[PD] help with pdf~ object (zexy)
IOhannes m zmölnig
zmoelnig at iem.at
Tue Oct 13 20:34:40 CEST 2015
On 10/13/2015 08:19 PM, Benoit Fortier wrote:
> Ok thanks IOhannes, it was very counterintuitive to me but I think I understand this now : in the help patch, the pdf~ object draws a probability density fonction of values between -1 and 1 in an array (the size of the array can be specified as an argument) by doing a density estimation of the flow of samples coming in the signal inlet of the object. Thats what I needed to understand, but if, for the record, anyone want to complete this explanation or correct me on something I get wrong, i’d be happy to read you.
the only thing that is not accurately correct, is that [pdf~] doesn't do
it just *counts* how often each (quantized) value appears and outputs
that measurement (normalized by the total number of samples encountered).
that's why the result gets better the more samples it observes.
-------------- next part --------------
A non-text attachment was scrubbed...
Size: 819 bytes
Desc: OpenPGP digital signature
More information about the Pd-list