, such as photons, can be entangled with each other, so that reading the spin (for example) of one photon tells you the spin of the other. And this relationship is instantly true, without regard to the speed of light. Physicists call this an assertion of nonlocality, which is supposed to be forbidden by relativity.
Einstein hated this, calling it “spooky action at a distance.” He and his colleagues spent a great deal of effort trying to disprove it, even as Einstein continued to make breakthrough quantum discoveries, such as the identification of Bose-Einstein condensates, which are superconducting gases that can be used to create qubits.
Today’s mainstream computers are subject to classical mechanics and limited by the speed of light. Quantum computers depend on quantum mechanics and, in their use of entanglement, are not limited by light speed.
The Nobel Prize for Physics in 2022 was awarded to physicists who showed that entanglement is real. So researchers in quantum computing who depend on entanglement can say, after Galileo: “And yet it computes.” (Galileo, on trial for asserting — correctly, as it turned out — that the earth is not at the center of the universe, is famously said to have whispered: “And yet it moves.”)
11. The Road Ahead
Quantum computing stands at an ‘awkward adolescence’: clear promise, growing investment, and early demonstrations of quantum utility — but no broad, uncontested advantage on headline industrial problems yet. The most realistic near-term path is hybrid: classical systems do what they’re superb at; quantum accelerators tackle subproblems where superposition, entanglement, and interference confer real benefit. As error rates fall and error correction scales, expect stepwise expansions of capability.
Clarifications & Common Misconceptions
Online references:
Coursera educational content:
