[sdiy] Observations of synthesized stretched-harmonic waveformsand subjective comments on their musical qualities

[Mike Pepper]

> > cheater cheater schreef:
> >
> > (...)
> >
> >> A bare oscillator on its own doesn't sound *very* special at all when
> >> you're not animating any of the parameters,
> >
> > Maybe that special sound only comes into play with some distortion or
> > nonlinearity. The effect of distortion will be different as the shape of
> > the non-cyclic waveform changes, animating the sound. I think this is
why
> > a distortion effect is far less impressive on a guitar-like synthesizer
> > sound than on that of a real guitar.
> >
> Yep, that's why I hope to one day be able to do this in analog - it
> would definitely sound more interesting than digital.
>

As the AI researchers discovered in the late 90's, (at least that's when
someone I knew was working in the field), noise is your friend. It can help
to add animation, just as it can help stop AI's getting 'pattern locked'.
The problem with implementation is the additional computation.

If you are experimenting with this type of thing, and you have a least a
pentium 3, why not download Pd and do some experimenting with that. It's
free and can allow some pretty complex csound generators. If you are running
out of horsepower for Pd, then there is always CSound which will run as long
as it needs to.

http://puredata.info/

www.csounds.com

It would make more sense to do something where you get audio results fast
and interactively. Learning either of the above, if you haven't already got
them in your arsenal of tools, will give you useful music making material at
the end of the process, as well as possibly extra useful skills. John
Chownings work on FM was done in one of the generations of MUSICx, the
ancestor of CSound, and it seemed to work for him. You never know, you might
end up with a fee and your name in another Yamaha patent....

||\/||ike