International audienceThe light pipes are innovative devices able to transport and distribute natural light without heat transfer in dark rooms. There are a lot of natural lighting applications able to predict the behaviour of light in a room through a traditional opening. Only few of them are able to model complex systems such as daylight guidance systems. Added to this, they seem to provide disparate and inconsistent results with respect to the actual performance of light pipes. The purpose of this publication is to present the approach undertaken and the results obtained to highlight the problem. To do this, a survey of the different programs has been carried out internationally to model tubular devices. Then, an inter-software comparative analysis was implemented for some of the applications listed. In order to assess the accuracy of numerical results, the results of an experiment -1:1 scale and in real weather conditions -were used as references to evaluate the chosen applications. We saw that the various selected programs tend to overestimate or underestimate the real phenomenon. The use of an experimental database permitted to put forward the most efficient applications. These results support the future goal to develop a new model. Future prospects of our study that can emerge are mainly based on the introduction of a new model for predicting the performance of light pipes and its integration in two software products developed within laboratory: CODYRUN (a multi-zone software integrating thermal building simulation, airflow transfers, lighting and pollutants) and HEMERA (a daylighting analysis software)