Experimenters tested the training benefits of an instructional strategy in which simulated events in a real-time simulation are made to occur faster than normal. Two experiments were conducted to evaluate applications of above real-time training (ARTT) for training air combat skills and emergency procedures. In the first experiment, experienced, Air Force F-16 pilots practiced emergency procedures and air intercepts using conventional, real-time simulation or ARTT at 1.5 times real time. The pilots trained using ARTT received the same number of training trials but less clock time in the simulator as pilots trained in real time. All pilots were then tested in real time. Pilots trained using ARTT performed emergency procedures and defeated bandit aircraft significantly faster than pilots trained in real time. In the second experiment, student F-16 pilots practiced using air-to-air radar in real time or ARTT. Students trained using ARTT received more training trials in approximately the same amount of clock time as the students trained in real time. ARTT students performed better on a real-time test than students trained in real time. It is concluded that ARTT which does not overload a trainee's working memory is more time efficient than conventional, real-time simulation and can improve performance by allowing more training events to be experienced within a given period of simulator time. ARTT also supported better real-time test performance under some conditions. As an instructional strategy, ARTT was found to be simple, inexpensive, and robust.