Long-term tillage and fertilizer experiments were conducted in rice in kharif followed by lentil in dry subhumid Inceptisols at Varanasi and Faizabad; horse gram at Phulbani and linseed at Ranchi in moist subhumid Alfisols in rabi during 2001 to 2010. The study was conducted to assess the effect of conventional tillage (CT), low tillage + interculture (LT1) and low tillage + herbicide (LT2) together with 100% N (organic) (F1), 50% N (organic) + 50% N (inorganic) (F2) and 100% N (inorganic) (F3) on productivity, profitability, rainwater and energy use efficiencies. The results at Varanasi revealed that CT was superior with mean yield of 2389 kg ha−1, while F1 was superior with 2378 kg ha−1 in rice. At Faizabad, CT was superior with mean rice yield of 1851 kg ha−1 and lentil yield of 977 kg ha−1, while F1 was superior with 1704 and 993 kg ha−1 of rice and lentil, respectively. At Phulbani, F2 was superior with rice yield of 1170 kg ha−1. At Ranchi, F2 with rice yield of 986 kg ha−1 and F3 with linseed yield of 224 kg ha−1 were superior. The regression model of crop seasonal rainfall and yield deviations indicated an increasing trend in rice yield over mean (positive deviation) with increase in rainfall at all locations; while a decreasing trend (negative deviation) was found for lentil at Faizabad, horse gram at Phulbani and linseed at Ranchi. Based on economic analysis, CTF1 at Varanasi and Faizabad, CTF2 at Phulbani and LT2F2 at Ranchi were superior.