This work proposes a modified differential evolution (MDE) based maximum power point tracker (MPPT) for photovoltaic (PV) system under partial shading condition. The proposed MDE does not involve any random numbers; hence, consistency of MPP tracking always prevails. Besides, it only contains one tuning parameter, i.e., mutation factor, which significantly simplifies the implementation strategy and therefore a low-cost micro-controller can be used for its realization. Despite the simpler MPPT structure, for each tested shaded curve, MDE always converges toward the global MPP within 12 perturbations. Performance wise, it outperforms another evolutionary algorithm, namely particle swarm optimization (PSO), which frequently traps at local MPP in shading conditions. The proposed MDE also works accurately for the measured data profile of a tropical country during 9.00 am to 5.00 pm, where it attains 99.5% average tracking efficiency.