Reliable and efficient techniques are now available to synthesize gain-scheduled controllers. Such techniques take advantage of linear matrix inequality (LMI) characterizations of gain-scheduled controllers and provide a global and systematic treatment of the gain-scheduling problem. This paper examines how such controller structures, designed in continuous time, can be discretized and implemented in practical applications. A special emphasis is placed on reducing the computational cost of discrete implementations. The validity of the proposed concepts and schemes are illustrated through a two-link flexible manipulator design example.