Date: Sat, 24 Oct 2015 10:14:29 +0200 From: magnum <john.magnum@...hmail.com> To: john-users@...ts.openwall.com Subject: Re: The meaning of status output of John On 2015-10-24 02:55, Ckcheng wrote: > I'm trying to test opencl speed of John the ripper on OSX. However, the status output of John is a little bit confusing. > > Here is my output during the run: > 4g 0:00:10:47 0.006182g/s 50500Kp/s 50500Kc/s 1157GC/s tofu4ta6..tucijisa > > Can someone tell me which statistic is equivalent oclhashcat's "Speed" field? From doc/FAQ: "The four speed metrics are as follows: g/s is successful guesses per second (so it'll stay at 0 until at least one password is cracked), p/s is candidate passwords tested per second, c/s is "crypts" (password hash or cipher computations) per second, and C/s is combinations of candidate password and target hash per second." > (I just want to know how many MD5 hashes were tried per second) I think that is ambiguous as written. It could mean c/s or C/s. I think what *Hashcat reports nowadays is equivalent to JtR's p/s. That figure will decrease with number of salts, so eg. with 10 salts the p/s will be a tenth of c/s. With just one salt (or unsalted hashes) p/s and c/s are always the same like in your example. Please note that for good OpenCL speeds you should run a newer version of Jumbo than the latest released (eg. a bleeding-jumbo snapshot from https://github.com/magnumripper/JohnTheRipper/archive/bleeding-jumbo.tar.gz) and to compete with *Hashcat at "fast hashes" you should use mask mode (alone or as an accelerator for some other mode). Regarding OSX there seems to be quite some regression in driver quality from Yosemite to El Capitan. Some formats now crash or fail self-test, whereas the Yosemite drivers (and late Mavericks iirc) was 100% working for me. I hope Apple will fix that soon. magnum
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.