Representation is a piece which explores the relationship between man and machine. The title comes from the classical utilization of the performing space and time, as well as a clear distinction between the performer and the computer as dramatis personae. All the electronic material is the result of the elaboration of sounds recorded in real time during the performance. The first buffer provides the material for the layers, the remaining 29 are the source for a complex granular synthesis. The recording is threshold-based so that the system does not require any human control. The synthesis algorithm used is 5 instances of Randy (see the section about software) controlled by several stochastic methods, at different time scales.

The piece is divided in three phases. In the first phase, all the 30 buffers are filled; after the 30th the computer gradually moves to the second phase. This part does not involve any action by the performer except for noticing when the computer gets to the third and last phase. This transition can be heard when all the material moves towards high static layers of sound. Then, the microphone on the tam-tam is ready to send the signal to the system: the amplitude follower affects almost all the parameters of the system. When the durations of the piece are not related to the action of the performer, they oscillate between considerable margins of randomness, including the transitions from one phase to the other.

Performances differ considerably from one another. This is due both to the strong influence of the buffers recorded during the first phase and the randomness present in all stages of the piece, except for the sequence of the three phases that determines the macro structure of the composition.