Most works featuring the capacity of pedestrian infrastructures have focussed on unidirectional movement base cases, studied relatively low-density situations and instructed their participants to behave ‘normal’. However, during large crowd movements at train stations and events often far higher densities are encountered, the flow situations are more complex, and pedestrians do not always behave ‘normal’. The objective of this study is to determine the impact of heterogeneity in walking speed and the flow situation on the global and local dynamics of the crowd. A large pedestrian experiment was performed by Delft University of Technology, coined CrowdLimits, to derive the answer to this question. A preliminary analysis of the participants’ movements illustrates that especially the introduction of differences in walking speed and distribution between flows influence the shape fundamental diagram.