tag:blogger.com,1999:blog-5832863639484084941.post928472741706199568..comments2018-05-17T12:10:01.932-07:00Comments on Into the symmetry: What the heck is RFC 5114?Antonio Sansonoreply@blogger.comBlogger4125tag:blogger.com,1999:blog-5832863639484084941.post-64529409066389936692017-01-05T02:21:57.171-08:002017-01-05T02:21:57.171-08:00thanksthanksRomilda Garethhttps://www.blogger.com/profile/04571828795230778384noreply@blogger.comtag:blogger.com,1999:blog-5832863639484084941.post-36270298044504718202016-10-30T15:39:06.672-07:002016-10-30T15:39:06.672-07:00"Why (p-1)/2 is not a safe prime?" If t..."Why (p-1)/2 is not a safe prime?" If the least significant [LS] 4 bits of the original prime are xx01 (where x is either 0 or 1), the the number formed by (p-1)/2 is even, and is divisible by at least 2. If the LS four bits are 0001, as is the case here, then the result is divisible by at least 2, 4, & 8. <br /><br />The fact that DH is missing rules for exceptions like these makes the whole process appear to be smoke, mirrors, and deception.JMKelleyhttps://www.blogger.com/profile/01988641797296765198noreply@blogger.comtag:blogger.com,1999:blog-5832863639484084941.post-82505178864690359932016-01-29T04:24:22.448-08:002016-01-29T04:24:22.448-08:00well but why making those factors sooooo small and...well but why making those factors sooooo small and soooo many ?Antonio Sansohttps://www.blogger.com/profile/13409233330078201207noreply@blogger.comtag:blogger.com,1999:blog-5832863639484084941.post-34345678247610472052016-01-29T04:07:18.237-08:002016-01-29T04:07:18.237-08:00I believe the Defense Contractor that wrote the RF...I believe the Defense Contractor that wrote the RFC was worried about compliance with NIST's Standard for Diffie-Hellman-like key exchanges: NIST Special Publication 800-56A. If you look at the original version of SP 800-56A and turn to page 28 you will find a table which REQUIRES a prime order subgroup of size 160,224, and 256 bits for 1024, 2048, and 2048 bit primes respectively. This is the requirement for DSA primes and not necessary for Diffie-Hellman The sizes are technically not optional or minimal but required to be what is specified. As far as I know NIST never enforced this requirement nor did they probably even mean to make it a requirement. They probably just copied it from the DSA standard. The Defense contractor might have thought they'd get an advantage by being the ONLY vendor with products that ACTUALLY satisfied the standard. Such is the world of marketing.Anonymousnoreply@blogger.com