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: <CAOLP8p4aevsd_Hzqr+51zTwWqg9qwWaRb9O6T+G8CS6J0A_bKA@mail.gmail.com>
Date: Sat, 5 Sep 2015 23:24:30 -0700
From: Bill Cox <waywardgeek@...il.com>
To: "discussions@...sword-hashing.net" <discussions@...sword-hashing.net>
Subject: Re: [PHC] Re: Where do authors get these numbers?

Seriously, you reported TwoCats on one thread to only generate1 GiB/s of
bandwidth?  This is simply an error.

On Sat, Sep 5, 2015 at 11:18 PM, Dmitry Khovratovich <khovratovich@...il.com
> wrote:

> There was a mistake related to our misunderstanding of Twocats
> specification, and this claim was removed in later versions of this report.
>
> Sent from my iPad
>
> On Sep 6, 2015, at 7:24, Bill Cox <waywardgeek@...il.com> wrote:
>
> BTW, the same paper says there's a 12X reduction in area-time cost against
> TwoCats.  They're smoking something good, and I want some!
>
> On Sat, Sep 5, 2015 at 10:09 PM, Bill Cox <waywardgeek@...il.com> wrote:
>
>> Just saw this:
>>
>>
>> https://www.cryptolux.org/mediawiki-esc2015/images/3/32/Proceedings_esc2015.pdf#page=13
>>
>> Looks like a ton of good effort by a lot of good people, but there's a
>> table claiming TwoCats generates only 1.26 bytes/cycle on a Haswell CPU
>> compared to some lame speed by their algorithm.  Huh?
>>
>> I just measured it on a similar CPU at 0.46 cycles/byte   There are
>> trade-offs,, and speed isn't everything.  It just happens to be fun :)
>>  Sorry, but the authors here made a mistake.
>>
>> I guess I can't get mad about other peoples mistakes.... that would be
>> seriously hypocritical.
>>
>> Bill
>>
>
>

Content of type "text/html" skipped

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ