Here is the link to my submission:
The piece is created by sonifying a time series of annual immigiration traffic into the US between 1820 and 1962 and another time series of internet traffic data in 11 European cities from 2005.
Here are the datafiles: Annual Immigration Data, Internet Traffic Data.
Here is the code: m220a-DataReader.ck, m220a-Duo.ck, m220a-Player.ck.
Questions:
(1) Why are the values for gain squared?
The squared mapping of the scaled values of data for loudness of the sound allows us to hear the intensity low where we have lower values of data and high when the data value is larger. Having a polinomial mapping instead of a linear one effects our audiotory perception such that the amplitude is more uniform and proportional.
(2) Why use MIDI keynums for the freq values?
Using the MIDI files provides a linear scaling where the frequencies changes logarithmically within an octave. Since the data is normalized, the range is from 0 to 1, resulting mapping it to a pitch range between G#5 and E7.
(3) What happens if the 100 ms update rate is increased or decreased significantly?
Because ChucK allows you to advance time, you can modify the frequency of updating one cycle of the code. Increasing 100 ms in this case makes the sound played slower since it advances time in a time slot longer than 100 ms. If it is decreased, the time is advanced in a shorter amount of time and we hear the sound is played faster.