Irrigated rice in China accounts for nearly 30% of global rice production and about 7% of global nitrogen (N) consumption. The low agronomic N use efficiency (AE_N, kg grain yield increase per kg N applied) of this system has become a threat to the environment. The objective of this study was to determine the possibility to improve the AE_N of irrigated rice in China by comparing the farmers' N-fertilizer practices with other N management strategies such as real-time N management (RTNM) and fixed-time adjustable-dose N management (FTNM). Field experiments were conducted in farmers' fields in four major rice-growing provinces in China in 2001 and 2002. The same experiment was repeated at the International Rice Research Institute (IRRI) farm in the dry seasons of 2002 and 2003. Agronomic N use efficiency was determined by the "difference method" using an N-omission plot. Maximum yield was achieved mostly at 60-120 kg N ha^-1, which was significantly lower than the 180-240 kg N ha^-1 applied in farmers' practices at the Chinese sites. With the modified farmers' fertilizer practice, a 30% reduction in total N rate during the early vegetative stage did not reduce yield but slightly increased yield and doubled AE_N compared with the farmers' practice at the Chinese sites. The total N rate in RTNM and FTNM ranged from 30 to 120 kg ha^-1 at the Chinese sites, but their yields were similar to or higher than that of the farmers' practice. Compared with the modified farmers' practice, RTNM and FTNM further increased AE_N at the Chinese sites. Overall, FTNM performed better than RTNM at the Chinese sites because the total N rate of FTNM was closer to the optimal level than RTNM. A quantum leap in AE_N is possible in the intensive rice-growing areas in China by simply reducing the current N rate and by allocating less N at the early vegetative stage.