On Feb 10, 2008, at 4:02 AM, Peter Gutmann wrote:
The device generates random numbers at a data rate of 2.0 megabits
a second, according to Toshiba in a paper presented at the
International Solid-State Circuits Conference (ISSCC) here.
I've always wondered why RNG speed is such a big deal for anything
but a few
highly specialised applications. For security use you've got two
options:
Assuming that it is impossible to introduce a bias externally and the
randomness can be specifically cryptographically qualified - and such
can be cheaply explained to an auditor - I can see a fair bit of use
to reduce the 'cost' you spend on convincing that same auditor that
your poker, roulette, etc site is fair, that you are keying all your
RSA/DH/whatever exchanges off the right randomness, etc.
I've had cases where a simple nonce (which was just required to be
different each time, so a i++ would do, not even unpredictable) ended
up being changed into some sha1() of some i++ ^ RNG -- as that was the
quicker way to get something argued live. So beeing able to wave a
magic wand over a large part of your infrastructure may be just the
ticked.
Dw
---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to [EMAIL PROTECTED]
