MathGroup Archive 2010

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

Search the Archive

Re: (-1)^(1/2.) on Mathematica!

  • To: mathgroup at smc.vnet.net
  • Subject: [mg107531] Re: [mg107518] (-1)^(1/2.) on Mathematica!
  • From: Murray Eisenberg <murray at math.umass.edu>
  • Date: Tue, 16 Feb 2010 03:52:16 -0500 (EST)
  • Organization: Mathematics & Statistics, Univ. of Mass./Amherst
  • References: <201002151048.FAA28364@smc.vnet.net>
  • Reply-to: murray at math.umass.edu

Nothing wrong there.  Mathematica, like any computatational engine, is 
like a sewer: what you get out of it depends on what you put into it.

The result 6.12323*10^(-17) + 1. i for (-1)^(1/2.) does not surprise me 
in the least. It just seems to be numerical garbage due to the 
limitations of finite precision. If you Chop the result, you'll get 
something closer to what you doubtless expected:

   (-1)^(1/2.) // Chop
   1. I

As an experiment, try, say:

   (-1)^(1/2.`20)


On 2/15/2010 5:48 AM, Legendre wrote:
> Hi,
>
> I tried the expression (-1)^(1/2.) on Mathematica 7.0, I got 6.12323*10^(-17) + 1. i. By using //N there is no problem. But using 2. it gives a wrong answer.
>
> Sinan
>

-- 
Murray Eisenberg                     murray at math.umass.edu
Mathematics & Statistics Dept.
Lederle Graduate Research Tower      phone 413 549-1020 (H)
University of Massachusetts                413 545-2859 (W)
710 North Pleasant Street            fax   413 545-1801
Amherst, MA 01003-9305


  • Prev by Date: variable graph size when including Legend and FrameTicks
  • Next by Date: Poor choice in PiecewiseExpand ?
  • Previous by thread: (-1)^(1/2.) on Mathematica!
  • Next by thread: Re: (-1)^(1/2.) on Mathematica!