3 ms·
Good analogy; and quite accurate. It's not just the number of bits that affects signal to noise ratio, though... It's also the complexity of the calculation; m
by smaddox 10y ago
Good analogy; and quite accurate.
It's not just the number of bits that affects signal to noise ratio, though... It's also the complexity of the calculation; more gate operations and longer storage in memory leads to more quantum decoherence. And AFAIK, each type of calculation must be implemented in hardware, because reversible quantum gates must be used. This means you would need a specialized chip for each algorithm. I haven't confirmed this understanding with someone that specializes in QC, but I don't see a way around it, other than to combine many algos onto a single chip, sharing as many gates as possible, but degrading S/N further.
- andai 10y agoRe: need custom hardware for each algorithm Something tells me, for an intelligence agency, building a quantum computer only for the purpose of breaking cryptographic keys would be a worthwhile investment. An investment they have already made: The effort to build “a cryptologically useful quantum computer” -- a machine exponentially faster than classical computers-- is part of a $79.7 million research program called “Penetrating Hard Targets.” article https://www.washingtonpost.com/world/national-security/nsa-seeks-to-build-quantum-computer-that-could-crack-most-types-of-encryption/2014/01/02/8fff297e-7195-11e3-8def-a33011492df2_story.html https://www.washingtonpost.com/world/national-security/nsa-s... leaked document http://apps.washingtonpost.com/g/page/world/a-description-of-the-penetrating-hard-targets-project/691/ http://apps.washingtonpost.com/g/page/world/a-description-of...