[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <CAFs9wnUrXQi7u+Ey9cG1VDNy0HL-6j4oPB_Z-VvrY_NTnNfMFQ@mail.gmail.com>
Date: Thu, 11 Mar 2021 17:13:36 +0100
From: Michał Majchrowicz <sectroyer@...il.com>
To: john-users@...ts.openwall.com
Subject: Re: Multi-gpu setup
> Right, this will not run any forked processes. There will be just one
> process per GPU. The range of node numbers merely allocates a portion
> of work to that process.
Thanks for clarification. It confused me since I mostly was testing
john on dual gpu systems and there it complains about lack of fork and
my mind assumed -se is some magical flag that "disables" fork/node
dependance :) I am already using different sessions for each john I
start (even on single machine) but always use full --session so didn't
notice it's basically same thing as -se. BTW your mask generation
trick worked and from 5 des I got back to 2. Cracked 3 used simply two
capital letters. For 2 des hashes my current setup currently needs
around a year to try full ?a so those "mixed masks" really helped :)
Anyway it is (as expected) clear that passwords people invent are not
random and some patterns can be applied to speed up the process. It
might be worth to use some neural networks to figure out those
patterns but still a lot of work before this :)
Regards.
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.
