Posts by charles_hermann 2 publicly visible posts • joined 6 Jul 2015
It's not scalable, in the sense that a scalable version of what is described would break the laws of both thermodynamics and quantum mechanics. I'll freely admit I don't know how to do that :)
If you believe that some different technology (valves -> silicon) might one day solve NP complete problems in P-time, that's up to you. I'd like to see the evidence, but this talk / paper doesn't offer any.
We're all already betting more that 100 EUR that this is nonsense ;)
A scalable method of solving any NP-complete problem in polynomial time / space would take out most forms of cryptography - including those relied upon by banks. Even though their preprint has been around since last December, I don't see widespread panic & stashing of gold coins under beds just yet!
If you look at their arXived paper, it's clear that scalability requires -complete- absence of noise, as the amplitudes of certain crucial oscillations decreases exponentially. No engineering magic is going to solve this one - even if they cool it to close to absolute zero they will run into quantum uncertainty before they can attack a practical problem.
They claim that a different encoding might get around this, but the one they reference only offers a quadratic advantage, which is a long way from what they need.
It's not scalable.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
