02-04-2013, 08:36 PM
(02-02-2013, 01:19 PM)Dolphin Wrote: I'm personally seeing a slight decrease in performance when upgrading to Catalyst 13.1 (up to 15% on some algorithms) but that's not really a reason to complain. I'm glad that the bugs have been fixed as I've been just turning my clock back for 2 months.
I really don't see the point in having 300 dictionaries. I understood the need to fix the issue of 20 dictionaries together but I really don't see why you would need to use 300 dictionaries at once. If you have multiple small files then just DL a program that combines multiple dictionaries together, there are heaps around.
Hardly a bug in my eyes but you could submit it https://hashcat.net/trac/
First, every one of us works in different ways, and that's advantageous in that it will allow us to, as a community, accomplish more things (crack more passwords, hashes, password types, and so on) than we would working individually.
Second, the core questions here are:
Is having a limit a hard constraint imposed by external forces? If so, if we're at that limit, we're done. If not, continue.
Is having a limit a soft constraint imposed by external forces (speed of cracking decreases even for only 2 dictionaries if the limit is increased, for example)? If so, if we're at that limit, we could hold a reasonable discussion of options, but there's a solid reason for the limit being there. If not, continue.
Do the developers want to impose an arbitrary limit? If so, that is their privilege, though I would personally advocate allowing the program to accept as many as it can before a soft or hard constraint is it.
I would respectfully request that regardless of hard, soft, or arbitrary, that once the limit is stable, it be listed in the help text.
Third, as far as 300+ dictionaries, I can understand techniques that split dictionaries into many subsections - common [words | names | surnames | jargon | sports X | sports Y | etc] , uncommon [...], rare [...], short [...], long [...], various keywalking sets, etc. This by itself could get fairly large; if someone does this for 30 or 40 languages, and regenerates keywalking sets for dozens of different keyboard layouts, it would multiply very quickly.