I am using the ramp function as shown in the code below.  When I run it on test data (e.g., a constant signal of amplitude 0.5) it glitches in the last few samples of the selection that I apply it to.  It smoothly ramps to the final volume as it should, then suddenly jumps to fairly random value.  This behavior is not 100% consistent: the existence and magnitude of the glitch seems to depend on the postion of the selection.

If I can send pictures or wave files to this mailing list I could show you more precisely what I mean.

;nyquist plug-in
;version 1
;type process
;name "Linear Fade In"
;action "Linear Fading In..."
(mult s (ramp 1))