Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!sun-barr!rutgers!cmcl2!esquire!rreid
From: rreid@DPW.COM (r l reid )
Newsgroups: comp.music
Subject: Re: Guitar Sounds (Guitarists read this...)
Message-ID: <3289@esquire.dpw.com>
Date: 10 Jun 91 13:03:07 GMT
References: <5081@orbit.cts.com> <1746@carto.mdavcr.UUCP>
Reply-To: rreid@esquire.UUCP (r l reid )
Organization: Playdoh's Republic
Lines: 42

In article <1746@carto.mdavcr.UUCP> sw@mdavcr.UUCP (Scott Wood) writes:
>A few issues ago, the Computer Music Journal had an article in which the
>Karplus-Strong algorithm was extended to incorporate distortion of the
>vibrating string (if memory serves me correctly). Has anyone heard the
>output of this algorithm? In fact, has anyone run the output of a
>Karplus-Strong algorithm through a guitar setup?
>-- 

I'm sure you are speaking of Charles Sullivan's extensions.  Fantastic
stuff, and you don't NEED to run it through a guitar setup, since effects
come with the software (I suspect most of us who are crazy enough
to go through the pain of software synthesis would consider it
to be a "unclean" act to hardware-process the signal anyway :-)

I've used portions of that code to implement my software kitharas,
and stringed just tonality diamonds.  Advantages over the hardware
approach?  Well, I can retune a kithara of 12 hexachords in milliseconds
instead of hours,  my glides are precise, I don't seed to keep
sanding the soundboard, and I never have to restring.
And in the case of the stringed just tonality diamond, I frankly
don't know how I could make it exist in phsyical space in
any playable way - too many strings occupyining the same space.

That's another advantage - my strings can be as long as I need them to be.

Disadvantages?  For me, the main one was the code was written assuming
you would be using octave.pitchclass notation - not the right thing
at all when you are working in ratios.  But that was an easy enough
hack to the code.  For others, the lack of fingerboard noise
is still bothersome - I happen to like it (my hardware kithara
can barely be heard through the noise of the movable bridge),
but Brad Garton has used Sullivan's stuff in the aresenal
available to his lisp-based Irish string band, with stunning
results - but I've heard him long for fingerboard noise.

K-S, as implemented in CSound (which was a vary good implementation
to my ears) sounds flat and lifeless in comparision
to these.  To me, Sullivan is a Computer Music Hero.

R Reid
rreid@dpw.com
rlr@woof.columbia.edu