What Is A Quantum Computer? The 30,000 Foot Overview

Title slide from the quantum computing presentationCHAD ORZEL

I was part of a panel at the 2018 meeting of the National Association of Science Writers on communicating quantum physics, and they asked me to do a brief overview of the field suitable for non-physicists. In 10-12 minutes.

I ended up going for what I think of as the view from 30,000 feet, covering the basic physics principles that make quantum computers interesting. Having spent a bunch of effort making this, I’m both thinking of ways to re-purpose it for other venues, and typing up the following blog version of what I said at the panel:


My goal here is not to describe any specific quantum computer, because all the various platforms that are under development are really complicated, and would take too much time. Rather, my goal is to describe the fundamental physics that makes quantum computing powerful and an interesting topic to study. In very general terms, that boils down to a one-sentence description:

A quantum computer exploits properties of quantum physics to perform certain types of calculations more efficiently than any classical computer

Of course, like any field more complicated than digging holes and immediately filling them back in, this one-sentence description needs a bit of unpacking, starting with the “properties of quantum physics” in play here.

As always when talking quantum, the first property that comes into play is the idea of wave nature. Quantum physics tells us that everything in the universe, including material objects that we normally think of as particles, has wave nature. The wave character of an electron picks out particular energy states for an electron inside an atom– you don’t go too terribly wrong if you imagine this as analogous to the standing-wave modes you see for sound waves in an organ pipe or on a guitar string.

These special states are the conceptual justification of the Bohr model of electrons in atoms, in which an electron can only be found in certain states of well-defined energy. Electrons move between these states by absorbing or emitting photons of light, with the light frequency corresponding to the energy difference between the initial and final states. (The actual states aren’t literally planetary-style orbits as in the actual Bohr model, but the conceptual picture isn’t wrong.)

What does this have to do with computing? Well, a quantum system with discrete states lets you pick out two of those states and call one of them “|0>” and the other “|1>” and treat them as the two states of a binary bit. You can drive this bit back and forth between states by applying light of the appropriate frequency, so these can form the base elements needed to do computation.

Waves and a jumping fish, highlighting some of the superposed wave patterns.CHAD ORZEL

Of course, if this were the only thing quantum physics had to offer, nobody would be interested in quantum computing. The wave nature of everything, though, leads to another interesting property that’s characteristic of waves. If you look at a picture of waves like the vacation photo shown above, you can identify multiple discrete wave patterns: there are several sets of concentric rings where something broke the surface of the water at a particular point, and also a larger pattern of straight-line waves running diagonally across the photo.

If you pick a random bit of water, such as the orange box highlighted, you can’t associate that element of water with a single wave. It’s affected by all of these patterns at the same time. That’s a very general property of waves, and carries over to quantum objects: they can be in multiple wave-like states at the same time.

This superposition property leads to some problems, as in the infamous Schrodinger Cat thought experiment with a boxed feline that’s both alive and dead. It’s the second property exploited for quantum computing, though: a qubit is not restricted to being either 0 or 1, but can be a superposition of both |0> and |1> at the same time. At the end of the computation, you’ll measure it to be one or the other, but during the computation process the exact state is indeterminate, and contains bits of both.

The ambiguous Necker cube and the two states it can resolve into.CHAD ORZEL

This two-things-at-once situation seems really strange, but if you’ve ever tried to learn three-dimensional drawing, you’ve seen a nice visualization of this: the Necker cube. You can see one of these wire-frame cubes as either facing up and to the right, or down and to the left, and with a bit of effort you can make your brain flip back and forth between the two. The unshaded cube is a mix of the two states, both up-and-right and down-and-left, in a manner analogous to the |0>-and-|1> superposition of a qubit.

This ambiguous cube also provides a nice way to illustrate the third property of quantum objects exploited for quantum computing, which is the idea of entanglement. If you put two Necker cubes side-by-side, you get a pair of objects whose individual states are indeterminate– both up-and-right and down-and-left at the same time– but where each is guaranteed to be the same as the other. You can’t really get your brain to see the cube pair in a split state where they face in different directions.

A pair of Necker cubes as a metaphor for entanglement.CHAD ORZEL