“I’m sorry. My responses are limited. You must ask the right questions.”
-Dr. Alfred Lanning, I, Robot
Well, maybe you could, one day. But it won’t help you beat your friends at DotA or stream Netflix faster.
The things quantum computers can do are so much cooler.
A big myth of quantum (a word which is sometimes synonymous with ‘magic’ nowadays) computers is that they bend the rules of space-time to spit out an answer due to effects like ‘superposition’ and ‘entanglement’ with ‘spooky action at a distance’. No matter how bizarre these concepts may seem, they all are based in hard science. Top academics, research universities, and high tech companies across the world are dedicating their lives and grant funds to figuring out the power of quantum computing.
What can a quantum computer do that a classical computer can’t?
“That, detective, is the right question. Program terminated.” -Dr. Alfred Lanning, I, Robot
You should think of a quantum computer as analogous to the Graphics Processing Unit (GPU) complementing the Central Processing Unit (CPU) in your PC. The GPU improves your video performance by being optimally designed for displaying your 3D graphics as quickly as possible through its use of special geometry and shading calculators that remove the mathematical load off the CPU. It’s a specialized design that is much better utilized running calculations related to cutting-edge games rather than word processing.
Quantum computers will enhance, not replace classical computers.
Quantum computers are the GPU to your classical computer. It takes the same amount of time to browse on your classical computer as it would on your quantum computer; your browser would still need to go through the steps of parsing the URL, forming an HTTP request, fetching resources, and rendering. A quantum computer does not mystically speed up that specific work process, and still needs to wait for each step to complete before proceeding.
Just as a GPU processes instructions related to graphics exponentially faster than a CPU, certain problems can be solved quadratically or polynomially faster with quantum algorithms, by exploiting the quantum effects of superposition and entanglement.
What’s the quantum difference?
Your computer at home carries information in binary: 0 and 1. A quantum bit, or a qubit, holds information in a superposition of 0 and 1 — where the state is a probability of the qubit being either a 0 or 1. And not only do qubits hold information as a superposition of two states, but qubits are also entangled so the state cannot be described without using the state of the qubits they are entangled with (check out the Einstein-Podolsky-Rosen pair, the “unit” of quantum entanglement). You can’t deconstruct entangled pairs to just describe one, without the other. But really what’s different is that quantum computing gives us access to the complex vector space called the Hilbert space, where qubits form the unit vectors, and basis, of the space — which gives us the surface area of a sphere of possibilities of states a qubit can be in, instead of just 0 or 1.
Once put into a physical implementation, like photons, superconducting circuits (two electrodes with a thin insulator in between, allowing for quantum tunneling to create a superimposed state), nitrogen-vacancy centers (a common natural defect in diamond), or other types of qubits, quantum computers are able to consider many solutions simultaneously by mapping superimposed states onto these qubits. The quantum amplitudes will constructively and destructively interfere, so we can measure the peak as the ‘correct’ response.
Whatever form it takes, the quantum computer power relies on a set of quantum algorithms that provide substantial speedup over classical computing algorithms. The most common examples are Shor’s algorithm (factoring) and Grover’s algorithm (search), but over 50 quantum algorithms exist to look at a wide range of problems and industries. For example, materials science is a core discipline that has influence in every other high-tech industry. Processing power in computers is no longer increasing as quickly as before, since silicon is not able to handle the load without overheating. Faster flight depends on better aerospace materials which can shield from radiation or stand up to heat. A quantum computer could possibly simulate new structures and their properties to find materials suited to these applications. A quantum computer, naturally, could also efficiently simulate quantum systems, as well as advance core physics research.
There’s been a lot of hype the last few years about quantum computing, especially its capabilities, the current state of the field, and how big the revolution will be. But commercialization of quantum computing is just beginning. The algorithms are not rigorously tested, but the biggest challenges are now engineering based, focusing on scaling up the system, instead of theoretical understanding of qubits or quantum effects. The pieces of the puzzle have been researched, but those pieces need to be put together into a large, fault-tolerant, scalable quantum system, tested with quantum algorithms, and given the right problems to attack. The most exciting thing is that this is coming sooner than you think.
What is another problem a quantum computer can solve?
How about efficiently designing a better quantum computer?
(Did we hit the singularity here?)