In papers, we can see people use NIST STS test suite to evaluate RNG.
However, I think that many people just evaluate short sequences, less samples, or only for several times (most are just a single time).
Perhaps, it is because people have tendency to report "SUCCESS", not facts.
Well …
more »