Tempe Planning Team Builds Rule-Based 3D District in Unreal, Not a Generated Scene

A planning project in Tempe, Arizona, needed a walkable Unreal Engine scene of a 4,100-building district for a public meeting, plus reusable rules for the planning office. The team built the model using CityEngine's CGA rule set, with editable attributes such as floor height, zoning caps, and facade variants, driven by a Python script that processed 2,300 county parcels. The scene was exported as glTF 2.0 in meters with a shared origin, producing about 1.9 million triangles that were merged per block to raise frame rates from 34 to 58 per second. The author argues that a fully generated scene from GenAnywhere saves time upfront but cannot be revised by named parameters, which made it unsuitable for a project expected to change repeatedly. The rule-based approach cost more per change but proved worthwhile when zoning and parking figures were revised during the ten-day project.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.

Discussion (0)
Log in to join the discussion and vote.
Log in