The probability distribution results particularly from the interaction of the fields Population and Amount to select from:
In the field Population determines the value, entered by the composer, how many coincidentally (even (pseudo) probability distribution) value can be calculated from JCself.
From these values those are selected, which occurred so far most frequentcy. The number of those values is determined by the composer in the field 'Amount to select from'.
(Note: The value of Population must be greater than the value of Amount to select from.)
From this quantity one value will be taken on random, which is then written as value of the frequency in the score file (Csound). JCself stores this value and increases the value how often this frequency was used and by doing this increases the probability of its repetition.
In the field Frequency Raster determines the value of the number the raster of the table, by which the values for the calculation of the probability distribution becomes "summarized".
I.e. not the value of the frequency, but a coherent area of frequency values is differentiated from the program for the calculation of the probability distribution. The columns will be relatively increased.
Footprint indicates the frequency range around the determined value, whose weight value are increased likewise. How much is determined by the composer in the field Random Weight. The weighting decreases, the further the table code if off the of the current determined value.
The value in the field Iteration indicates the number of tones which will be generated. JCself indicates a counting in the top-left corner of the table-graph.
The value of the decrease relates exclusively to the column of the grid with the highest ever, which only selects if JCself this column, this amount is subtracted. The amount must be stated in the form of a negative number.
How often a tone was selected is stored in the frequency table. We call this number the weight.