Where are the limits that our computers can do, and how big do they have to get? With its new “Bristlecone” Chip that has 72 quantum bits and is designed to be used for quantum computers and computing, Google is trying to reach new barriers, claiming in its Google Research Blog that it will reach “quantum supremacy” aimed at solving the most complex real-world problems. But, what will this “supremacy” actually mean for the future of computing?
Do size and power matter?
Currently, it seems that computer science is taking giant steps toward negating long-held computing postulates, like Moore’s Law, that says that density of transistors has to constantly double.
The researchers and developers like those that came up with Intel’s “Nirvana NNP” are trying to resolve complex problems that are currently beyond our reach with on-the-edge technological solutions.
One of those could be the new “Bristlecone” Quantum computing chip being developed by Google. The claim that this chip can produce 72 qubits (quantum bits) might not be so impressive, but as MIT experts Martin Giles and Will Knight explain, the previous best quantum chip was developed by IBM last year and it was a 50-qubit processor. It takes only a few quantum bits to outrun even the most complex supercomputers.
As Knight explains, quantum computers offer a possibility for calculations that conventional supercomputers are not supposed to do.
According to him, such calculations would make possible easier discovery of new materials, for example.
Skepticism remains
Still, the term “supremacy” does not impress all scientists who, like Simon Benjamin, a quantum expert at Oxford University, are not yet certain what quantum computers will actually be able to do. The problem lies in the fact, as Giles and Knight explain, that the more qubits a quantum machine has, there is a greater possibility of computing errors.
Also, there are problems of keeping the chips in a quantum state, as they are very sensitive to even minute vibrations, known as "noise". There are correctional possibilities with clever software, for example, but takes a toll on computational capacity of a certain machine.
Google though seems to be aware of all the pitfalls and in an attempt to resolve them it plans to use a quantum machine equipped with its new chip to solve an algorithm that is at the edge of the limits current supercomputers have.
Although, if done, this will not be a practical achievement, as Google’s John Martinis, head of this project states, “we’re going to want to show that a quantum machine can do something really useful.”
