Looper's Delight Archive Top (Search)
Date Index
Thread Index
Author Index
Looper's Delight Home
Mailing List Info

[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Date Index][Thread Index][Author Index]

Re: OT: True Sound Morphing



At 5:44 PM -0800 2/15/09, S V G wrote:
>      Doesn't the Hartmann Neuron synthesizer do nothing but true 
>sound morphing?  Or am I not understanding your definition...

Yes, the Hartmann will do morphing, but that is not all that it will 
do.  To me at least, "morphing" means gradually changing from one 
recognizable sound to another recognizable sound -- much the same way 
that those famous little PC widgets in the 90's "morphed" between two 
different photographs.

The Neuron is a Resynthesizer.  That means that it primarily takes a 
recorded sound, then converts it to merely a collection of sine 
waves.  This is not unlike taking an FFT snapshot of a sound, for 
instance.  However, an FFT is merely a static description of a sound 
(high-end synth/samplers like the Fairlight and Synclavier 
notwithstanding).  Once converted, these sine waves can be 
manipulated just like any other parameter on a synthesizer.

One of the things the Neuron can do is load two sounds (referred to 
in Hartmann-speak as Models) and interpolate between them -- thus 
accomplishing the "true sound morphing" we're speaking of.  However, 
you can load in a single Model -- essentially loading a null Model 
into the second slot -- and manipulate its sine wave makeup without 
having to bring another Model into the mix, if you don't want.

One of the biggest advantages of Resynthesis in general is that you 
don't have any of the standard munchkinization that you find with 
normal samplers.  You can change pitch and time of a sample 
independent of each other, simply by manipulating the tuning and 
duration of the sines, for instance.  You can also monkey with the 
harmonic makeup in ways you wouldn't find to be normally easy with a 
standard sample.  The Modelmaker software has you define which 
parameters of the wave you want to manipulate in realtime when you do 
the analysis to change the sample into a Model, so you have some 
options.

There's a lot you can do.  Pity the darn thing was discontinued early 
(due to bankruptcy).  It's a CPU hog, buggy as hell, and crashes or 
locks up pretty regularly -- which is why I'm normally loathe to 
recommend it.  But the results don't sound half bad, and it always 
manages to produce something interesting.

        --m.
-- 
_____
"we're no longer sure where home is; homesickness is our only guide"