In this talk I will review my work on a 3D Virtual Reality model of the Roman city of Pollentia (Alcúdia, Mallorca, Spain). The story of this city began with the Roman conquest in 123 BC, but its first structures date back to 70/60 B.C. It spread and developed from the 1^st century BC until the 3^rd AD, when a great fire destroyed part of the city. After that episode, the site passed through different phases, including a Late Antique occupation.

This Virtual Reality project is directly related to my PhD thesis that will be defended in 2012. It is focused on the architecture and urbanism of Pollentia in the period between the 1st cent. B.C. and the 3rd cent. A.D. In a first stage the state of the art has been studied. The research on the ancient excavations from the 20s and 30s of the 20th century has been particularly useful regarding the relocation of lost archaeological data, and for the topography of the city. The following step has been to document the buildings in detail, analyse their architecture and contextualise them within the Western Mediterranean. Hypothesis about their elevations, chronologies and urban organization raised from these analysis, and I proposed myself to build a 3D VR model of the city for a single phase, as a way to focus on architectural features from the elevations, to present the results of my research, and to visualize the volumetry of the buildings.

The VR model takes into account the buildings standing during the 1^st century AD in the forum of the city: the Capitolium, the insula of tabernae, the square and two small temples. They have been built with Autodesk AutoCAD and 3ds Max, in part during a research stage at the Computing Archaeology Research Group. In order to represent my hypotheses on the urbanism and topography of the city, and to simulate a urban background, a model with Procedural CityEngine of the rest of the city was created, and then exported to the model in 3ds Max.

The location of the city, in an isthmus between two large bays in the NE part of Mallorca, offers a great visibility of maritime traffic of the area, and it was one of the reasons for the foundation of Pollentia. I thought it would be interesting to represent this landscape in my model, and this has been achieved with two different processes. The first one was to create a terrain model of the surrounding geography. The second one was to use a spherical photo as a spherical environment, simulating both the remote landscape (horizon) and the sky at a time. The final result is a model where hypothesis on architecture, urbanism and topography are represented in a feasible geographical environment.

The further purpose of this model is to be a tool for other analysis as illumination and circulation patterns. It will be specially relevant to apply structural simulations to this model, which I hope to be one of the issues in the future.