The record‐playback method of making noise comparison tests provides realistic answers with a minimum of equipment and sophistication. Such a method is especially attractive in problems handled by making experimental changes which, taken individually, usually produce very small changes in level and quality of noise.In automotive work, playback of single‐channel recordings of interior car noise sensed by a single microphone sound unreal because of an emphasized bass, the cause of which is not known to the author. Directional effects also contribute to the lack of realism.The use of a binaural magnetic tape recorder improves realism, which in turn enhances the resolution of smaller differences.