Way back in 1965, Intel CEO Gordon Moore published a paper predicting that computing power (or number of transistors per microchip) would double each year for at least a decade to come. More than fifty years later, that prediction has become the basis for what we commonly refer to as “Moore’s Law” — which states that computing power will increase exponentially year over year, with the cost of that performance trending in the opposite direction. In recent years, it became clear that we may finally be reaching the physical limits of Moore’s Law, suggesting that his 10-year prediction turned forty-year reality was finally coming to a close, and people in the tech industry began looking for computing alternatives that weren’t hindered by the physical limits of standard microchips.
That pursuit led scientists to the shadowy world of quantum mechanics, where reality sounds like science fiction and science fiction comes off as pretty tame. The computers we use today may offer a highly streamlined interface, but the dirty work behind the scenes is still conducted via bits that register as either a 1 or a 0. Quantum computing turns that concept on its head, using qubits that can exist in multiple states at once instead of traditional bits. As a result, quantum processors can work with twice as much information and handle complex calculations not just faster than modern super computers… but much, much faster.
How much faster, you ask? According to an announcement made by Google on Friday, their 54-qubit chip quantum computer called Sycamore just completed a problem in 3 minutes and 20 seconds that would have taken the world’s most powerful supercomputer 10,000 years to solve. That claim isn’t just bluster either — it comes along with a 77-author, peer reviewed paper confirming the success of their test and the validity of what they call, “Quantum supremacy.”
“It is likely that the classical simulation time, currently estimated at 10,000 years, will be reduced by improved classical hardware and algorithms,” Brooks Foxen, a graduate student researcher in Martinis’ lab, said in a statement. “But since we are currently 1.5 trillion times faster, we feel comfortable laying claim to this achievement,” he said, referring specifically to the “quantum supremacy” claim.
A pair of bits can represent just one possible representation of 1s and 0s at a time (00, 01, 10 or 11), whereas qubits can exist as both 1s and 0s simultaneously. That means a pair of qubits doesn’t have to represent just one pair of numbers as shown above, but rather can store all four at once. Because of that, computing power increases exponentially with each additional qubit: three qubits would store 8 combinations, four would store 16. That means the 54 cubits powering Google’s Sycamore can work with more than 10,000,000,000,000,000 (10 quadrillion) combinations at once.
“We’re looking forward to giving humanity a new tool for solving what would otherwise be impossible problems,” Erik Lucero, a hardware engineer at Google’s quantum research lab, said.
The possibilities for this kind of computational power are endless, but Google has its sights set on a few possibilities first. Sycamore or computers like it could be used to solve complex problems with lots of variables almost instantly, like optimizing package delivery routes or logistical processes. A computer this powerful could lead to utterly unhackable forms of encryption or entirely new advancements in molecular scale materials. One thing Google hasn’t spoken at length about but seems inevitable is the use of these powerful computers in conjunction with machine learning and artificial intelligence. An AI powered by a supercomputer is one thing, but one with processing power that’s thousands of times more powerful than a supercomputer is another entirely.
“It stands to reason this could be a very valuable resource for machine learning,” Google’s Hartmut Neven told The Financial Times. “We are playing around with this.”
Way back in 1965, Intel CEO Gordon Moore published a paper predicting that computing power (or number of transistors per microchip) would double each year for at least a decade to come. More than fifty years later, that prediction has become the basis for what we commonly refer to as “Moore’s Law” — which states that computing power will increase exponentially year over year, with the cost of that performance trending in the opposite direction. In recent years, it became clear that we may finally be reaching the physical limits of Moore’s Law, suggesting that his 10-year prediction turned forty-year reality was finally coming to a close, and people in the tech industry began looking for computing alternatives that weren’t hindered by the physical limits of standard microchips.
That pursuit led scientists to the shadowy world of quantum mechanics, where reality sounds like science fiction and science fiction comes off as pretty tame. The computers we use today may offer a highly streamlined interface, but the dirty work behind the scenes is still conducted via bits that register as either a 1 or a 0. Quantum computing turns that concept on its head, using qubits that can exist in multiple states at once instead of traditional bits. As a result, quantum processors can work with twice as much information and handle complex calculations not just faster than modern super computers… but much, much faster.
How much faster, you ask? According to an announcement made by Google on Friday, their 54-qubit chip quantum computer called Sycamore just completed a problem in 3 minutes and 20 seconds that would have taken the world’s most powerful supercomputer 10,000 years to solve. That claim isn’t just bluster either — it comes along with a 77-author, peer reviewed paper confirming the success of their test and the validity of what they call, “Quantum supremacy.”
“It is likely that the classical simulation time, currently estimated at 10,000 years, will be reduced by improved classical hardware and algorithms,” Brooks Foxen, a graduate student researcher in Martinis’ lab, said in a statement. “But since we are currently 1.5 trillion times faster, we feel comfortable laying claim to this achievement,” he said, referring specifically to the “quantum supremacy” claim.
A pair of bits can represent just one possible representation of 1s and 0s at a time (00, 01, 10 or 11), whereas qubits can exist as both 1s and 0s simultaneously. That means a pair of qubits doesn’t have to represent just one pair of numbers as shown above, but rather can store all four at once. Because of that, computing power increases exponentially with each additional qubit: three qubits would store 8 combinations, four would store 16. That means the 54 cubits powering Google’s Sycamore can work with more than 10,000,000,000,000,000 (10 quadrillion) combinations at once.
“We’re looking forward to giving humanity a new tool for solving what would otherwise be impossible problems,” Erik Lucero, a hardware engineer at Google’s quantum research lab, said.
The possibilities for this kind of computational power are endless, but Google has its sights set on a few possibilities first. Sycamore or computers like it could be used to solve complex problems with lots of variables almost instantly, like optimizing package delivery routes or logistical processes. A computer this powerful could lead to utterly unhackable forms of encryption or entirely new advancements in molecular scale materials. One thing Google hasn’t spoken at length about but seems inevitable is the use of these powerful computers in conjunction with machine learning and artificial intelligence. An AI powered by a supercomputer is one thing, but one with processing power that’s thousands of times more powerful than a supercomputer is another entirely.
“It stands to reason this could be a very valuable resource for machine learning,” Google’s Hartmut Neven told The Financial Times. “We are playing around with this.”
For now, however, this burgeoning technology is still years away from having an appreciable impact on the world at large — but with such a dramatic leap in computer power in our grasp, it seems the era of Moore’s Law may be over, but the age of dramatic computational advancements may only be just beginning.
Support the SOFREP community with a lifetime membership. Sign Up Now
COMMENTS
There are on this article.
You must become a subscriber or login to view or post comments on this article.