MathGroup Archive 2007

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

Search the Archive

Re: Hole/Disk function

  • To: mathgroup at smc.vnet.net
  • Subject: [mg76091] Re: [mg76024] Hole/Disk function
  • From: Murray Eisenberg <murray at math.umass.edu>
  • Date: Wed, 16 May 2007 05:26:51 -0400 (EDT)
  • Organization: Mathematics & Statistics, Univ. of Mass./Amherst
  • References: <200705150845.EAA16651@smc.vnet.net>
  • Reply-to: murray at math.umass.edu

Let Mathematica do more of the work:  Use Boole, e.g.:

   holeyDisk[{x_,y_},holeSize_] :=Boole[x^2+y^2<=holeSize^2]

I made holeSize an additional parameter to holeyDisk rather than leave 
it global.  It's often safer and clearer to make all parameters 
explicit like that.

If you prefer, you could use the form holeyDisk[x,y,holeSize], with 
three arguments, instead of the two I used, where the first is a list of 
the coordinates.

You could, if you wished, make holeSize optional by specifying some 
default value, e.g.:

  holeyDisk[{x_, y_}, holeSize_: 1] := Boole[x^2 + y^2 <= holeSize^2]

Note my use of x^2+y^2,=holeSize^2 so as to avoid taking the square 
root.  Probably of very little consequence in practice, but my intuition 
is that calculating squares is more efficient than calculating square 
roots.  (I just tested doing tht with 100,000 random reals between 0 and 
1, and there a difference on the order of 0.02 seconds.)


Mathieu G wrote:
> Hello,
> Is there a better way to define a Disk/Hole function than that?:
> 
> DHole[x_, y_] := If[Sqrt[x^2 + y^2] <= HoleSize, 1, 0];
> 
> I would like to use it for a convolution using the definition as shown 
> in http://oldsite.vislab.usyd.edu.au/CP3/Four5/mod2-01.gif
> 
> Best regards,
> MG
> 

-- 
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: How to use the Simpson 1/3 rule to write this program
  • Next by Date: Re: Solve & RotationMatrix
  • Previous by thread: Hole/Disk function
  • Next by thread: Re: Hole/Disk function