Re: [ROOT] Strange behaviour of TF2->Integral()

From: Rene Brun (brun@pcbrun.cern.ch)
Date: Sat Sep 04 2004 - 08:19:21 MEST


Before you sent this mail to roottalk, I already replied to your private 
mail that this problem had been fixed in the CVS version.
The problem occuring for symmetric functions in x and y.

Rene Brun

On Fri, 3 Sep 
2004, Smbat Grigoryan wrote:

> Hi,
> 
> I have a problem with 2-dimensional integral.
> Here is a test macro showing the problem:
> 
>   Double_t func(Double_t *x, Double_t *par) {
>     Double_t r = 1 - x[0];
>     //    Double_t r = x[0];
> 
>     if (r < 0.5)
>       return par[0];
>     else
>       return 0;
>   }
> 
> void test() {
> 
>   TF2 *f2 = new TF2("f2",func,0,1,0,1,1);
>   f2->SetParameter(0, 1);
> 
>   cout << f2->Integral(0,1,0,1) <<endl;
> 
> }
> 
> After .x test.C  I get 0 instead of 0.5,
> but if I uncomment the 3rd line and comment the 2nd one
> I get 0.5 as it should be. (One dimensional integral is OK for 
> both options).
> I am using ROOT v. 3.10/02
> Any idea?
> 
> Thank you very much in advance,
> Smbat
> 



This archive was generated by hypermail 2b29 : Sun Jan 02 2005 - 05:50:09 MET