Now, the company shows that it can put actual logical qubits in the same hardware variant. In previous versions of that equipment, the resonator cavity had a single post and supported a single frequency. The new iteration had two posts and two frequencies. Each of these frequencies creates their own quantum resonators in the same cavity with their own set of modes. “This ensemble of photons within this cavity creates a logical qubit,” Remir told ARS.
Additional quantum information that can be stored in the system allows for identification of errors that are more complex than photon loss.
Catch it but does not fix the error
The company has conducted two experiments with this new hardware. First, we tested our ability to perform multiple rounds of error detection on data stored in logical Qubit, acting essentially like quantum memory and retaining the information stored there. The system was corrected due to photon loss, but no other errors were corrected. These occurred at a rate just over 2% each round, so by the time the system reached its 25th measurement, many instances had already encountered an error.
For the second time, the company repeated the process and discarded the instance where the error occurred. In almost every example, it meant that the results were discarded long before they passed 20 measurements. However, at these later stages, none of the remaining instances were in the wrong state. This indicates that all errors have been fixed successfully, that is, that the team has not tried, can fix all detected issues.
“When you do this, there are no errors left,” Remir said. “And this builds confidence in this approach. This means that when you build next-generation code that can fix these errors detected with this two-mode approach, you need a flatline that doesn’t cause errors over a long period of time.”
