expfit vs harminv - which is faster/smaller ?

[Sergei Steshenko]

--- On Tue, 12/16/08, Francesco Potortì <Potorti at isti.cnr.it> wrote:

> From: Francesco Potortì <Potorti at isti.cnr.it>
> Subject: Re: expfit vs harminv - which is faster/smaller ?
> To: sergstesh at yahoo.com
> Cc: help-octave at octave.org
> Date: Tuesday, December 16, 2008, 5:41 AM
> >there is 'expfit' in the forge package and there
> is 'harminv' which is a
> >standalone program (and a library if one needs).
> >
> >If I understand correctly, they both do the same thing.
> >
> >It looks like both are based on BLAS, but maybe
> somebody has experience with
> >both and knows which of them is faster and which of
> them has smaller memory
> >footprint.
> >
> >I think they are about equal, but just want to make
> sure.
> 
> By reading the docs, it seems to me that they use different
> methods, are
> thought for different problems and require different
> inputs: expfit
> requires you to give it a number of harmonics to find,
> while harminv
> wants a range where to search for them.
> 
> That said, I should make it clear that I have no experience
> whatsoever
> with those methods.
> 
> -- 
> Francesco Potortì (ricercatore)        Voice: +39 050 315
> 3058 (op.2111)
> ISTI - Area della ricerca CNR          Fax:   +39 050 315
> 2040
> via G. Moruzzi 1, I-56124 Pisa         Email:
> Potorti at isti.cnr.it
> (entrance 20, 1st floor, room C71)     Web:  
> http://fly.isti.cnr.it/

This is what I read in 'man harminv':

"
       -f nf  Specify a lower bound nf on the number of spectral basis functions (defaults to 100), setting a lower bound on the number of modes to search for.  This
              option is often a more convenient way to specify the number of basis functions than the -d option, above, which is why it is the default.
".

I am now running 'harminv' and am using this option - I believe it's also
number of hramonics.

So far it seems 'harminv' is less memory hungry and it hasn't yet crashed
on me.

Thanks,
  Sergei.