Re: Scilab is now GPL v2 compatible - as the site claims

[bharat pathak]

Hello Sergei,

    In octave and matlab there is a function called remez
    which returns the transfer function given order, fvec,
    avec, wghts parameter.

    So far I have done enough signal processing using Octave
    and never felt limited. When certain functions were not
    present I build my own library of 50 functions using the
    base functions.

    The examples mentioned in the books

    "Digital Signal Processing Using Matlab : by vinay k ingle
    john proakis" work well using octave too. I have personally
    tried all of them myself. And they work seamlessly well.
    I think there are more than 100 examples listed in the book.

    Regarding Scilab I must say the language takes time to learn.
    I had initial hiccups. Example, in octave there is command
    called filter.m which does time domain filtering just in 1 line.

    Scilab has a function which is called rtitr and to emulate the
    same behaviour in Scilab it took me 30 min to figure out
    and took 10 extra lines to write the code.  One thing I like
    about Scilab is Scicos (which is free equivalent of simulink),
    though with base functionality, hope it improves over time.
    Scilab should focus more on having more blocksets.

   About me : My name is Bharat Pathak and have worked
   extensively in DSP and image/video processing for over
   12 years and run a company named "Arithos Designs"
   www.Arithos.com. Primarily into DSP Design Consultancy
   and Corporate Trainings.

Regards
Bharat Pathak

P.S: This mail is not to hurt any of sentiments of Scilab Users.
I too have Scilab installed on my desktop and use it now and
then, just that my comfort level working with Octave is much
higher.








--------------------------------------------------
From: "Sergei Steshenko" <address@hidden>
Sent: Sunday, September 21, 2008 12:27 PM
To: "Søren Hauberg" <address@hidden>
Cc: <address@hidden>
Subject: Re: Scilab is now GPL v2 compatible - as the site claims

> --- On Sat, 9/20/08, Søren Hauberg <address@hidden> wrote:
>
> > From: Søren Hauberg <address@hidden>
> > Subject: Re: Scilab is now GPL v2 compatible - as the site claims
> > To: address@hidden
> > Cc: address@hidden
> > Date: Saturday, September 20, 2008, 11:35 PM
> > lør, 20 09 2008 kl. 23:28 -0700, skrev Sergei Steshenko:
> > > Scilab has a really nice function for filter
> > synthesis: frep2tf. I think
> > > it is better than its octave-forge counterpart. Better
> > in the sense of
> > > accuracy.
> >
> > Patches are more than welcome :-)
> >
> > Søren
>
> ???
>
> If anybody is interested, he/she first has to verify my claim -
> my understanding of the algorithm is superficial, but, still, I think
> frep2tf has a functional piece its octave-forge counterpart doesn't have,
> and that is why the function is more accurate.
>
> If my claim is correct, then octave-forge maintainers willhave to decide
> whether to add a new function based on fre2tf or to change the existing 
> ne.
>
> ...
>
> Anyway, frep2tf has a subtle, though not serious, flaw.
>
> IIRC, it uses sum of abs deviations rather than sum of squares of abs
> deviations as its error metric. I mean, strictly saying, it's not least
> squares approximation as it claims. It was one of the things I changed in
> my local copy.
>
> Regards,
>  Sergei.
>
>
>
>
> _______________________________________________
> Help-octave mailing list
> address@hidden
> https://www-old.cae.wisc.edu/mailman/listinfo/help-octave