How is Google Scycamore made

Google: A quantum computer shows what is currently possible and what is not

The development of quantum computers is slowly making progress. Google managed to build a freely programmable chip called Sycamore with 54 superconducting qubits, 53 of which work. Test results published ahead of time made headlines shortly afterwards, according to which the so-called quantum superiority was demonstrated. A quantum computer did a calculation in three minutes that would have taken 10,000 years on a supercomputer. It is a statement that is both correct and yet goes to the limits of what can still be called an invoice.

The test that was carried out was specially developed to demonstrate the potential superiority of quantum computers in practice as quickly as possible, without having to develop an actually operational computer beforehand. The chip still has too few qubits to perform calculations such as the decomposition of large numbers into prime factors. In addition, the qubits are too unstable and the calculation steps performed are too imprecise for error-free calculations to be possible.

The benchmark has no practical application

There are options for error correction, but these require even more qubits and also a minimum level of stability and freedom from errors, which has not yet been achieved. The test therefore consists of selecting a random sequence of manipulations of the qubits, executing them several times and recording the results. An attempt is then made to use a supercomputer to calculate the distribution of the results of the previously selected calculation steps.

This method was selected as proof of the so-called quantum superiority, because the simulation of additional qubits and further calculation steps entails exponentially increasing effort. In fact, the quantum chip was able to deliver its results faster than they could be simulated by a computer. As expected, the distance grew with increasing complexity.

  1. Windeit Software GmbH, Berlin, Bad Oldesloe
  2. ING singles in Germany, Frankfurt

The researchers emphasize that these are real calculations in which one qubit has been manipulated with quantum gates alone or in dependence on another qubit. So not only was a quantum state created and its natural physical development measured, as has been the case in many previous experiments. However, the random sequencing of calculations carried out in this way has no practical application.

  1. 1
  2. 2