Hi Dario,
can't you just start with the proper fdn examples from reverbs.lib? And later
implement some fancy feedback matrix stuff? There's hadamard(n) i.e. ready. A
custom matrix would look something like (so you can inspect various signals by
setting some coeffs to 0):
matrix(n) = bus(n) <: bus(n*n) : par(i,n,h(i)) : bus(n) with {
h(i) = par(j,n,(*(COEFF(i,j)))):>_ ;
COEFF(i,j) = 0; //n*n coeffs here
//... COEFF(0,0) = 0.1; COEFF(0,1) = 0.2; etc.
};
--
Urban Schlemmer
- Diplomtonmeister -\\
- Formation supérieure aux métiers du son (FSMS) -\\
http://chiselapp.com/user/jcage/repository/rdk/doc/www/www/revdev.html
> Gesendet: Sonntag, 03. Februar 2019 um 17:32 Uhr
> Von: "Julius Smith" <j...@ccrma.stanford.edu>
> An: "Dario Sanfilippo" <sanfilippo.da...@gmail.com>
> Cc: ur...@gmx.de, "Faust users" <faudiostream-users@lists.sourceforge.net>
> Betreff: Re: [Faudiostream-users] Routing signals to output
>
> Hi Dario,
>
> How about just letting everything go to the output? E.g.,
>
> //output(n) = si.bus(order*2) :> si.bus(channels);
> output(n) = si.bus(order*2);
>
> I often route whatever I want to see to the output so I can take a
> look in Octave.
>
> Another good practice is to define several intermediate functions, and
> set process to them individually for testing purposes. This also
> makes the block diagram more hierarchical.
>
> - Julius
>
> On Sun, Feb 3, 2019 at 10:12 AM Dario Sanfilippo
> <sanfilippo.da...@gmail.com> wrote:
> >
> > Hi, Urban.
> >
> > On Fri, 1 Feb 2019 at 16:48, <ur...@gmx.de> wrote:
> >>
> >> As I understand, routing is the problem here. It can be done; but you have
> >> to rewrite the feedback loop - which scrambles your code. I have a nested
> >> allpass somewhere with an output tap from the inner loop. Is that what you
> >> mean?
> >
> >
> > Your allpass example could be helpful, are you willing to share it?
> >
> > This is my situation at the moment:
> >
> > order = 4;
> > channels = 2;
> >
> > // SECTIONS
> >
> > input = si.bus(order);
> > matrix = si.bus(order);
> > delay(x) = si.bus(order);
> > nltf = si.bus(order);
> > fb = si.bus(order*2) :> si.bus(order);
> > output(n) = si.bus(order*2) :> si.bus(channels);
> >
> > // MAIN
> >
> > process(x) = (input : matrix : delay(x) <: si.bus(order*2)) ~
> > ((si.bus(order) :> _/(order) <: si.bus(order)) , nltf :
> > ro.interleave(order, 2) : fb) : output(channels);
> >
> > I would need to inspect the signals coming out from, for example, the "fb"
> > section. I believe that there must be a generalised proceedure to send any
> > recursive signal to the output and rewrite the process accordingly but I
> > haven't found it yet. I'll find a way eventually.
> >
> > Thanks,
> > Dario
> >
> >
> >>
> >> I resorted to viewing the output of the nested structure und understand
> >> that. Kind a complex. I sometimes write test routines containing
> >> simplified versions of the nested structures. Pattern matching seems
> >> limited in this regard.
> >>
> >> --
> >> Urban Schlemmer
> >> - Diplomtonmeister -\\
> >> - Formation supérieure aux métiers du son (FSMS) -\\
> >> http://chiselapp.com/user/jcage/repository/rdk/doc/www/www/revdev.html
> >>
> >>
> >> Gesendet: Freitag, 01. Februar 2019 um 10:42 Uhr
> >> Von: "Dario Sanfilippo" <sanfilippo.da...@gmail.com>
> >> An: "Faust users" <faudiostream-users@lists.sourceforge.net>
> >> Betreff: [Faudiostream-users] Routing signals to output
> >> Hello.
> >>
> >> I searched the archives and there's an email from 24/07/2017 (Debugging
> >> "peripheral" values that are not the main signal path) but it didn't get
> >> an answer, at least not publicly. I'm having the same problem so I'm
> >> bringing it back.
> >>
> >> I have a network with several modules and nested feedback loops. I need to
> >> inspect some of the signals which are not part of the main output to make
> >> sure that their behaviour is correct.
> >>
> >> When those signals are within the feedback paths, it seems less
> >> straightforward to route them to the output so that they can be inspected
> >> with faust2plot. A send2output function would be great but I don't that
> >> there's something like that.
> >>
> >> Do you have some procedure to send any signals to the output without
> >> changing most parts of the process?
> >>
> >> Thanks,
> >> Dario
> >> _______________________________________________ Faudiostream-users mailing
> >> list Faudiostream-users@lists.sourceforge.net
> >> https://lists.sourceforge.net/lists/listinfo/faudiostream-users
> >
> > _______________________________________________
> > Faudiostream-users mailing list
> > Faudiostream-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/faudiostream-users
>
>
>
> --
>
> Julius O. Smith III <j...@ccrma.stanford.edu>
> Professor of Music and, by courtesy, Electrical Engineering
> CCRMA, Stanford University
> http://ccrma.stanford.edu/~jos/
>
_______________________________________________
Faudiostream-users mailing list
Faudiostream-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/faudiostream-users
