back to article Startup Diraq taps GlobalFoundries to forge silicon-based quantum chips

Quantum startup Diraq is to produce sample devices at GlobalFoundries fabs, making it another developer following Intel down the route of using standard CMOS production techniques to build toward full-scale quantum systems. The contract manufacturer is to produce sample devices that combine both quantum and classical circuitry …

  1. JRStern

    I still prefer unobtainium

    It bifurcates the zero point dark matter more efficiently.

  2. Michael Wojcik Silver badge

    Yes, yes, totally believable

    Dzurak [claimed] "Quantum computing can be made much cheaper and more accessible than competing qubit technologies."

    Did he actually say that? Because that is meaningless, and certainly does not inspire any confidence.

    "Commercially relevant" isn't a goalpost. DWave have sold some of their machines, so they're "commercially relevant", even though as adiabatic systems they have rather constrained applications and aren't useful for general QC. (I've seen claims that they could be, if the noise could be reduced enough; but since annealing depends on noise, that seems to be impossible in practice.)

    Quantum-dot approaches to general QC appear to me to be more advanced than topological qubits but not as far along as non-quantum-dot photonics or superconducting qubits, but I confess I know only the barest bones of what's happening in this field. Nonetheless, I'd be very surprised to see a general QC with millions of error-corrected qubits in five years. Or ten. Or fifteen.

    I also wonder if that "99.9%" is for the whole machine, or per qubit. A 0.1% error rate per qubit when you have a significant number of qubits would be problematic for most applications.

    In any case, the real applications for general QC are in chemistry and physics. It'd be great to be able to satisfy those applications, but for the vast majority of IT, QC will probably always be a curiosity. Very few actors will ever be able to make the economics of quantum cryptanalysis work, and then only for specific targets, and by the time they can, we'll have mostly been using PQC for a while. QC will not be useful for things like database search; yes, Grover's Algorithm is A Thing, but it's not a useful thing for most real-world unstructured-search problems. QC does not solve NP-Complete problems. It does not magically make everything faster.

    1. diodesign (Written by Reg staff) Silver badge

      "Did he actually say that?"

      Yes, that's what he was quoted as saying.

      C.

POST COMMENT House rules

Not a member of The Register? Create a new account here.

  • Enter your comment

  • Add an icon

Anonymous cowards cannot choose their icon

Other stories you might like