[sdiy] physical modelling

[Stefano Costa]

Dave Krooshof wrote:

> So the difficultness correlates with how deep you want to get into it.
> I guess plain ordinairy FM is not that hard. Maybe the filters are.
> But to me, analogue synths do sound simple compared to physical,
> mechanical instruments.

No, it's a different complexity. In mechanical instruments you have to implememt
a precise model of the underlying mechanical thing, like a multiple resonant
oscillator and a friction model for a guitar. In the analogue electrical world,
there are an host of physical properties that complicate thing, ranging from
noise, impendance, capacitance that even with the same components may change
behaviour with a different current input. The analogue world must take carre of a
lot of different things, while let me say, the physical world has to implement
one thing very well.

> But I see you can get into this heavily.
> (BTW is this why the mooger fooger RTAS plugin in protools
> is so heavy on processor time?)

Don't know...

> But this is exactly the area I do find interesting.
> Still your modelling the mechanical world, altough the settings are odd.
> But incase of an analogue synth, what is the physical world your modelling?
> Are you actually about to model it to component level?

At a certain extent, this is already the case, i.e., the modelling is purely
theoretical, with ideal components, not real-life ones.

> As in 'change the instruments from silicium to germanium diodes while
> playing'?

Of course. Or playing an analogue synth with super-conductive components, with no
resistance and almost no noise.

> Or a model of our famous all tube BBD?

Everything....more or less.

Stefano