On 13/03/2019 18:31, Gilles Sadowski wrote:
Dieharder did eventually finish the first 8 runs. Perhaps the test took a long time to determine if it was a pass/fail since it was borderline. Anyway they all eventually failed that test. I tried the following: XorComposite: A composite of two rngs using the xor operation: return new IntProvider() { @Override public int next() { return rng1.nextInt() ^ rng2.nextInt(); } }; SerialComposite: A composite of two rngs using alternating output: return new IntProvider() { int flip; @Override public int next() { return ((flip++ & 1) == 0) ? rng1.nextInt() : rng2.nextInt(); } }; I ran: XorShiftXorComposite: XorComposite using XorShift1024Star + XorShift1024StarPhi with the same seed XorShiftSerialComposite: SerialComposite using XorShift1024Star + XorShift1024StarPhi with the same seed SplitXorComposite: XorComposite using XorShift1024Star + TwoCmres (this is a control) FAILURE counts for Dieharder: XorShiftXorComposite : 89, 105, 104, 104, 105, 106, 105, 104 XorShiftSerialComposite : 27, 23, 22 SplitXorComposite : 0, 0, 0
BigCrush has finally finished: XorShiftSerialComposite : 0, 0, 0 XorShiftXorComposite : 0, 3, 0SplitXorComposite : 0, 1, 0 --------------------------------------------------------------------- To unsubscribe, e-mail: dev-unsubscr...@commons.apache.org For additional commands, e-mail: dev-h...@commons.apache.org
