Power and energy consumption is a major issue and a design constraint in many types of systems. However, the knowledge of average power consumption is not enough, since many issues may arise due to dynamic power characteristics, such as IR drop, and side channel attacks in cryptographic circuits via DPA. Analog simulation (e.g. Spice), which has been used for decades as an accurate mean for estimating current and power, is no longer an option when it comes to medium to large circuits, due to the unacceptable simulation time. This paper presents an open framework for fast and accurate dynamic power estimation based on gate-level simulation, and standard tools and libraries. Our solution was thoroughly evaluated using standard benchmark and cryptographic circuits. Relevant speed and accuracy measures were compared to the ones obtained with analog simulation, such as speed-up, peak and average power, and cross correlation. A comparison with other published solutions was conducted, when possible. Our solution achieved a speed-up of three orders of magnitude in simulation time as compared to analog simulation, which is one order of magnitude better than other published solutions. The results, such as average and peak power and correlation, are in the same range as the published solutions. When applied to cryptographic circuits our solution has shown similar results to benchmark circuits indicating a feasible solution for DPA evaluation usage. Our solution is totally based on standard simulators and libraries. It is fully open and it has disclosed documentation and code, available for public use.