IBM introduces the groundbreaking 1,121-qubit quantum processor, Condor, but shifts emphasis to enhancing error resistance over size.IBM adheres to a quantum computing roadmap, doubling qubits annually. The Condor chip, unveiled on December 4, features 1,121 superconducting qubits arranged in a honeycomb pattern, succeeding previous record-setting bird-named machines like the 127-qubit chip in 2021 and the 433-qubit one last year.Quantum computers hold the promise of executing computations inaccessible to classical computers, leveraging quantum phenomena like entanglement and superposition. These features enable multiple qubits to exist in various collective states simultaneously.
However, quantum states are notoriously unstable and susceptible to errors. Physicists address this challenge by coordinating multiple physical qubits, encoded in superconducting circuits or individual ions, to collaborate and represent a single 'logical qubit' of information.
In a strategic shift, the company introduces Heron, a chip boasting 133 qubits with a groundbreaking record-low error rate, three times less than its predecessor's quantum processor.
Experts generally assert that cutting-edge error-correction methods demand over 1,000 physical qubits per logical qubit. Achieving practical computation capabilities would consequently necessitate machines with millions of physical qubits.
Recent months have seen physicists expressing enthusiasm for an alternative error-correction approach known as quantum low-density parity check (qLDPC). This method holds the potential to reduce the required qubit count by a factor of 10 or more, as outlined in a preprint by IBM researchers. In response, IBM announces a shift in focus towards developing chips capable of accommodating a few qLDPC-corrected qubits within approximately 400 physical qubits. The strategy involves networking these chips together for enhanced quantum computing capabilities.
According to Mikhail Lukin, a physicist at Harvard University, the IBM preprint is "excellent theoretical work." However, Lukin acknowledges the formidable challenge of implementing this approach with superconducting qubits, suggesting that it might take years before a proof-of-concept experiment is attempted on this platform. Lukin and his collaborators have undertaken a similar study, exploring the potential implementation of quantum low-density parity check (qLDPC) using individual atoms instead of superconducting loops.
The challenge lies in the qLDPC technique, necessitating each qubit to be directly linked to at least six others. Conventional superconducting chips have qubits connected only to two or three neighbors. However, Oliver Dial, Chief Technology Officer of IBM Quantum and a condensed-matter physicist at IBM’s Thomas J. Watson Research Center, reveals that IBM has a solution. The company plans to enhance the design of its quantum chips by adding a layer to accommodate the additional connections mandated by the qLDPC scheme.
IBM reveals a new quantum research roadmap, envisioning practical computations like simulating catalyst molecules by the end of the decade. Oliver Dial expresses the significance of this achievement, stating that what was once a distant dream is now coming closer, and the outlined path is seen as enormous progress.
doi: https://doi.org/10.1038/d41586-023-03854-1
References
