- has anyone created one here?
What is the basic principle?
S
-
Grain Delay
boonier
-
Sorry to ask, but what does one sound like?
|] [] |.| ][|-| -- http://soundcloud.com/domxh
-
Not at all
Heres a few examples:
http://koen.smartelectronix.com/KTGranulator/
and i just found this (from the horses mouth i guess)
http://www.audiomulch.com/~rossb/rb-gst/BencinaAudioAnecdotes310801.pdfboonier
-
there is a pretty good description of what it does on the smart electronix page there:
It works like this: incoming (mono) sound is fed into a delay line from which small pieces of various durations and at different moments in the past are selected. Each of these pieces is then amplified, transposed and enveloped to form a "grain". Each grain is also randomly panned and the whole mix is sent out to a stereo output stream. Feedback of the grain output back into the delay line is also provided, and the delay line can also be frozen so that the grains are only taken from what is currently stored in the delay line.
one way to make one would be to use the basic engine of the pitch-shifter from the pd docs.　
you'd create a [delwrite~] and then read back at various speeds using [vd~]
-
(you'd have to create a 'voice' abstraction containing a [vd~] driven by [vline~], and signal operations for the amplitude, pan and envelope...and then invoke multiple instances of that abstraction in your patch to make a chorus of grain voices)
-
Hardoff - you were on the right track
I found this on the pure data org web site
A working example beats any manual i guess!
boonier
