[Cryptech Tech] Statistical Checks on the Random Numbers

[Russ Housley]

Joachim:

> Thanks Russ for taking the time to do the testing.
> Which implementation of FIPS 140 tests are you using?

The only difference in the various versions of FIPS 140 tests are the values in the Runs Test.  I do not recall if the code has the earlier ones or the later ones.  Probably the earlier ones.  That is easy to change if desired.

I'm not surprised that diehard is showing similar results.

Mixing seems to be very valuable.

Russ