I have used the original Diehard tests to assess the quality of an in-house handrolled PRNG in FORTRAN 77. While I was reading some of the original documentatio

I have used the original Diehard tests to assess the quality of an in-house handrolled PRNG in FORTRAN 77. While I was reading some of the original documentation[0] I stumbled upon this: The CDROM I will soon be distributing contains some 60 ten-meg binary files of random numbers for use in serious Monte Carlo calculations as well as for input to DIEHARD. Most of those files combine several of the best deterministic generators with white noise and black noise, (the latter from a CD of rap music). I am not sure if that was a joke or a real thing. My guess is a joke. [0] http://www.stat.fsu.edu/pub/diehard/cdrom/source/diehard.doc http://www.stat.fsu.edu/pub/diehard/cdrom/source/diehard.doc