hashcat Forum

Usually when you run a bruce force attack, hardware can generate & compare millions or even billions passphase per seconds. But in wordlist attack, store billions of passphase in text file already take GBs of harddrive & time to generate & prepare. If i want to run a wordlist attack with trillions of passphase. Is there any trick for this to make it fast or take less space?

You can assemble the passphrases from a one-word-per-line wordlist, using princeprocessor, and then pipe it to hashcat, using one of the prince_*.rule rules files to supply more work to the GPUs.

Combinator Attack (-a 1) lets you merge two wordlists, and you can append rules as well.

Rules are your friends! I've gained a lot of respect for the mad geniuses that came up with one rule.
https://github.com/NotSoSecure/password_cracking_rules

That plus rockyou has a stupid amount of success.

Well tuned masks are lethal as well.
https://blog.netspi.com/netspis-top-pass...-for-2015/
https://blog.korelogic.com/blog/2014/04/...topologies

One consideration when approaching a session is that masks and rules are handled in the GPU, all other generation and feeding is done by the cpu; pre-generation can actually slow you down a lot.

This article was a real eyeopener for me: https://hashcat.net/wiki/doku.php?id=fre...full_speed