However, because of its nature as an average, the image does not provide an indication of the changes in the sound over time. To fix this weakness, several images recorded milliseconds apart are superimposed on one another, creating a typographic model of the behavior of the sound over a staggered time period. Peacock explains the spectral image created by the Fast Fournier Transform by analyzing the image made by a clarinet playing middle-C concert. The hallmark clarinet characteristics (strong odd harmonics and high energy in the low frequencies) are evident in the spectrograph, as well as features of the tone's behavior as it is sustained through time. Having examined the features of a standard clarinet tone, Peacock moves on to the spectral image of the multiphonic tone....

Key differences in the images are clear. The three loudest pitches in the multiphonic image are only a step away from each other, deviating from the hallmark of the clarinet and resulting in a sound that is decidedly un-clarinet-like. In addition, the multiphonics show heightened irregularity in both partial amplitudes and pitch over time. While the standard tone showed changes in amplitude over time, those changes were uni-directional and preditable, whereas the multiphonic changes appeared at different times and unpredictably.
Peacock concludes by pointing out that the deepened understanding of the complexity of multiphonics that is made possible through this computer imaging can play a crucial role in controlling and manipulating these tones in future performances.