[sdiy] Filter banks & variability & MaxMSP

[Henry Till]

Hello,

Yes, FFT in Max can sound good in some cases, but in
totally horrible in other cases.  Some of the filters
sound fine to my ears, but I have heard complaints
from others.    

I use Max/MSP at home and work, and I think it is
wonderful, though I never look to play the role of
something like my hardware synth.  It can be a
powerful way to prototype ideas and experiment with
ideas that would be far too complex to model quickly
in hardware.     And for soundfile playback and
processing, it can do great things.  With its Jitter
extensions added, it powerful video and 3d
processing/imaging, and much much more.  Very robust,
to use that oh so popular software expression.  And
for you colored spaghetti lovers, tt has a
patchcord-object visual interface, akin to modular
patching.  

I encourage the curious uninitiated mind to take a
look:

http://www.cycling74.com/

Another related piece of software is PD, or Pure Data,
which is the open source baby of Miller Puckette, who
was the original author of Max back when it was a
project at IRCAM.  It has a very very similar paradigm
to MaxMSP, in fact you could say they were siblings. 
Soft-hard bridgers should take note that this can be
compiled on a great number of platforms: windows,
unix, linux, os x, and there is even hacked version
for windows ce for running on handhelds.  So, with
cheap pc and some added dc-enabled adc/dac hardware
possibilities exist for an relatively inexpensive
computer-based addition to a modular system...  

That's at:
http://www.pure-data.org/ 

-Henry


--- jbv <jbv.silences at club-internet.fr> wrote:
> 
> 
> Tim,
> 
> >
> > I'm thinking this is one of those times DSP can
> come
> > to the rescue. There is a way to use FFTs to do
> > filtering. That way you just draw the filter shape
> you
> > want, and presto!
> >
> > I wonder if PCs are now fast enough to be able to
> do
> > DSP stuff in real time using a sound card for I/O?
> 
> I did extensive experiments with Max-DSP, mostly
> with 512 points FFT and granular amplitude control.
> Sounds great at times, but it depends on the kind of
> sound source. At other times it can sound really
> weird
> (you don't recognize the sound source, but the
> result
> isn't sonically interesting). I guess it comes from
> the fact
> that each FFT bin isn't really a BPF...
> 
> JB
> 


	
		
__________________________________
Do you Yahoo!?
Win a $20,000 Career Makeover at Yahoo! HotJobs  
http://hotjobs.sweepstakes.yahoo.com/careermakeover