[Cryptech Tech] Seeking comments on a proposal for changes to the Cryptech RNG design.

[Rob Austein]

I'll leave analysis of the crypto impact per se to others who know
more, but given the amount of testing that the TRNG has already had
(some of which we may not be able to repeat because it was done by
third parties who may not volunteer to test a new version), there may
be some value in considering the minimal necessary change.  Which
would appear to be replacing SHA-512 with SHA-256 along with the
minimum set of changes necessary to support that, on the theory that
members of the SHA-2 family are mostly interchangeable.

I'm not opposed to the more ambitious options Joachim proposed, just
trying to find the right balance.  What I most want out of the
proposed revision is the ability to run the FPGA at 100MHz without
breaking the good TRNG we already have; if we can get an even better
TRNG and free up some resources, cool, but the issue that's really in
our faces at the moment is the clock speed restriction.