[sdiy] Frequency shifters, again

[Czech Martin]

> Excellent. If you can add some nice user interface (lots of pots),
> you will have a really good FS!

This was NOT intended. This suite of little tools just gives me the possibility
to examine new (old) ideas via building a signal chain in a modular
way. Without any headroom headache, because all numbers are in %e format.
This way I could concentrate on signal processing instead on
bit fumbling. Of course: I/O of such programms is slow.

If I find something usefull, I can create an optimised code,
like the "ideal" multiplier without
alias or the fast convolution.

All in shell mode , no fancy GUI. Wouldn't be portable to other OS.

btw.: the creation and testing of these basic signal processing
tools was more enlightening than 2 years of extra university education,
I bet. I regret that I could not afford a reasonable computer
when I was a student....

> 
> 
> > -mixing the original signal with the shiftet signal will not give
> >  any interesting results. Only some amplitude modulation 
> could be observed,
> >  no interesting phasing pattern.
> 
> This is interesting.

yes, I hoped so.


> 
> 
> > The absence of interesting phasing patterns makes me think
> > about the role of the "dome filter" in analog FS.
> > Does the pase shift which is created in that filter add
> > some effects to mixing sounds? 
> > If this is so (I think yes), a filter method SSB modulator
> > would perhaps be not so desireable for music purposes.
> 
> I think yes, it's the dome filter. If you look at the "classic"
> FS design (with dome filter), you can forget about the FS 
> background completely: Just think of the two multipliers
> and the quadrature LFO as a means to slowly crossfade
> (or is it "vector-crossfade" ?) between two different phase
> shifter channels. For a fixed position (sin of LFO = 0,
> cos of LFO = 1), you just get a simple, static, phase shifted
> signal. Same for the other extreme (sin = 1, cos = 0). 
> Of course it is important to have the 90 degree phase
> difference between the two all pass filters, in order to get
> a true all pass filter response for intermediate values
> (sin != 0, cos != 0) as well. But in the end it's the 
> all pass filters that create the notches, when mixed with 
> the straight input signal - not the modulation process.
> The latter is just needed for the "right" interploation.
> 

I think you are right.
Let's see what the additional all pass will do.

m.c.