[linux-audio-dev] FM, Phase Modulation and The Wiki

[Jens M Andreasen]

On Tue, 2005-06-28 at 21:13 +0100, Simon Jenkins wrote:
> On Tue, 2005-06-28 at 15:18 +0200, Jens M Andreasen wrote:
> > On Tue, 2005-06-28 at 14:14 +0200, Jens M Andreasen wrote:
> > 
> > > I am currently looking [in wikipedia] at the Casio section near the end (regarding
> > > phase distortion.) Wasn't this technique developed way back in the 60's
> > > by french academics?
> > 
> > Let me refrase that to:
> > 
> >  I am almost certain that this technique was developed at the "french
> > national institute of electronic music" (or some such?)
> > 
> > Question:
> >   What is the actual name of that institute, and can anybody cunning in
> > french help me search in their wayback archives?
> > 
> > /ja
> > 
> > 
> > 
> 
> Its unlikely that this was invented at IRCAM in the 1960's:
> 
> http://www.ircam.fr/62.html?&L=1
> 
> I rest my case.

I agree that there is nothing to see there ...

Am I the only one to remember that Yamaha sued Casio, and Casio got home
free because phase modulation was already published by IRCAM?

(honestly; That's a weird feeling ...)

> 
> Phase distortion gets its own short section in Curtis Roads' "the
> computer music tutorial" where it is again attributed to the Casio
> corporation. Phase /modulation/ OTOH, gets just a very brief mention as
> a variant of frequency modulation. 

Yamahas smash-hit 'DX7' is definately based on phase modulation. So if
'Curtus Road' haven't figured that one out yet, I just wonder ... How
much more from that source is solely based on Corporate Newsletters?


> It seems the difference is that PD works on individual cycles of the
> modulated waveform. The wiki author guesses that there may be an extra
> synching oscillator but in fact...


This thread is about correcting some of the wrongs in the Wiki! (you
damned dummy!)

> 
> "The scanning interval speeds up from 0 to [pi] and then slows down from
> [pi] to 2 [pi]. The overall frequency is constant, according to the
> pitch of the note, but the output waveform is no longer a sine".
> 
> ...somebody just did the math(s).
> 
Did not! Did perhaps? Did :)

Where is this from? The illustrations I have seen in Wiki, is (wrongly)
contradicting that frequency should be a constant (although, I guess
that most people around here would testify that a constant phase offset
equals a constant frequency! And I will personally haunt them if they
don't)



Well ... 

Never mind. The FM synthesis section looks somewhat less misleading now.
At least to these old eyes ;)
 

> Cheers
> 
> Simon

Cheers to you too!

/ja
> 
--