This study concerns the numerical simulation of airflow and the prediction of comfort properties in a visualisation room of a research centre. Because simulation accuracy depends on the modelling level, special care has been given to fine details. Particular interest has been concentrated on the four-way ceiling air supply diffuser, on the furniture and on the thermal conditions given on computers and on mannequins.The geometric model was created using the parametric features of the pre-processor Gambit, in combination with elements created with the Rhinoceros NURBS modelling tool.Mass, momentum, energy, turbulence, species and radiation conservation equations were solved using Fluent commercial flow solver. Computed airflow and heat transfer parameters were used for numerical prediction of indoor air quality quantities based on ISO 7730. Basic parameters included air temperature, relative humidity and air velocity. These were combined with clothing insulation and metabolic activity measures to obtain standard IAQ indices such as the mean age of air, the predicted mean vote and the predicted percentage of dissatisfied room occupants. Post-processing was carried out with standard Fluent elements and also using Vu, a multi-platform tool which also allows data to be displayed in the Cave immersive environment. © 2004 Elsevier Ltd. All rights reserved.