The solution of a 2 1/2 degrees of freedom H-infinity/H-2 stochastic optimal control problem is considered for SIMO or multiloop systems. The optimization is based on an extended cost-function which allows separate costing of the terms due to the feedback and reference controllers. This enables the robustness of the feedback-loop to be shaped independently of the tracking performance. Simplifications can be achieved over the more general multivariable problem which is important in self-tuning control problems. The tracking problem allows a model-following capability to be introduced, and a predictive control feature is also available. The solution is valid for 1, 2 and 2 1/2 degrees of freedom designs.