After the sensorial exploration of Talvipuutarha, the initial sonic vision involved gamelan-type sounds and non-tempered scales. Intuitively, we felt that the glass and metal structure would resonate and shimmer with inharmonic, broad-spectrum, and complex sounds.
Based on this intuition, a large amount of initial sound material was then recorded with the percussionist Petteri Kippo at the studios of University of the Arts’ Sibelius Academy. The Sibelius Academy’s percussion collection was used, concentrating on metal, glass, and ceramic instruments. A total of three hours' of sound material was recorded. The above photos depict Petteri Kippo in the recording studio. Close mic’ing techniques were used in order to ‘zoom’ into the sound and retrieve as much detail as possible. The Talvipuutarha greenhouse acoustics are very reverberant and may cause a loss of detail in sound. It was thus very important for us to have high precision sound material to start with.
The resulting sound bank served as a basis for the composition, each composer employing his own set of techniques and software. A difference in approach was noted here: O. Lähdeoja used multitrack techniques to formulate a tape-music inspired approach to sound composition and spatialisation, whereas J. Moreno used spatial engines in order to move sound events within the space.
The final diffusion engine was made with two computers running max/MSP — one for each room with sound (Talvipuutarha’s third room was left without sound in order to preserve a social space for the visitors). In the Palm Room, the engine was programmed to create a generative mix between Lähdeoja’s and Moreno’s compositional materials, interweaving both composers into a continuous, ever-changing texture. In the Cactus Room, the approach was entirely generative, based on granular synthesis and partly controlled by external weather data.