Intel's director of quantum hardware, Jim Clarke, holds the new 17-qubit superconducting test chip
- On Tuesday, Intel delivered something called "a 17-qubit superconducting test chip" to QuTech, Intel's research partner in the Netherlands.
- This is the world's second 17-quantum bits, or qubits, quantum computing chip. The first was introduced by IBM in May.
- This chip allows Intel to put a stake in the ground in one of the strangest, and potentially game-changing, new forms of computing that researchers are currently developing.
On Tuesday, Intel came out with its state-of-the 17-qubit superconducting test chip, matching the biggest quantum computer chip ever to be produced, by IBM.
Until now, quantum computing has been, in some regards, a two-horse race between IBM and Google. In April, Google showed off its research on a nine qubit computing chip and has advanced other research that could allow it to break some quantum computing records by the end of the year. In May, IBM showed off the first-ever 17 qubit chip.
IBM's work is based on research done at Yale through professor Robert Schoelkopf (the IBM team includes many of his Ph.D. and post-grad students). Google's work is based on research from University of California at Santa Barbara under professor John Martinis, an effort that was backed and absorbed by Google in 2014.
All the researchers from IBM, Intel, Google and elsewhere, like Microsoft, are all in a race to to build a 50 quibit chip. That's the size needed to build a supercomputer which would be vastly more powerful than any of today's supercomputers. No one even knows what kinds of problems a computer that fast and smart could solve.
Quantum entanglement
Quantum computers are different than today's computers, which are digital. A digital computer thinks in two states: zero and one (or off and on). But a quantum computer uses combinations of zeroes and ones to creates multiple states, which can be a zero, a one, both at the same time or (and this is the weird part) something in between, a mysterious zero/one state that's hard to describe or determine.
These messy states are called "entanglement," and there are already several well-known mathematical formulas (aka algorithms), that can make use of these states to calculate things that traditional computers aren't powerful enough yet to do. For instance, quantum computers can work with billions of variables at the same time, like the interaction between molecules in chemistry.
They are also great for machine-learning tasks. These computers are expected to help find new drugs, create new forms of computer security. It is also believed that this type of computing could lead to computers that can think and reason to create humanoid robots, or deliver medicine that is personalized to each human's own, unique chemistry.
If all this sounds hard to grasp, don't worry, you aren't alone. Microsoft is also betting big on quantum computing and yet Bill Gates admits that, despite all the physics and math he knows, even he doesn't really understand how quantum works. That's how complicated it is.
Colder than space
For now, the challenge is simply to build bigger quantum computers.
QuTech
QuTech Professor Leo DiCarlo in the quantum computing lab at QuTech
That kind of condition is hard to create and maintain.
It's not just the cold that's a problem. As a quantum computer grows in size by adding more qubits, it can malfunction in a lot of ways.
But progress is going fast. In May of 2016, IBM launched a five-qubit machine and the world's first cloud service. Flash forward a year and these chips are already more than triple the size.
Google expects that it will have created a test computer so big and powerful by the end of this year, that it will be able to perform certain calculations that traditional supercomputers cannot do, a concept called "quantum supremacy," Martinis told Motherboard.
In the meantime, Intel just threw itself headlong into the game. Here's a closer look at what it's up to:
Get the latest IBM stock price here.