john-users - meaning of p/s, c/s and C/s

Follow @Openwall on Twitter for new release announcements and other news

[<prev] [next>] [thread-next>] [day] [month] [year] [list]

Date: Sun, 10 Jan 2016 14:46:56 +0100
From: Patrick Proniewski <patpro@...pro.net>
To: john-users@...ts.openwall.com
Subject: meaning of p/s, c/s and C/s

Hello,

I've read the FAQ about the meaning of the status line, but I think I need a bit more info. 
I'm running a john session with --fork=4 ans --mask=[a-zA-Z0-9][a-zA-Z0-9][a-zA-Z0-9][a-zA-Z0-9][a-zA-Z0-9][a-zA-Z0-9][a-zA-Z0-9][a-zA-Z0-9] that yields to this kind of status lines:

1 0g 0:00:02:16  (8) 0g/s 933.4p/s 3819Kc/s 29306KC/s yoHaaaaa..BqHaaaaa
2 0g 0:00:02:16  (8) 0g/s 937.1p/s 3835Kc/s 29425KC/s OwHaaaFp..RyHaaaFp
3 0g 0:00:02:16  (8) 0g/s 935.3p/s 3829Kc/s 29386KC/s GsHaaaaF..JuHaaaaF
4 0g 0:00:02:16  (8) 0g/s 960.6p/s 3933Kc/s 30182KC/s qmIaaaFU..toIaaaFU

I'm targeting 31394 passwords (traditional crypt(3) [DES 128/128 AVX-16]), with 4092 different salts.

So if I get it correctly: 
- john is testing about 940 passwords per second for each forked process.
- john is computing about 3850000 hashes per second for each forked process, it should be equal to (number of salts)*(p/s): non-salted hashes should give same number for p/s and c/s.
- many passwords are sharing the same salt, so every hash computed can be tested against ~ 31394/4092 passwords. C/s is then equal to (p/s)*(remaining number of passwords), or (c/s)*(number of passwords)/(number of salts).

And, according to the poor rate of ~940 passwords per second, and the mask covering a 62^8 space, I'll be finished in 1871 years.

is that correct?

thanks,
pat

Confused about mailing lists and their use? Read about mailing lists on Wikipedia and check out these guidelines on proper formatting of your messages.