Problematic expression for expand



-----maxima-bounces at math.utexas.edu wrote: -----

>I?don't?know?if?there?is?a?problem?in?expand?here?that?can?be?fixed
>without?rethinking?expand?in?a?major?way,

Expanding this expression requires about 6.4 million calls to simptimes and
about 3.3 million calls to simplus. (I neglected to set timer for
simpexpt.)
I made the mistake of printing the expanded form--apparently it's a mess.
I'm
not so sure that this shows a weakness or problem with expand--the expanded
form
is huge. Those GCDs ratsimp does help a great deal.

(%i3) timer(?simplus,?simptimes,?simpmexpt);
trace: ?simpmexpt has no functional properties.
(%o3)                        [simplus, simptimes]
(%i4) expand(f)$

(%i5) timer_info();

      [ function          time//call          calls     runtime
      gctime ]
      [                                                                    ]

      [ simptimes  6.216851747439563E-4 sec  6395586  3976.041 sec
      0    ]
(%o5)
      [                                                                    ]

      [  simplus   7.403910983151573E-4 sec  3337166  2470.808 sec
      0    ]
      [                                                                    ]

      [   total    6.623870617477975E-4 sec  9732752  6446.849 sec
      0    ]

Barton