MathGroup Archive 2005

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

Search the Archive

Re: Integration under Mathematica 5.0

  • To: mathgroup at smc.vnet.net
  • Subject: [mg57439] Re: [mg57417] Integration under Mathematica 5.0
  • From: Daniel Lichtblau <danl at wolfram.com>
  • Date: Sat, 28 May 2005 05:39:02 -0400 (EDT)
  • References: <200505270856.EAA07672@smc.vnet.net>
  • Sender: owner-wri-mathgroup at wolfram.com

José Carlos Santos wrote:
> Hi all,
> 
> At another newsgroup, someone has transcribed this Mathematica
> session:
> 
> In[1]:=$Version
> 
> Out[1]=5.0 for Microsoft Windows (November 18, 2003)
> 
> In[2]:=Integrate[Sqrt[1+Sqrt[u]]/(E^(a*u)*(2*Sqrt[2]*Sqrt[u])),{u,0,1}]
> 
> Out[2]=(1/3)*(4 - Sqrt[2])
> 
> This makes no sense, because the integral does depend upon the parameter
> _a_ (just ask Mathematica to plot the graph). I do not have access to
> Mathematica 5.0, but under the version 4.0 I do not get that answer;
> what I get is the same integral written in a slightly different form.
> 
> Can someone reproduce that behaviour? And, if it is possible to
> reproduce it, can someone *explain* it?
> 
> BTW, the value (1/3)*(4 - Sqrt[2]) is the value of the integral when
> a = 0.
> 
> Best regards,
> 
> Jose Carlos Santos

I can explain it. It's a bug.

It will be fixed in a future release.

Daniel Lichtblau
Wolfram Research


  • Prev by Date: Complex Oddity
  • Next by Date: global fit
  • Previous by thread: Integration under Mathematica 5.0
  • Next by thread: Re: Integration under Mathematica 5.0