Date: Sun, 18 Dec 2011 20:54:29 +0400 From: Solar Designer <solar@...nwall.com> To: john-dev@...ts.openwall.com Subject: Re: 1.7.9-jumbo On Sun, Dec 18, 2011 at 08:46:34PM +0400, Solar Designer wrote: > Re-introducing the two "&& md4_size < PLAINTEXT_LENGTH" checks restores > almost the original performance. I suspect that the "target < targetEnd" > were needed as well. We probably incur extra loop iterations on --test > without these. I don't know whether this affects actual cracking runs as > well or not. Perhaps not. "./john -fo=nt -i=all8 hashes-all.txt-20.nt" for the two versions: stolorat (eissat) stolovol (divate) sheastar (deatrick) guesses: 1578 time: 0:00:01:00 0.00% c/s: 274653M trying: cysq2r15 - cysq2r2m stolorat (eissat) stolovol (divate) sheastar (deatrick) guesses: 1578 time: 0:00:01:00 0.00% c/s: 276034M trying: mshtbin7 - mshtbite (manually interrupted at 1 minute). The john.pot files match exactly. BTW, the raw speed is around 18M c/s: 275000/15928 = 17.2652 275000/(15928-1578) = 19.1638 Alexander
Powered by blists - more mailing lists
Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.