On Sep 9, 3:12 am, Peter Jeremy <peterjer...@acm.org> wrote: > On 2010-Sep-06 21:29:35 +0100, "Dr. David Kirkby" <david.kir...@onetel.net> > wrote: > > >I would be very weary of any random number generator that claims to be a good > >source of random numbers if the output differs by platform or compilation > >mode. > > It depends what you mean by "differs". For "real" random numbers, you > can only discuss different statistical properties of the sequences > because the actual sequencee of numbers cannot be predicted or repeated. > > For pseudo-random sequences, you can additionally talk about the > actual sequence of numbers returned by the rng. The documentation for > most rng's talks about sequences being repeatable by calling an > initialisation function with the same seed but generally does not > state the scope of the repeatability - so actual sequences generated > by a specific seed could potentially vary on different platforms. > > I would suggest that the bigger problem is that the GAP doctest relies > on a specific sequence of random numbers being used by GAP but does > not control or test the actual sequence of random numbers. > I don't quite see what you mean by the latter. GAP, without the patch I proposed in the spkg in this thread (trac #9867), uses a random seed in an endianness-dependent way. The said patch fixes this. Nobody seems to be willing to review the ticket, however, as if the problem got resolved in a way that escapes me.
Best, Dima > -- > Peter Jeremy > > application_pgp-signature_part > < 1KViewDownload -- To post to this group, send an email to sage-devel@googlegroups.com To unsubscribe from this group, send an email to sage-devel+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/sage-devel URL: http://www.sagemath.org
