[Maxima-bugs] [ maxima-Bugs-1007094 ] sum wrong with numer flag

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

Bugs item #1007094, was opened at 2004-08-11 02:17
Message generated for change (Settings changed) made by robert_dodier
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=104933&aid=1007094&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: Xmaxima
Group: None
>Status: Closed
Resolution: None
Priority: 5
Submitted By: Nobody/Anonymous (nobody)
Assigned to: Nobody/Anonymous (nobody)
Summary: sum wrong with numer flag

Initial Comment:
(%i64) build_info ();

Maxima version: 5.9.0.9beta1
Maxima build date: 19:5 6/25/2004
host type: i686-pc-mingw32
lisp-implementation-type: Kyoto Common Lisp
lisp-implementation-version: GCL 2.6.2

I am testing a sum. but the sum result is not the same 
when 'numer' is on/off, 289327664 compared to 
0.00825405701325.

(%i67) u;
(%o67) 'SUM((-1/(8*i+6)-1/(8*i+5)-2/(8*i+4)+4/
(8*i+1))/16^i,i,1,a)

(%i68) u, a=2, simpsum;
(%o68) 1618091/196035840
(%i69) %, numer;
(%o69) 0.00825405701325

(%i70) u, a=2, simpsum, numer;

RAT replaced 0.0625 by 1//16 = 0.0625
(%o70) 289327664
------------------ %o70 is wrong------

(%i71) u, a=2, simpsum, bfloat;
(%o71) 8.25405701324819B-3
----------------- bfloat is right --------




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

Comment By: Robert Dodier (robert_dodier)
Date: 2005-04-16 01:09

Message:
Logged In: YES 
user_id=501686

Another data point.  I get the same results (exactly) as
Barton Willis.

(%i6) build_info();

Maxima version: 5.9.1.1cvs
Maxima build date: 18:54 4/12/2005
host type: i686-redhat-linux-gnu
lisp-implementation-type: GNU Common Lisp (GCL)
lisp-implementation-version: GCL 2.6.6



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

Comment By: Barton Willis (willisbl)
Date: 2004-08-17 13:30

Message:
Logged In: YES 
user_id=895922

Maybe this is a bug with GCL 2.6.2; here is the calculation 
using beta2 and GCL 2.7.0

(%i3) display2d : false;
(%o3) FALSE
(%i4) u : 'SUM((-1/(8*i+6)-1/(8*i+5)-2/(8*i+4)+4/
(8*i+1))/16^i,i,1,a)$

(%i5) u, a=2, simpsum;

(%o5) 1618091/196035840
(%i6) %,numer;
(%o6) 0.00825405701325
(%i7)  u, a=2, simpsum, numer;


RAT replaced 0.0625 by 1//16 = 0.0625
(%o7) 0.00825405701325
(%i8) u, a=2, simpsum, bfloat;

(%o8) 8.25405701324819B-3
(%i9) build_info();

Maxima version: 5.9.0.9beta2
Maxima build date: 13:27 7/28/2004
host type: i686-pc-mingw32
lisp-implementation-type: Kyoto Common Lisp
lisp-implementation-version: GCL 2.7.0

Barton

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

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=104933&aid=1007094&group_id=4933