[sdiy] transducers

[Metrophage]

I know that Karplus-Strong algorithms are a bit limited, they sound
very much like that pipe delay, and perhaps the one modelled in
"Absynth". The reason why I mentioned it is that I think modal
synthesis is on the complex side to implement, so I decided to mention
something simpler as well. The reason why I mentioned modal synthesis
is because jhno said he'd like to know of any kind of digital reverb
which could actually describe three axes of movement in a spring, I
think this would be a starting point for making an accurate digital
>spring< reverb. I am not going to do it, there are other systems I'd
model to play with, and perhaps I will eventually get to that. Just
like jhno prefers using real springs, it is still probably the most
economical and best-sounding way to get this effect.

Anyway, a "model" might only be an "analogy" which is described with
just enough detail to be convincing! >;]P

C "Enough of these ivory-tower concepts, Caliban. Let us venture forth
into the marshland with the 707 and DrumFire!" J

--- CCartCat at aol.com wrote: 
> In a message dated 11/9/04 12:26:30 PM, richard at skydancer.com writes:

> > Karplus Strong isn't so much a physical model as a physical
> analogy. It's
> > hugely simplified compared to what happens in a real string, which
> is part
> > of a real set of strings, somewhere near a real soundboard, in the
> real 
> > world.
> > 
> > It's also not a good reverb model.
> > 
> > And - worse still - it's a DSP technique. ;)
> > 
> > If you want a spring reverb, the simplest way to make one is buy a
> long
> > piece of drain pipe, run a long spring along the middle, fit a
> cheap
> > speaker at one end and a mike at the other. It'll sound - unusual -
> but
> > most spring reverbs sound less than stellar, so at least you'll
> have
> > something original.
> > 
> > You can also buy a reverb tank - most guitar amp makers stock them
> as
> > spares for their own units - and build a simple driver circuit
> around it.
> > 
> > If you want a *good* reverb, that's a whole other issue. Some
> familiarity
> > with convolution reverbs probably wouldn't hurt at that point.
> > 
> > Or you can go find a large space and tile it. That will work too.
> > 
> > Richard
> > 
> 
> Ignorantly piggybacking on the above:
> 
> Yeah, for a digital version of a spring reverb, how about using the
> sound of 
> an impulse thru a spring reverb in a convolution reverb?   Do
> different spring 
> reverbs.   Get really weird with other springs/strings (as previously
> 
> mentioned)/resonant objects.
> 
> No doubt done to death already in some major music lab, but why
> should 
> whitecoats have all the fun?
> 
> Okay, back to lurking,
> Kevin Seward
> 



		
__________________________________ 
Do you Yahoo!? 
Check out the new Yahoo! Front Page. 
www.yahoo.com