About declicking. I mostly use software instruments, not sampled sound, but
the same ideas should work.

Every instrument I make uses a damping envelope to suppress clicks at the
beginning and ending of notes:

; Define an attack, sustain, and release period (values are variable and
depend on instrument):
iattack = .008
isustain = p3
irelease = .05
; Make p3 the sum of attack, sustain, and release,
; to make sure any very short notes are lengthened so their envelopes aren't
cut off with resulting clicks.
p3 = iattack + isustain + irelease
; Define a damping envelope (in addition to any other envelopes used):
kdamping linseg 0.0, iattack, 1.0, isustain, 1.0, irelease, 0.0
; Apply the damping envelope after all other processing.
aoutput = kdamping * asignal
out aoutput

Sometimes I make the damping envelope an arate variable, but usually I
simply set ksmps = 1.

I have had very few problems with clicks since I started doing this. Usually
if I do have clicks, they come from instruments whose internals generate
noise, use delay lines, etc., and I can apply another damping envelope to
the problematic part of the signal path. I still seem to get sharp enough
attacks, but sometimes I have to experiment to get the proper balance
between a sharp attack and no clicking.




-----Original Message-----
From: jim altieri 
To: csound@maths.ex.ac.uk 
Date: Friday, April 02, 1999 12:37 AM
Subject: declicking


>hi everybody,
>  i've heard a bunch of you mentioning at one time or another about
>declicking a csound piece.  what are some techniques to do so?  it's really
>become an issue with the piece on which i'm currently working.  if this is
>too elementary of a question, feel free to respond privately, or with a url
>for a tutorial that answers this question.  thanx.
>-jim altieri
>
>
>__________________________________________________________
>http://mp3.com/tweeg/
>http://www.oberlin.edu/~jaltieri
>jim.altieri@oberlin.edu
>ocmr 76
>oberlin, oh 44074
>(440) 775-6243
>