Recently working on a piece which uses data sets Bitsphere will introduce later this year the question how to map sound to a 2D representation rose again. The most common application of this mapping is the step sequencer where one dimension is applied to the time and the other to a specific sound.
To give an overview over the topic Bitsphere prepared an incomplete list of possible mappings in this post.
The basic question is how to represent a given mathematical two dimensional set to a sound by applying sound control to the values of the x and y axis. A graphical representation of a data st may look like this:
There are some obvious and less obvious technologies in the list following:
x -> Time – y -> Frequency
This might be the most common application as seen on score sheets: One dimension shows the time range counting forward the other indicates the pitch of the melody. This is what a sequencer program like the old tracker programs basically were designed for: Determine at what time which note has to be played.
x-> Time – y -> Instrument (Sound)
The correlation between time and instrument is rather simple. In the two dimensional representation one axis shows the elapsing time the other axis’ values are assigned to an instrument. Obviously to determine which instrument has to be played an integer value is required but a good program may use the values in between to trigger the volume of the two instruments at the limiting integers.
x-> Time – y-> 2. Harmonic
With this mapping the pitch of the tone evolving is not altered but its quality when manipulating the 2. harmonic. This raises various possibilities for example some more harmonics may be bound to one of the 2D axis. Or just the odds. Check back the resources on the net for harmonic additive synthesis.
x-> Time – y-> Filter
This is a common practice in live use where filters are altered over time manipulating the sound. Disk Jockeys do this by hand though this may be implemented in code too.
x-> Time – y-> Volume
Mapping sound to volume may seem a rather boring thing, it is indeed a very powerful tool. In classical music the effect is known as dynamics. Most of you will know the signs pp, p , mf, f, ff and their numerous variations. Its extrems are ‘mute’ – ‘on’ which is the binary of all sounds existence and becomes very dynamic when differently applied to multiple sounds at a time.
As stated above this list is incomplete. Bitsphere appreciates any comment on this post with other possible mappings. The real power of this mappings lies in the multidimensionality when more than one of them is used in the same piece. This might be the topic of a post to come.