Intel has made a major breakthrough toward quantum computing with the delivery of a 49-qubit test chip called “Tangle Lake” to its research partner QuTech.
The company said it thinks more than one million qubits would be necessary for a working quantum computer, but that the new chip should help researchers develop error-correction techniques, an important aspect of complex quantum systems.
The chip is named after a chain of lakes in Alaska, a reference to the extremely cold temperatures needed to keep quantum bits stable and the “entangled” quantum state that allows them to represent multiple values at the same time.
Tangle Lake follows two months after Intel delivered a 17-qubit superconducting test chip.
In a keynote at CES in Las Vegas, Intel chief executive Brian Krzanich said quantum computers could help solve problems in drug development, financial modelling and climate forecasting.
Intel said it is also looking into another type of system that uses spin qubits, which it said are much smaller than the superconducting qubits used in Tangle Lake. The company said it has already developed a fabrication flow for spin qubits using its 300nm technology.
IBM, another company investing in the field, is demonstrating a 50-qubit quantum computer at CES.
Krzanich also highlighted Intel’s work on neuromorphic computing, a type of computer processing based on the functioning of the human brain.
Intel Labs has developed a neuromorphic chip called Loihi that it said could make machine learning more power efficient, and plans to deliver it to university and research institutions in the first half of this year.
The chip could be used in areas such as smarter security cameras and the infrastructure needed for real-time communication with autonomous vehicles.
“This has been a major research effort by Intel and today we have a fully functioning neuromorphic research chip,” Krzanich said in the keynote.
The wide-ranging talk also covered Intel’s work in fields such as autonomous cars, bolstered by its $15 billion (£11bn) acquisition of Israel-based Mobileye last year.
Mobileye’s Road Experience Management (REM) has been built into 2 million vehicles sold by BMW, Nissan and Volkswagen and are to begin actively collecting crowdsourced data this year. The data is to be used to build high-definition maps for future autonomous vehicles as well as nearer-term car navigation systems.
Krzanich said Intel is working with car maker SAIC and digital mapping company NavInfo, both from China, to collect crowdsourced map data in the country. SAIC is to build vehicle automation products using Mobileye technology, Krzanich said.
Intel said a new automated driving platform will use automotive-grade Atom processors with Mobileye EyeQ5 chips to power automated and autonomous driving systems.
Other topics covered in the presentation included the demonstration of Intel-based flight control technology for a vertical take-off passenger aircraft called the Volocopter and Intel’s move into immersive filmmaking with a studio that facilitates the use of immersive and VR filming techniques.
Krzanich said Intel will use its True VR technology to deliver live and on-demand VR broadcasts of 30 Olympic events at the PyeongChang 2018 Olympic Winter Games.
Also at the show Intel announced its 8th-generation Core processor, which combines AMD Radeon RX Vega M graphics and 4GB of second-generation, high-bandwidth memory (HBM2).
The chips are used in four-inch-square mini-PCs called NUCs, demonstrated at CES, which are aimed at gaming and VR applications. The NUCs are set for availability this spring.
Would a rose by any other name smell just as sweet? Decide for yourself with our tech company name quiz!
Trolls beware. Twitter releases feature that will deliver a 'reconsider prompt' for users, if they…