Why it matters: As the quantum future looms closer, hundreds if not thousands of companies and research groups race towards constructing the first quantum computer that can outperform traditional supercomputers. However, the competition is not just between organizations, it's also between competing methods of quantum computing.
IonQ was founded on a gamble that 'trapped ion quantum' computing could outperform the silicon-based quantum computers that Google and others are building. As of right now, it does. IonQ has constructed a quantum computer that can perform calculations on a 79-qubit array, beating the previous king Google's efforts by 7 qubits.
Their error rates are also the best in the business, with their single-qubit error rate at 99.97% while the nearest competitors are around the 99.5 mark, and a two-qubit error rate of 99.3% when most competitors are beneath 95%. But how does it compare to regular computers?
According to IonQ, in the kinds of workloads that quantum computers are being built for, it's already overtaking them. The Bernstein-Vazirani Algorithm, a benchmark IonQ is hoping will take off, tests a computer's ability to determine a single encoded number (called an oracle) when the computer can only ask a single yes/no question.
When the algorithm is run for every number between 1 and 1023, a conventional computer gets a 0.2% success rate. IonQ's quantum computer gets a 79% success rate.
"After two years of work, our against-the-grain bet is paying off," IonQ's CEO, Christopher Monroe, believes trapped ion quantum computing is the best bet. "The IonQ System is robust and industrial strength. Even at this early stage, the results show the ion trap design has all the advantages we expected and more."
All quantum computers "isolate and manipulate quantum systems to create quantum versions of computer bits, called qubits" reads IonQ's website. Quantum computers replace the traditional 0 or 1 logic gates processors rely on and replace them with 0 and 1 quantum gates, which are simultaneously 0 and 1 during calculations but output 0 or 1. This funky math has the potential to reinvent computing in fields like chemistry, medicine, energy, logistics and future fields like AI.
The specific 'trapped ion technology' the IonQ's quantum computer relies on replaces the supercooled silicon that Google, IBM and Rigetti use with ytterbium, a silvery rare earth metal. The ionized ytterbium is suspended in an oscillating electromagnetic field, where it's manipulated by engineers who program the lasers that input, store and retrieve information.
While 'trapped ion' quantum computing still has some hurdles to overcome, namely slow operation times and massive sizes, the accuracy and scalability of the technology means that IonQ will be letting companies use its computer sometime next year. It's also got a peer-reviewed journal article on the developments that will be published in the coming months.
Quantum supremacy, the moment that the best quantum computer is better than the best traditional computer, is approaching rapidly. While even IonQ will admit that they don't know what the "killer app of quantum computers" is yet, it doesn't seem like it'll be too long before we're all taking it for granted.