It would be pointless for me to share my measurements because you wouldn't (and shouldn't) trust them. I based my "ultrahard" assessment on iterative median logfreedom and kernel skew measurements with Dyspoissometer, which turned out to be of comparable hardness to the bimobius function. (The bimobius function is asymptotically indistinct from binary noise, if the Riemann Hypothesis is true.) You can see this for yourself in the Enranda demo. But knock yourself out with your own statistical test suite. If you find anything suspicious, which would most likely be caused by a bug as opposed to a design flaw, email me.
For your convenience, Enranda comes with otpenranda (One-Time-Pad Generator with Enranda), a command line utility which allows you to create power-of-2-sized files comprised of unmodified Enranda output. You can then analyze the resulting file with dyspoissofile or an application of your own choosing.
No comments:
Post a Comment