Interfacing of microprocessors with isokinetic dynamometers has enabled the rapid quantification of many parameters of muscle function including peak torque, angle-specific torque, work, power, torque acceleration energy, and various endurance indexes, and the measurements with these devices can be made isometrically at various angular positions and isokinetically (concentrically or eccentrically) with a large scale of angular speeds. Many of these parameters, however, lack evidence of validity, reproducibility, and/or clinical relevance. Peak torque has been and still is the most properly studied isokinetic strength testing parameter and its use can be recommended for research and clinical purposes. Concerning testing of muscular endurance, the absolute endurance parameters (for example, work performed during the last five repetitions and total work in a 25-repetition test with a speed of 240 degrees/s) are the best for use. Many internal and external factors in the isokinetic testing procedure can have an undesirable effect on the test result. However, through proper education and strict adherence to the test instructions, it is possible to successfully control the confounding variables. In scientific work, isokinetic devices have greatly expanded the possibilities for studying dynamic muscle function. There is also little doubt about their usefulness in documenting the progress of muscular rehabilitation. A disadvantage of isokinetic devices is that isokinetic movement seldom occurs in actual human performance tasks and that the isokinetic training effect is, therefore, quite (although not completely) specific to that type of movement. In addition, being normally an isolated joint exercise, isokinetic training can produce large loads on the involved joints and may, therefore, under certain conditions be dangerous for healing tissues.