<p>Granular materials are ubiquitous in nature and commonly used in engineering. In geoscience, the movement of debris flow, the instability of slopes, and the slip of fault zones are all related to the movement of granular materials. The central issue in granular materials movement is its self-organizing intermittent behavior, which is called the avalanche process or crackling noise in physics and materials science. In this study, the influence of granular shape and hardness on avalanche characteristics has been investigated by continuous toroidal shear testing of granular materials and synchronous acoustic emission signal monitoring. Our results reveal that the energy distribution follows the form of <i>P</i>(<i>E</i>) ~ <i>E</i>^−<i>ε</i>e^{−(<i>E</i>/<i>Λ</i>)}, with the exponent <i>ε</i> = 1.53 and <i>Λ</i> = 10⁶ aJ. Omori aftershocks follow a power law with an exponent near unity for about four decades. The Omori sequence is identical for pre-shocks and aftershocks within experimental resolution, and this identity agrees with the previous epidemic-type aftershock sequence (ETAS) model prediction. The waiting times are power law distributed with 1 − <i>υ</i> = 0.9 and 2 + <i>ξ</i> = 2.1. The relative magnitudes between mainshock and the largest aftershock remain constant at 0.75. Changes of disk shapes (circles, octagons, and pentagons) show little influence on the avalanche characteristics. Variations in bead materials show the same exponent <i>ε</i> = 1.53 for polymethyl methacrylate and aluminum with excellent statistical relevance, and <i>ε</i> = 1.71 with a lower confidence level for soft nitrile rubber. Our experiments extract a fuller set of avalanche parameters with discs of different shapes and hardness. These results are particularly pertinent for research in geoscience, materials sciences, and physics where avalanches are commonly observed.</p>