Here https://news.ycombinator.com/item?id=10698317 https://news.ycombinator.com/item?id=10698317 and also more importantly here https://news.ycombinator.com/it

Here https://news.ycombinator.com/item?id=10698317 https://news.ycombinator.com/item?id=10698317 and also more importantly here https://news.ycombinator.com/item?id=10707442 https://news.ycombinator.com/item?id=10707442 TL;DR: Google researchers demonstrated that DWAVE2X quantum annealing against one classical (i.e. normal computer) algorithm (on one CPU core) - simulated annealing, was asymptotically (i.e. more than just by constant factor) better, but was not against another classical algorithm quantum monte carlo (QMC) that is, the difference was constant. In fact, it was 10^8 times faster and that is generating all the hype. Now, it was mentioned in the paper that Selby algorithm on classical computer would actually produce better results than DWAVE2X but it was not compared against it (maybe because it would minuscule the amazing 10^8 times difference). Also it is worth mentioning that special solutions in silicon (ASIC) are demonstrated to produce about 10^6 faster results compared against the same algorithm on a single CPU core (it does not follow that such speed up is possible for every classical solution).