The project operates on a 8 computer client-server architecture to create the shared virtual environment, where each person has their own avatar based experience. Each of the 6 graphic clients independently generate their own particular environment state and view, with each client communicating environment event changes (such as avatar location, object triggering, etc.) to the other clients on the network and to the server. The server coordinates timed transformations such as scene changes and character construction.

The piece runs with custom software developed for the project, written with a game engine, NetImmerse from NDL, Inc. Face scan data is taken from each scan and translated into the appropriate file formats, assigned by the server to particular characters and distributed to each client computer. Each client computer uses Nvidia GeForce 4 cards for display. And the displays are created specifically for the physical installation, consisting of Hitachi LCOS projectors, projected sideways to create a 7’ x 5’ portrait orientation screen for each kiosk.

The project software and content developed over a year with two primary programmers, first Ryan McKinley, and then Craig Donner. To facilitate creative development, an application called the SceneEditor was created which allowed for quick changes in project content and in scripting interactions in the project. In particular, one area of particular interest is the way in which the project uses viewpoint choreography that is a further development of montage strategies that Sheldon Brown has been developing which combine cinematic and gaming vocabularies.

The modeling was done with 3D Studio Max, with Wes Middleton and Chris Berg assisting with the development of 3D models, images and video.

The physical installation was designed by Sheldon Brown and built by Sparks Environments. The displays and interfaces were designed and built by Sheldon Brown with assistance from David Poulin.