Rainfall is strongly affected by topography. Orographic rainfall, which is forced by topography, is water resources for human societies and becomes the cause of natural disasters such as flooding. It is, therefore, important to estimate orographic rainfall quantitatively and to understand its characteristics such as intensity and frequency. To understand orographic rainfall and to project future changes of its amounts and its characteristics are quite important for projections of the regional variations of water resources due to climate change in future. A gauge-based analysis of daily precipitation over East Asia by Xie et al. (2007) is characterized by including rainfall due to topography. We use this dataset to examine precipitation simulated by a global 20km-mesh atmospheric GCM from the point of view of the grid-scale relationship between precipitation and topography. In comparison with experiments in the AGCM with different (55km, 110km, 190km, 280km) horizontal resolutions, it is revealed that the 20km-mesh AGCM shows the superiority in simulating orographic rainfall not only its location but also its amount. It is also shown that the vertical profile of mean rainfall intensity and rainfall amounts are more realistic as the horizontal resolution of the AGCM is higher. We also examine the relationship between precipitation and slope of topography in terms of its magnitude and its direction.