Acoustical design will always involve a decision of the conditions required in terms of physically measurable quantities followed by design techniques aimed at achieving the required results in the final building. Studies by the authors were initially concerned with the use of acoustical models as an aid in the design techniques. The requirements for the model surfaces and two acoustical models have been studied but finally an impasse was reached when measurements sufficient for testing model similarity were not known. The response of particular listeners to music and speech in typical auditoriums and to sounds with artificially added echoes showed a reasonably broad tolerance of acoustical conditions. Different individuals had varying requirements and with program variation also the over‐all tolerance for a hall is quite large. These considerations suggest that the wealth of detail supplied by acoustical measurements is not matched by any corresponding subjective response and that further search for more critical acoustical parameters will be unrewarding.