There are now several ways of predicting the illuminance produced for different configurations of light pipe. This study is an overview of two such methods for two slightly different forms of light pipe diffuser. Both models are partly empirical in nature but have been formulated using significantly different techniques. Their outputs are also different in presentation, although both models deal with illuminance predictions (with one of the models offering luminous flux predictions also). The models have also used separately obtained measurements in their formulation and so can be seen as being two independent methods, using similar logic, to achieve similar goals. While the two sets of predictions cannot be compared to each other directly (due to the difference in diffuser in each case), the two models can be presented as an indication as to how modelling procedures of this type can be approached. (c) 2004 Elsevier Ltd. All rights reserved.