Accurate knowledge of the precipitation estimates with high quality and fine spatiotemporal resolutions is crucial to the precipitation science communities. Reanalysis and satellite-based precipitation products are two primary sources of precipitation estimates for various applications. In this study, three latest reanalysis and satellite-based precipitation products, namely ERA5 (released for public in 2018 and 2020), ERA5-Land (released for public in 2019) and IMERG-Final (released for public as V06B in 2019), are selected in order to figure out their error characteristics against rain gauge observations at multiple scales over Jiangxi province, south central China, in 2019. The main conclusions of this study include but not limited to: (1) considering the accumulated yearly precipitation amount, both reanalysis precipitation data and satellite-based precipitation products have similar spatial patterns and show overestimations; (2) except for the performance revealed by POD, IMERG-Final generally outperforms ERA5 and ERA5-Land at multiple temporal scales, especially in terms of CC, FAR and MFI; (3) ERA5 and ERA5-Land precipitation products have similar spatiotemporal error characteristics, and ERA5-Land, which has finer spatial resolution, performs better than ERA5; (4) in the respect of the capabilities of capturing precipitation events, the spatial characteristics of IMERG-Final are the closest to those of rain gauges, while ERA5 and ERA5-Land significantly overestimate the event duration and underestimate the mean event precipitation rate. These findings could help the state-of-the-art reanalysis and satellite-based precipitation products improve the data quality in the future generations.