Intel has entered the race to develop quantum computing technology, recognizing that before it can revolutionize computing, significant advancements must be made. The company has created a quantum processor known as Tunnel Falls, which it plans to offer to research labs aiming to make practical strides in this groundbreaking field.
Tunnel Falls, announced recently, contains 12 qubits, the fundamental units of data processing in quantum computing. This achievement signifies a crucial milestone in Intel’s endeavor to surpass its competitors by developing advanced hardware for quantum computing.
Unlike many rivals in the quantum computing arena, Intel manufactures its qubits using individual electrons within computer chips that are similar to those powering millions of PCs. Although the company is currently behind industry leaders like IBM, Google, Quantinuum, and IonQ, Intel believes that leveraging its expertise in conventional chip technology will ultimately lead to faster progress.
Jim Clarke, Director of Quantum Computing Hardware at Intel Labs, emphasized the natural inclination to utilize existing tools rather than creating entirely new ones. Intel manufactures its own quantum computing chips at its D1 fab facility in Oregon.
While individuals may not purchase their own quantum computers, the technology’s potential impact on various aspects of life is substantial. Financial services companies seek more profitable investments, materials science researchers aim to enhance battery performance, pharmaceutical companies strive to design superior drugs, and governments endeavor to decipher encrypted communications of adversaries—all of which can be facilitated by quantum computing.
Addressing challenges that conventional computers cannot handle, quantum computing harnesses the peculiar properties of the ultrasmall world. Although practical quantum computers are still some years away, physicists and engineers have consistently made progress in this field.
Intel, renowned for its expertise in large-scale manufacturing, intends to accelerate quantum computing advancement by producing numerous quantum chips, referred to as quantum processing units (QPUs). The University of Maryland, one of the institutions benefitting from a US government initiative to expedite quantum computing progress, will utilize Intel machines in its research efforts.
The competition in the field of quantum computing
One of the remarkable aspects of quantum computing is the wide range of approaches being pursued by different companies. Intel, for example, is utilizing electrons and their quantum mechanical property known as spin, which is analogous to the spinning direction of a top. On the other hand, IBM and Google are employing small superconducting electrical circuits, while IonQ and Quantinuum manipulate charged atoms confined in a trap. Other approaches involve neutral atoms and even photons, the most fleeting particles.
According to Seth Lloyd, a quantum computing pioneer and MIT researcher, at an extremely small scale, quantum mechanics becomes the dominant force in physics, allowing almost anything to serve as a qubit. The challenge lies in finding ways to manipulate these entities effectively for computation.
In contrast to the traditional computer chip market, where different companies compete head-to-head, the race in quantum computing involves diverse technologies competing against each other. It’s akin to a horse competing against a falcon, a motorcycle, and an Olympic sprinter.
Intel is confident in its chosen approach. While Tunnel Falls is already in production, the company is currently finalizing the design for its successor and has even started planning the subsequent model. Although the current 12-qubit configuration is only a small fraction of what is required for practical quantum computers, Intel has adopted a simple approach that allows for rapid improvement and sustained progress over the coming years.
The next significant milestone for Intel is reaching a few thousand qubits. This quantity would enable quantum computer engineers to address the frequent errors that occur during qubit operations. According to Jim Clarke, this milestone is likely to be achieved within the next three to five years. However, it won’t be until the early to mid-2030s, when Intel anticipates having a million qubits, that the world-changing potential of quantum computing will be fully realized.
Intel’s focus extends beyond quantum processing units (QPUs) alone. The company is also investing in the development of crucial data links that connect each qubit to the external world. While current quantum computers may resemble sophisticated chandeliers with metal communication conduits, this bulky design won’t be viable for systems with thousands or millions of qubits. Intel believes that its control chips and chip interconnect technology will be essential components of a comprehensive quantum computing system.
There is a multitude of competitors in the quantum computing field
IBM, one of Intel’s major competitors, is already offering multiple quantum computers with 127-qubits for research and commercial purposes, and they have a functioning machine with 433-qubits as well.
Jerry Chow, the leader of IBM’s quantum computing hardware division, stated that their plan involves scaling up to hundreds of thousands of qubits using superconducting qubits. IBM is actively working on various quantum computer chips, each with unique code names such as Egret, Heron, Condor, and Crossbill. These chips are designed to explore new technologies that aim to reduce errors and enhance the connections between qubits, which are crucial for the performance of quantum machines.
IBM has also made notable progress in their research. Recently, they achieved recognition in the prestigious journal Nature for their work demonstrating that their 127-qubit Eagle quantum computing chip outperforms traditional computers in simulating materials physics, including phenomena like magnetism.
In contrast, Intel decided against pursuing the superconducting qubit approach. Instead, they opted for spin qubits, which are much smaller, enabling them to fit around 25,000 qubits on a single 300mm silicon wafer in their chip fabrication plant. When faced with challenges during quantum chip manufacturing, Intel focuses on adapting the qubit technology to fit into traditional chip manufacturing processes, rather than the other way around.
Difference in approach between Intel and some of its competitors regarding quantum computing
Others, however, remain unconvinced and hold different opinions regarding Intel’s approach to quantum computing. Google, for instance, is committed to using superconducting qubits and believes they are the leading technology for future quantum supercomputers. Google highlights the processing speed of superconducting qubits and their progress in error correction, which enables more accurate calculations over longer durations. They envision scaling their technology to large-scale, error-corrected machines that can be widely utilized.
On the other hand, IonQ’s CEO, Peter Chapman, expresses skepticism about Intel’s approach, deeming it too rigid for practical and large-scale quantum computers. IonQ is pursuing ion trap machines, which involve moving charged atoms to facilitate qubit interactions for computation. According to Chapman, the fixation of qubits onto the surface of a chip introduces significant complexity to computations. He asserts that traditional silicon-based processors, which have been successful in classical computing, are not the optimal solution for the quantum era.
These divergent viewpoints on the best approach to quantum computing may be resolved over time as the technology evolves and larger-scale machines are developed. Intel’s strategy capitalizes on its manufacturing expertise, leveraging its experience in building some of the most complex electronic devices worldwide. Jim Clarke emphasizes the advantage of having a fabrication facility like Intel’s, recognizing that not every competitor possesses such capabilities.