[Maxima-bugs] [ maxima-Bugs-1399818 ] bad answer from simpsum?

[SourceForge.net <no...@so...>]

Bugs item #1399818, was opened at 2006-01-08 09:36
Message generated for change (Comment added) made by robert_dodier
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=104933&aid=1399818&group_id=4933

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
>Category: Lisp Core
Group: None
>Status: Closed
>Resolution: Fixed
Priority: 5
Submitted By: David Bremner (bremner)
Assigned to: Nobody/Anonymous (nobody)
Summary: bad answer from simpsum?

Initial Comment:
Maxima version: 5.9.2
Maxima build date: 17:19 11/8/2005
host type: i686-pc-linux-gnu
lisp-implementation-type: GNU Common Lisp (GCL)
lisp-implementation-version: GCL 2.6.7


The attached test case shows different results
depending on whether the simpsum flag is set.

The number got by brute summation is the same as
that got by maple.

here is a transcript

[ 12 quixote ~/research/path/calc ]  maxima
Maxima restarted.
(%i1) display2d:off;

(%o1) false
(%i2) load("test.max");

(%o2) ?test\.max
(%i3) val1;

(%o3) 'sum(-3*b*2^(b+1)-3*2^(b+1)+1536*b+1536,b,3,8)
(%i4) val2;

(%o4) 35424
(%i5) val3
;

(%o5) 56880
(%i6)                

----------------------------------------------------------------------

>Comment By: Robert Dodier (robert_dodier)
Date: 2006-01-14 01:12

Message:
Logged In: YES 
user_id=501686

Dunno what's going on here exactly, but these bugs are fixed
by recent (post-5.9.2) changes to sum and product functions,
which are now in cvs and will be included in the 5.9.3 release.

By the way subst ([l=10, j=3], expr); has the same effect as
nested subst's. (See also sublis.)

Here's what I get when I try this on a recent cvs version:

(%i1) display2d:false;

(%o1) false
(%i2)
expr:sum(b*(j-1)^(l-1)*j+(j-1)^(l-1)*j-b*(j-1)^(b+1)*j-(j-1)^(b+1)*j,b,3,l-2);

(%o2) 'sum(b*((j-1)^(l-1))*j+(j-1)^(l-1)*j-b*((j-1)^(b+1))*j
                            -(j-1)^(b+1)*j,b,3,l-2)
(%i3) val1:subst(l=10,subst(j=3,expr));

(%o3) 35424 /* <-- OK !! */
(%i4) simp:false; /* <-- TURN OFF SIMPLIFICATION SO WE CAN
EXTRACT THE SUMMAND */

(%o4) false
(%i5) val1:subst(l=10,subst(j=3,expr));

(%o5) 'sum(-1*((-1+3)^(b+1))*3+(-1)*b*((-1+3)^(b+1))*3
                              +(-1+3)^(-1+10)*3
                              +b*((-1+3)^(-1+10))*3,b,3,-2+10)
(%i6) summand:part(val1,1);

(%o6) -1*((-1+3)^(b+1))*3+(-1)*b*((-1+3)^(b+1))*3
                         +(-1+3)^(-1+10)*3+b*((-1+3)^(-1+10))*3
(%i7) simp:true;

(%o7) true
(%i8) val2:sum(ev(summand),b,3,8);

(%o8) 35424 /* <-- OK !! */
(%i9) val3:ev(val1,simpsum);

(%o9) 35424 /* <-- OK !! */


----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=104933&aid=1399818&group_id=4933