We empirically show how applying a pure greedy algorithm cyclically can recover compressively sampled sparse signals as well as other more computationally complex approaches, such as orthogonal greedy algorithms, iterative thresholding, and $\ell_1$-minimization.