[115098] in cryptography@c2.net mail archive
Toshiba shows 2Mbps hardware RNG
daemon@ATHENA.MIT.EDU (David Wagner)
Thu Feb 14 17:44:39 2008
From: David Wagner <daw@cs.berkeley.edu>
To: cryptography@metzdowd.com
Date: Wed, 13 Feb 2008 13:49:39 -0800 (PST)
Crawford Nathan-HMGT87 writes:
>One of the problems with the Linux random number generator
>is that it happens to be quite slow, especially if you need a lot of
>data.
/dev/urandom is blindingly fast. For most applications, that's
all you need.
(Of course there are many Linux applications that use /dev/random
simply because they don't know any better, but that's a pretty weak
argument for a fast hardware RNG.)
A fast hardware RNG could be useful but I'm not convinced high
speed matters all that much for most applications. Grab 128 bits
for a hardware RNG, feed it through AES-CTR to generate an unending
stream of pseudorandom bits -- that's good enough for most applications.
(Yes, I know there are exceptions where pseudorandomness is not
enough. But even the exceptions rarely need true random numbers at
a rate of several Mbps.)
---------------------------------------------------------------------
The Cryptography Mailing List
Unsubscribe by sending "unsubscribe cryptography" to majordomo@metzdowd.com
