[Octave-bug-tracker] [bug #30114] perms generates different results to matlab

[Rik]

Follow-up Comment #7, bug #30114 (project octave):

I agree with David that reverse engineering an interface that may change
probably isn't worth your time, although you are welcome to do it.  The
factorial function grows so quickly that perms isn't all that useful unless N
is small.  As a quick test,

z = perms(1:10);
zs = zeros(size(z));
tic; zs = sortrows(zs); toc;

The time taken to sort 3.6 million rows was 2.94 seconds.  The storage
required for Z is already 290 MB.  You might be able to push N a little
higher, but you're going to run out of memory well before sortrows becomes a
significant time sink.

    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?30114>

_______________________________________________
  Message sent via/by Savannah
  http://savannah.gnu.org/