WAPPEN FIELD: 12-channel Spatialization vs. Stereo Reduction ============================================================ SPEAKER LAYOUT Wappen Field consists of a set of 12 suspended helmets arranged in the form of an isosceles trapezoid. The longest side of the trapezoid faces outward, usually towards the opening of the space in which the installation is presented. The helmets are equi-distant, at a radius of four feet. The entire installation, not including buffer space around the helmets, is approximately 20 feet wide by 10 feet deep. 9 10 11 5 6 7 8 0 1 2 3 4 IMMERSIVE EXPERIENCE The audio emitted by each helmet is loud enough to hear at a distance. It is also possible to place one's head inside a single helmet to focus on a single channel of audio. The diagram above labels each of the 12 helmets with a number (ranging from zero to eleven), each of which corresponds to an audio channel used by SuperCollider. STEREO REDUCTION A stereo reduction effectively eradicates the experience of being in a field of multi-channel spatialized voices. In addition, the frequency range of the speakers in our installation differ from most home stereo systems (ranging from 180 Hz to 6kHz) -- they capture the essential frequencies of the human voice and project more strongly and sharply than home entertainment stereo systems. Thus the mix for the installation is tuned specificially for that equipment and the consequent assumptions about the room in which they are hung. Nonetheless, these stereo reductions capture, albeit obliquely, the dynamic and intent of the Wappen Field installation. Each song of Wappen Field is made up of tracks derived from an improvised performance of seven vocalists, captured and isolated in the recording studio. The number of tracks used in each reconstituted song ranges from five to twenty-six. The installation frequently mixes tracks from multiple live performances, making possible combinations beyond what could be achieved in the studio. Even when tracks are mixed in combinations similar to the original ensemble performance, the voices are always remixed and spatialized in ways to enhance the relationship between voices by using proximity, distance and movement. Two examples of SuperCollider code in this directory demonstrate the difference between the stereo and 12-channel versions. If you read SuperCollider, or are just curious, see the files "supercollider--stereo.scd" and "supercollider-12channel.scd". These header files map speaker positions to descriptive variable names. The stereo channel reduction arbitrarily splits the field of 12 channels, sending the six channels on the left half of the trapezoid to stereo left, and all others to stereo right. There is no way to capture the distinctions between linear, circular and angular movement through the matrix with only a stereo pair. SPATIALIZATION and ALGORITHMIC PLAYBACK Voices are sometimes compartmentalized into sections of the speaker grid. Sometimes they emerge from or converge towards a single node of the grid. Other times, they swirl around the listener as a group, or drift aimlessly as if by their own intentions. For ensemble pieces that match the original, tracks are usually played from beginning to end, though they are moved about the space by algorithmic processes. In other cases, individual tracks are split up or re-constituted using mathematical functions to produce effects that would be unnatural, tiring or impossible for the human voice, but which capture the essence of Wappen Field. All processes that spatialize and choose from streams of sampled sound are effected by mathematical processes that have a clear structural outline, but also contain randomness that shapes the experience of the ensemble and sometimes gives it a spontaneous feel. Thus, no matter how long the Wappen Field installation runs, no song ever repeats exactly. In some cases the effect of two different performances can be quite drastic, but never so much to lose the essential message and impact of a given song.