Microphysical properties play crucial roles in physical processes related to the development of precipitation. In this study, Global Precipitation Measurement (GPM) dual-frequency precipitation radar (DPR) data were processed to demonstrate the microphysical properties of different precipitation systems (PSs) that are objectively classified with the k-means clustering algorithm. Four types of regular/non-extreme PS (high-latitude shallow PS, subtropical shallow PS, moderate PS, deep PS) and four types of extreme PS (extreme deep PS, strong PS, extreme strong PS, and marine extreme PS) were recognized. These eight types of PS exhibit differences in spatial-temporal features and convection characteristics, such as storm height, rain intensity, and vertical structures. For example, with the highest radar echo top and the largest mean mass-weighted mean diameter (D<sub>m</sub>), the extreme strong PS mainly locate over tropical continent, while the high-latitude shallow PS have the least precipitation rate and mean normalized intercept parameter (N<sub>w</sub>) values. The relationships between convection features and microphysical properties also vary among the eight types of PSs. For extreme PS, maximum precipitation rate near the surface generally exceeds 100 mm h<sup>-1</sup> and balanced breakup and coalescence processes play a dominant role compared with non-extreme PS. In contrary, the coalescence processes dominate near the surface in two types of shallow PS. These results highlight the diversity of global precipitation microphysics and emphasize the necessity of global studies to increase the understanding of precipitation processes.