“OK”, you say, “so you told us in your last post that quantum computers are about to emerge big time and that they are really fast compared to the ordinary computer today. How does being ‘quantum’ make them fast?”
That’s a key question, of course. I’ll try to give you an idea without lapsing into science-speak too much.
In the 1980’s some really smart people, among them Richard Feynman, Yuri Manin and Charles Bennet, realized that certain phenomena that only occur on the quantum level can be used to find solutions to certain mathematical problems such as transmitting spy-proof messages, the teleportation of information, the generation of true random numbers and the breaking of cryptographic codes hitherto deemed unbreakable. Since then, a lot of other problems have been formulated in a way which makes them amenable to quantum computing.
At the heart of ordinary computers rests a central processing unit, a so-called CPU, which performs all calculations. Each CPU can only do one computation at once, but since it can do them very quickly, it appears as if it were doing several of themin parallel.
The basic kernel of information on which a CPU performs its calculations is called a “bit”, representing the smallest amount of information possible: either 0 or 1. A bit can have only one of those two values. But that’s perfectly OK, because all it takes is a sufficiently long string of bits to represent what ever you want: large numbers, letters, images, sound, videos, etc. Everything can be represented by bits.
A quantum computers works not on bits but on “qubits” (not to be confused with “Q-tips”, the little sticks with cotton tips you swab your ears with). Qubit stands for “quantum bit” which suggests a close correspondence with classical bits but on the quantum level.
“What’s the difference?” you ask. Well, the main difference is that while a classical bit can either be 0 or 1, a qubit can be a bit of both to varying degrees. There is an infinity of possible states a qubit can be in, compared to only two states of a classical bit. This difference comes to play when you look at several qubits.
Eight bits can be strung together to represent any number between 0 and 255. Eight qubits, however, can represent any number in that range at once. And in that lies the power of the quantum computer: it can perform many calculations truly concurrently.
The difficult part is how to make qubits, keep them alive long anough and have them interact in specific ways to perform calculations. The handling of qubits is the part where research is needed most. But already a large number of scientists have found many different approaches to this problem. Looking at the speed at which quantum computers have progressed in the past couple of years alone, 5-10 years in the future will have them starting to move into mainstream computing.