Controllers, especially PIDs, are massively used in industrial practice, and there has been an ongoing search for methods that make controller parameter setting easier, with low requirements on the operator's knowledge. At best, they should be fully automatic, requiring no human participation. A recently-developed method for controller parameter setting fulfils these needs of industrial control practice. The method is aimed at achieving the recommended values of one or more control quality indicators. This is done from an experimentally performed evaluation of excited frequency responses without knowledge of the mathematical model of the controlled process. The indicators can therefore be obtained in control loops involving nonlinearities even in the controller. All data processing is carried out by a program added to the control algorithm. This paper presents another frequency response evaluation technique. It is applied in the search for state feedback gains. By adapting the state feedback gain, the changing dynamics of the controlled process is always tuned in such a way that the model for a controller design remains practically the same. Then, controller setting adjusted for compensating the process dynamics does not need to be changed.