News for hackers. Stuff that matters.
Re:A riddle wrapped in a curve by pbsd on Thursday October 22, 2015 @12:43PM (#10433223)
Loosely speaking, you need n + epsilon qubits and 4n^3 operations to break RSA-n, whereas you need ~6n qubits and 360n^3 operations to break ECC-n. For n = 256, you need around 1536 qubits, whereas you need at least 3072 for RSA-3072.

This suggests a criterion to reject P-256: it needs fewer than 2048 qubits to break. P-384 is above this threshold, at around 2.5k qubits. Hey, it's as good speculation as any.

Re:A riddle wrapped in a curve by api on Thursday October 22, 2015 @01:20PM (#10433472)
Does this mean that barring a quantum-hard alternative we could get effectively quantum-hard crypto by using crazy key sizes like RSA-131072 or ECC-4096?
Re:A riddle wrapped in a curve by pbsd on Thursday October 22, 2015 @01:29PM (#10433528)
Yes, though it would be quite impractical. See for example http://cr.yp.to/talks/2010.05.28/slides.pdf
Re:A riddle wrapped in a curve by Tomte on Thursday October 22, 2015 @02:30PM (#10433969)
"Key almost fits on a hard drive" :-)