
The operating schedules of synchrotron sources encourage relatively rapid data collection. Consequently, very high count rates may be met in powder diffraction experiments and high-resolution single-crystal studies. A rigorous treatment of the losses due to pulse pile-up is given, as well as a discussion of possible losses due to windowing. The code for a simple correction algorithm is provided and used to illustrate the effect of losses on the intensities of two of the reflections in the powder spectrum of cimetidine.