A new benchmark is being undertaken to assess the impact of wheel-rail contact modelling assumptions on the simulation of railway vehicle dynamics. The benchmark is split into two distinct simulation cases: the first, Case A, using a single wheelset to pinpoint the differences between the contact models and the second, Case B, using a simplified railway vehicle to assess the effect of the different contact models on the simulation of vehicle behaviour. After an open discussion of the Case A specification, the initial call for contributions was made in November 2006. The discussion of simulation Case B specifications was opened in April 2007 and to date is ongoing. This paper briefly introduces the new Manchester Contact Benchmark and presents some of the initial findings from simulation Case A.