Software size estimates provide a basis for software cost estimation during software development. Hence, it is important to measure the system size reliably as early as possible. Two of the best known specification level metrics, Albrecht's function points (A.J. Albrecht, 1979) and DeMarco's function bang metrics (T. DeMarco, 1982) are compared by a simulation study in which automatically generated randomized dataflow diagrams (DFDs) were used as a statistical sample to automatically count function points and function bang in a built CASE environment. These value counts were correlated statistically using correlation coefficients and regression analysis. The simulation study permits sufficient variation in the base material to cover most types of system specifications. Moreover, it allows sufficient sampling sizes to make statistical analysis of data. The obtained results show that in certain cases there is a relatively good statistical correlation between these metrics.>