[693] in cryptography@c2.net mail archive
Re: Random numbers from the '60's...
daemon@ATHENA.MIT.EDU (Marc Horowitz)
Mon May 5 22:34:59 1997
To: perry@piermont.com
Cc: Phil Karn <karn@qualcomm.com>, brettc@tritro.com.au,
stewarts@ix.netcom.com, cryptography@c2.net
From: Marc Horowitz <marc@cygnus.com>
Date: 05 May 1997 21:26:15 -0400
In-Reply-To: "Perry E. Metzger"'s message of Mon, 05 May 1997 19:37:31 -0400
"Perry E. Metzger" <perry@piermont.com> writes:
>> I'm not sure that I trust this, if only because much of the "noise"
>> might not be thermal noise but instead things like the board picking
>> up stuff from the rest of the machine, most of which is decidedly
>> non-random...
This is what a mixing function is for. See rfc1750. As long as some
of it is random, and you have a good upper bound on the proportion,
you can still use this as a source of entropy.
Marc
