The stochastic parallel gradient descent algorithm (SPGD) is a promising control algorithm for adaptive-optics (AO) system, which is independent of wave-front sensor and can optimize the system performance directly. Based on SPGD, an adaptive-optics test-bed was built with a 32-element deformable mirror and a CCD. The principle of SPGD control algorithm was demonstrated through examining the effects of gain and perturbation amplitude on correction capability and convergence rate. Experimental results show SPGD can correct static distorted wave-front successfully when the gain and the perturbation amplitude were appropriate for distorted wave-front. The main effects on correction speed were discussed through analyzing experimental results.