Sets of reinforced concrete beams are tested to evaluate analytical results of nonlinear finite element as well as to investigate sensitivity of FEM results to some important issues such as method of solution, mesh discretization, concrete constitutive model and also crack models. Since specimens' shear span to depth ratio a/d is rather low (1.5), they are classified in deep beam category and most probably their behaviors are dominated by shear resistant mechanism. It is however observed that almost all of the beams failed in shear mode in experiment. On the other hand analytical loaddeflection response was also dominated by brittle failure and sudden drop after the peak load without any trace of yielding in longitudinal reinforcements. The results however showed considerable sensitivity to several parameters addressed above and will need more concern to properly select those methods, models and parameters such as solution procedure, fracture energy and shear retention factor. Those parameters and methods are examined and evaluated by test results where in the most acceptable one is selected for investigation of this study.