Morj

@svuorela@helvede.net Trick question because you can break 1024 bit RSA with a contemporary quantum computer already, but it's going to run as only a randomness oracle Increase this to 3096 and then it becomes interesting

@svuorela@helvede.net Exactly my point. So it's better to pick a benchmark that can't be solved with this approach. Otherwise you'll get into the Kolmogorov's problem where the irrelevance of the QC to the computation becomes harder and harder to gauge

@svuorela@helvede.net I'll remind you that the quantum circuit doesn't do full computation, it's used as an oracle for some functionalities in the big algorithm ran on a normal processor. Depending on how you structure the algorithm, it may be hard to impossible to say if the QC result is useful to the computation or not. It's easier to change the initial problem than to define what constitutes a quantum computer that is doing a useful function.