Suji Desilva: Thanks, Alan. And then – thanks, Alan. And then, the Advantage 2 coming online the next I think a few quarters, several quarters. Can you talk about the addressable opportunity, if there’s any increase there of any new end markets that might be tapped with the higher qubit count machine versus the three you’ve been focusing on in the past?
Alan Baratz: Yes. So I think that the work we did that resulted in the Nature paper two months or so ago really points out where we’re going to see a lot of value in Advantage 2 over the current Advantage system. So in that work as I think I mentioned the last time we got together, we basically proved that our systems do coherent quantum and kneeling. And we proved that while they’re doing coherent quantum and kneeling, they’re delivering a significant speed up over classical on hard optimization problems. However, once we leave the coherent regime, then that speed up starts to wane. And so as we increase coherence, which is a big component of Advantage 2, what we are able to do is to get to optimal solutions faster and on larger problems.
So the fact that Advantage 2 has not only more qubits and more connectivity, both of which allow us to embed larger problems and significantly higher coherence, which means we’re able to get to the solutions faster than in Advantage is what represents opportunity for addressing more applications than we are able to address today. Now, how to map what I just said into specific applications. The easiest way to think about that is that the number, the way we solve commercial problems for our customers is through our hybrid solvers, because all of their applications today are larger than can be directly embedded in our quantum computer. And so the classical part of the hybrid solver takes in the problem, it finds the hard sub [ph] problems and sends those off to the quantum computer to be solved.
Okay. Now, yes, there’s a relationship between the size of the problem that comes in and the size of the problem that goes to the hybrid solvers. And as the problems coming in are larger, the problems that need to go to the quantum computer from the hybrid solver become larger. And to give you two data points, today, if we solve an employee scheduling problem it can have a couple of 100,000 variables associated with it. And the sub problems that are generated out of a couple of 100,000 variables are on the order of what can fit on our 5,000-qubit quantum computer with 15 way connectivity. Now, if we wanted to solve global routing for FedEx, that would require 50 million variables, not a couple of 100,000 variables. And the size of quantum computer we would need to be able to solve that particular problem would be, well, not so much more qubits, because with 7,000 or 10,000 qubits we could deal with it, it would be more connectivity.
And the way we get more connectivity is with more coherence. And so by improving the coherence and improving the connectivity, we’re able to solve those larger and larger problems. Now, Advantage2 won’t get us to full up global routing for FedEx, right. But what it will do is it will take us from the realm of employee scheduling to larger logistics management problems than we’re able to solve today with a current Advantage quantum computer.
Suji Desilva: Okay, helpful, thanks. And thanks for the detailed answer.
Operator: And our next question comes from Richard Shannon from Craig-Hallum. Please go ahead, Richard.
