[Octave-bug-tracker] [bug #58800] BIST for rng sometimes fails

[Hg200]

Follow-up Comment #12, bug #58800 (project octave):

With Markus' patch (i mean that one from comment #10), Rik's code does not
error (tested with N = 10000) and the "tst("rng")" passes wo errors.

The Matlab statement for "rng("shuffle")" is: "... initializes the generator
based on the current time ...". What I want to know is the output of Matlab on
Windows 7 for Rik's code in comparison to Octave. If anyone can do that, I
would be interested.

@Michael:  " ... to do some fixed operation on the state (in the simplest
case, just add 1)". The implementation in randmtzig.cc looks "memoryless" to
me. That is, "init_mersenne_twister()" returns the same state if the same
entropy is given at initialization. It is not intended to add information
about the past. I would not recommend touching the implementation unless the
original algorithm (i.e. written in papers) does provide such a feature.


    _______________________________________________________

Reply to this item at:

  <https://savannah.gnu.org/bugs/?58800>

_______________________________________________
  Message sent via Savannah
  https://savannah.gnu.org/