Re: [ROOT] Getting a function maximum position

From: Rene Brun (Rene.Brun@cern.ch)
Date: Tue Jan 28 2003 - 18:04:30 MET


TSpectrum is not the solution to this problem. TSpectrum was designed
to find peaks in distributions with many channels and sharp peaks like
the distributions in gamma ray spectroscopy. In this case, sigma=1
is a good input.

Rene Brun

Allister Levi Sanchez wrote:
> 
> Hi Nick,
> 
> Yeah I have tried TSpectrum.  But its constructor accepts only TH1, not
> TF1.  Oh, by the way, what's the rule of thumb for setting sigma in
> TSpectrum::Search()?
> 
> Cheers,
> Allister
> 
> Nick van Eijndhoven wrote:
> 
> >Hi Allister,
> >To my opinion, the TSpectrum class does what you want.
> >However, when I tried it some time ago, I got the opinion
> >that the found peak positions were sometimes somewhat
> >displaced w.r.t. the actual ones.
> >Perhaps this has now been fixed in some more recent Root versions.
> >
> >                                           Cheers,
> >                                            Nick.
> >
> >
> >



This archive was generated by hypermail 2b29 : Thu Jan 01 2004 - 17:50:08 MET