Near the bottom of
this page you explain that the optimum way to use hashcatplus is for us to sort our wordlists by line length into separate files. I understand the explanation and I can see why it would help.
However I wonder what “run oclHashcat-plus in directory mode†means ? I am guessing that you mean run each line length at a time, separately. Although I am wondering if there is a specific command called “directory mode†that I am not using. I cannot find such a command in the --help.
I think "directory mode" just means that you point hashcat-plus to the directory cointaining the wordlists instead of pointing it to a list of dicts.
(11-18-2011, 10:22 PM)SmilingWolf Wrote: [ -> ]I think "directory mode" just means that you point hashcat-plus to the directory cointaining the wordlists instead of pointing it to a list of dicts.
Oh I see I didn't know you could do that, I thought it meant one text file at a time containing the same length passwords. I'll give it a try.
Did you read your PM ? I am waiting to hear from you so I can make the wiki page about your script. I just need a few details and your permission to put it up there thats all.
Thanks.
not exactly. you know there is also an stdin-mode. so we have wordlist-mode and stdin-mode. we should mention that on wiki maybe. the wordlist-mode has some advantages. 1. it can fully parse the keyspace at startup and therefore calculate an ETA. 2. it does not block the user-input device - in wordlist-mode the stdin is used for user-input and therefore the user can send commands to pause/resume.
So I am guessing the best way to explain this to a new user would be that…
The optimum way to run hashcatplus is to sort all words into individual text files by length. Then run each text file separately through hashcatplus in wordlist-mode and there is no specific (files are pre sorted into length) command to be typed into hashcatplus when doing this.
Note:
I suspect the most optimum way would be to do the above and then use rules. This fits in nicely with my idea about lists….
well. in theory you are right. but if the user runs some serious big masks it would require to much space on hdd. yes, we are talking about TB's
I have done some tests and I hope this isn’t the placebo effect !!
Using alphabetically sorted word list length ranging from 8 – 15 direct attack on a WPA handshake capture, I get a fairly sustained rate of…
90589 real and 92354 GPU
Using the same list pre sorted words by length I actually saw 101.6k GPU !!!
The “real†figure was still around the 90673 but what about that 101.6k ? I haven’t seen that before….ever !
This dropped back to 90776 real 92085 GPU minutes later. My temp was constant and hovered around the 71c – 72c area.
I am using XP Pro SP3 and a HD5870 card. I was also using the toggle 2 rule in an effort to optimise the GPU.
Just an update to this, different capture files are reacting to this differently. Some don’t seem to improve speed and some do. I am not sure if it is the capture or what but I haven’t seen a change in performance like this since I dropped (a certain commercial software) for hashcatplus !!!
wow, 100k with HD5870? is it overclocked? looks like record for this card...
btw, nice avatar
(11-20-2011, 12:24 AM)atom Wrote: [ -> ]wow, 100k with HD5870? is it overclocked? looks like record for this card...
Is it really !!
It doesn't stay there long when I refresh (pressing s) it quickly drops back to around 91, 92, or 93,000 where it stays mostly.
I have "tweaked" it a little, but not much. I was wondering if that was a bad idea when WPA cracking ? Its one thing when a overclocked graphics card misses a few pixels on a game but when making these calculations I often wonder if errors creep in and knowing my luck just at the time the actual password is being checked !!
I have tested occasionally with known caps and passwords and it has never skipped past a know one. Any advice ?
(11-20-2011, 12:24 AM)atom Wrote: [ -> ]btw, nice avatar
You're jealous aren't you !! Ha ha !!