Mathematica 9 is now available
Services & Resources / Wolfram Forums
-----
 /
MathGroup Archive
1997
*January
*February
*March
*April
*May
*June
*July
*August
*September
*October
*November
*Archive Index
*Ask about this page
*Print this page
*Give us feedback
*Sign up for the Wolfram Insider

MathGroup Archive 1997

[Date Index] [Thread Index] [Author Index]

Search the Archive

Re: Bug in SymbolicSum, or am I just stupid?

  • To: mathgroup at smc.vnet.net
  • Subject: [mg5624] Re: [mg5593] Bug in SymbolicSum, or am I just stupid?
  • From: Sherman Reed <Sherman.Reed at worldnet.att.net>
  • Date: Wed, 1 Jan 1997 21:04:54 -0500
  • Sender: owner-wri-mathgroup at wolfram.com

At 06:59 AM 12/27/96 +0000, you wrote:
>I'm wondering if I'm just doing something really dumb, or if there
>is some profound reason why Mathematica 2.2 seems to be finding different
>symbolic infinite sumations for the same expression:
>
>
>
>Mathematica 2.2 for SPARC
>Copyright 1988-94 Wolfram Research, Inc.
> -- Terminal graphics initialized -- 
>
>In[1]:= <<Algebra`SymbolicSum`
>
>In[2]:= ?SymbolicSum
>SymbolicSum[f, {i, imin, imax}] attempts to find the value of Sum[f, {i, imin,
>   imax} ] for symbolic imin,imax. SymbolicSum[f, {i, imax}] evaluates the sum
>   of f with i running from 1 to imax.
>
>In[2]:= Simplify[(2)( 1/(2^n )  - ((n+1)/2)(1/(2^n))]  (* expression 1 *)
>
>        3 - n
>Out[2]= -----
>           n
>        2 2
>
>In[3]:= Simplify[(2 - (n+1)/2) (1/(2^n))]  (* expression 2--should be same *)
>
>        3 - n
>Out[3]= -----
>           n
>        2 2
>
>In[4]:= SymbolicSum[(2)( 1/(2^n) ) - ((n+1)/2)(1/(2^n)),{n,1,Infinity}]  (*1*)
>
>        1
>Out[4]= -
>        2
>
>In[5]:= SymbolicSum[(2 - (n+1)/2) (1/(2^n)),{n,1,Infinity}]  (*2*)
>
>          3
>Out[5]= -(-)
>          4
>
>In[6]:= SymbolicSum[Out[3],{n,1,Infinity}]   (* simplified *)
>
>          3
>Out[6]= -(-)
>          4
>
>
>Thanks very much anyone.  Sorry if this is a FAQ or obnoxious newbie question.
>
>Sean Luke
>U Maryland at College Park, very late a night
>seanl at cs.umd.edu
>
>
Sean,

I was able to duplicate your results on 2.2.3 under Win95, which are wrong.

Using 3.0, where SymbolicSum has apparently been redone, is called simply
SUM, and is autoloaded, I got the correct answers.

looks like a bug in 2.2.2 and 2.2.3

sherman c. reed



  • Prev by Date: RE: HELP: Legendre transform
  • Next by Date: Help with 'Help' files
  • Previous by thread: RE: HELP: Legendre transform
  • Next by thread: Re: Bug in SymbolicSum, or am I just stupid?