[13238] in cryptography@c2.net mail archive
Re: CipherActive?
daemon@ATHENA.MIT.EDU (Joseph Ashwood)
Sun May 11 13:08:03 2003
X-Original-To: cryptography@metzdowd.com
X-Original-To: cryptography@metzdowd.com
From: "Joseph Ashwood" <ashwood@msn.com>
To: <cryptography@metzdowd.com>
Date: Sat, 10 May 2003 19:52:31 -0700
----- Original Message -----
From: "Rich Salz" <rsalz@datapower.com>
Subject: CipherActive?
> Anyone heard of these guys? An Isreali technology firm that claims to
> have a new patent-pending process for modexp that's 3-6 times faster
> depending on playtform.
>
> URL is at http://www.cipheractive.com/technology/technology.htm
It appears that either their sampling methods aren't at all accurate, or
that they are unclear. In both figure 2 and figure 3 the measurements are
performed on percentages and 1000'of bytes. Looking closely at these
samplings you'll note that there is only 1 shared sampling point, 32 1000's
of bytes, looking closely at the two graphs it is fairly easily noted that
"standard" RSA modexp uses apparently a little less than 50% of the
processor, using CipherActive's modexp method results in taking a little
less than 50% of the processor. In both cases it is easier to read the
location of the top of the bulk encryption portion, and in both cases it
lies slightly above the 60% marker. In fact the measurements are so close
that if you move them both into a picture editer, and stretch them to the
same distance between the markers (figure 2 is actually smaller), you will
see that the bars line up on a pixel by pixel basis.
In short unless their measurement methods have varied substantially between
the two measurements, the two methods are identical.
Joe
Trust Laboratories
Changing Software Development
http://www.trustlaboratories.com
---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo@metzdowd.com
