Summary

• Physical models can be used as a compositional tool and they can render extraordinary results as is the case of the Maracas. There are some performance techniques which are impossible in real time performance like changing the amount of seed in one maraca or in both and also changing the size of the shell. This is possible in a computer generated gesture, adding expression to a musical gesture.

• The model of one instrument, the maracas in this case, or rather PhISEM, is capable of making a family of groups of sounds which might be related or not giving flexibility for composition as well as simplifying manipulation of the algorithm.

• There are many parameters to control in a physical model. While real time brings spontaneity and a great deal of responsiveness, actual control of the model is somewhat difficult. Rendering values for expression in an off-line program results in an efficient way for handling performance. In this way the physical model doesn't depend on a playing technique but rather on a characteristic sound an in many new ways on how the sound may be perceived.

• Some research about the real physics for the instrument is helpful to derive how musical expression can accompany the sound. In the computer synthesis of an instrument like the maracas a very diverse set of parameters can be discovered by experimenting with various parameters which can extend the usual instrumental technique to other domains like instrument morphing, dynamic range, delta tempos and spatialization.