In this paper, the authors present a method to analyze different implementations of stream-based applications on heterogeneous multiprocessor systems. They take both resource usage and performance constraints into account. For the first aspect they use an empirical cost model. For the second aspect they build a network of cycle-accurate processor simulators. The simulation and resource cost estimation have been integrated in an existing framework, allowing one to generate fast exploration simulations, cycle-accurate simulations and FPGA implementations from a single system level specification.