18 May 2017. Leonello Tarabella, collaborator of Signal and Images laboratory, helds a seminary at Universitat Pompeu Fabra in Barcelona titled "The experience on computer music of Pietro Grossi in Pisa, Italy, in the centenary of his birth".
I hereby report the approach of M° Pietro Grossi in using computers of the ‘70s and ‘80s for producing music.
Due to the low power of computers of those years (clock at 10 MHz and RAM memory of not more than 1Mbyte at the best) it was impossible to synthesize in real time audio signals of some relevance. The technique commonly used was the so called “offline” synthesis: this means that the computed samples of an audio signal were first cumulated sequentially on mass memory supports such as magnetic tapes or disks and then, in order to get sound, back-read and sent at the proper sampling rate to a Digital-to-Analog converter. As a result, depending on the complexity of the sound synthesis model and the number of voices of a composition, the process required even hours of computation for few minutes of music.
The philosophy of M.Grossi was that of “real-time”. For that, the special device “TAU2 audio terminal” was developed in Pisa at CNUCE and IEI, two institutes of C.N.R. (National Research Counsel) deeply active on researches on computer science.
Based on a hybrid electronic architecture (digital in control - analog in synthesis) TAU2 was able to play with polyphony of 12 voices under the control of a mainframe 370/68 IBM computer.
Once the TAU2 was assembled and put at work by the many engineers and technicians of both the CNUCE and IEI institutes, M.Grossi developed by himself the music programs writing down hundreds and hundreds code lines in FORTRAN language, in those years the most popular language used by the scientific community.