Understanding the energy consumption of flash devices is important for two reasons. First, energy is emerging as a key metric for data management systems. It is thus important to understand how we can reason about the energy consumption of flash devices beyond their approximate aggregate consumption (low power consumption in idle mode, average Watt consumption from the data sheets). Second, when measured at a sufficiently fine granularity, the energy consumption of a given device might complement the performance characteristics derived from its response time profile. Indeed, background work which is not directly observable with a response time profile appears clearly when energy is used as a metric. In this paper, we discuss the results from the uFLIP benchmark applied to four different SSD devices using both response time and energy as metric.