Frequency issues exist in generation, transmission and load sectors. From the load point of view, frequency sensitive devices like engines or clocks require quasi constant frequency to work properly. Whereas in the transmission system, power losses like hysteresis and eddy currents in transformers are frequency dependent. At the generation side, it is essential to prevent rotor angle from exceeding maximal threshold value; thus avoiding generators becoming out-of-step. Many techniques and model predictive control were published to solve the load frequency control (LFC) problem. Moreover, current optimization methods and high performance computers allow engineers to optimize complicated linear and nonlinear problems within reasonable time. Among optimization techniques, genetic algorithm (GA) optimization is utilized by control designers. It is also supported by Optimtool -MATLAB toolbox. This paper presented a comparative study of conventional and sliding mode control (SMC) designs for LFC. Optimized conventional controllers (PI and PID) were applied into a three -area system. Generated responses from conventional controllers were compared to responses from systems equipped with decentralized SMC.