An Intel Tunnel Falls quantum chip on a silicon wafer. The wafer is capable of up to 12 qubits of quantum processing.
Intel
The hardware giant sees this as a step along its path towards ‘full-stack’ quantum computing solutions.
On Thursday morning, Intel announced the release of its newest quantum computing chip, which it calls ‘Tunnel Falls’. The chip is aimed at the quantum computing research community, and as part of the announcement the hardware giant said that it will be providing chips to the Sandia National Laboratory as well as labs at the University of Maryland, the University of Rochester and the University of Wisconsin-Madison.
Intel’s announcement comes on the heels of its release earlier this year of a software package for developers that simulates a full quantum stack. But it also comes on the heels of its competitor IBM announcing a major breakthrough in its quantum technology in a paper published with the University of California Berkeley in Nature this week.
They’re not the only ones. Silicon Valley heavyweights from Google to Microsoft to Amazon are all working on quantum systems, not to mention a slew of startups and even unexpected big companies like Honeywell. That’s because for certain applications, such as artificial intelligence, logistics planning, chemical simulation and encryption, quantum computing promises to be a game-changer. In a report earlier this year, McKinsey estimated that the quantum computing market alone could top $90 billion by the year 2040, with the economic value generated potentially in the trillions.
The reason for that lies in basic physics. A classical computer is fundamentally limited because each transistor can only exist in one of two states–a zero or a one. Every computer you use is built on this principle. But a quantum bit (or “qubit”) can exist in multiple states, which rapidly accelerates certain types of computing applications. The catch, though, is that qubits are extremely fragile, which makes them prone to error and hard to scale.
It’s because of that that experts suggest that practical quantum computing is still years away. That’s particularly the case because unlike classical computing, which nearly entirely exists in the form of integrated circuits and transistors on a microchip, there are actually multiple potential hardware solutions for quantum computing, each with their own pros and cons.
Intel’s chip is based on what it knows best: silicon chip technology, using some of the same techniques it uses to make classical computing chips. Silicon is something of an underdog in the quantum computing realm right now, as the technology to make chips with lots of qubits is still in its infancy–Tunnel Falls, for example, has 12 qubits, while IBM has developed a chip built around superconductors that has over 400 qubits. But a recent article in Nature suggested that silicon could have the long-term benefit of being able to scale to an industrial level.
“We’re piggybacking on what we know about transistors and that’s what sets us apart from others,” James Clarke, Intel’s director of quantum hardware told a group of reporters at a briefing earlier this week. “Our goal is to change as little as possible from leading edge technologies to make these,” he added.
In addition to building its quantum systems utilizing as many classical chip manufacturing techniques as it can, Intel is also not limiting its shipbuilding to specialized groups or labs. Clarke says that the new Tunnel Falls chips will be fabricated at its factory in Oregon, which is the company’s largest operations site.
In terms of what Intel’s ultimate plans for its quantum technology is, Clarke says it’s too soon to tell. He says the company plans to make quantum computing a full-stack solution but declined to state whether that would also include quantum-as-a-service solutions like Microsoft and other competitors are developing.
“Let’s build the quantum system first and then see,” he says.
MORE AT FORBES
