phc-discussions - Re: An argument for Catena-2

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

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

Message-ID: <CAOLP8p4o-AXO3ALH8XJ4nLhEHYDes=WyuscYggHtft7YQ0+YEQ@mail.gmail.com>
Date: Sun, 26 Jan 2014 15:53:34 -0500
From: Bill Cox <waywardgeek@...il.com>
To: discussions@...sword-hashing.net
Subject: Re: An argument for Catena-2

On Sun, Jan 26, 2014 at 3:47 PM, Bill Cox <waywardgeek@...il.com> wrote:
> Also, if instead of a simple linear chain in the first row, if we put
> a Catena-3 graph, the recomputation penalties are much higher.  I get
> over 140X for this modified Catena-2 graph when trying to pebble it
> with half the memory.

With a Catena-8 graph in the first row, my poor pebbling algorithm
throws in the towel and goes home.  I can only get small graphs to
pebble in a reasonable time, and the penalty is many thousands times
more computation for 50% of the memory.

Is this enhanced Catena-2 as good or better defending against TMTO
attacks as Catena-3?

Bill