This paper presents a theoretical and experimental study of the ability of common class-D switch-mode control loops to suppress various errors signals. Such errors are categorized as either of conventional continuous- time (CT) nature, e.g., hum and thermal noise, or as discrete-time (DT) nature being impulses at the pulse edges, e.g., distortion due to power stage deadtime or timing jitter. It is shown both theoretically and experimentally that DT error suppression can be dramatically worse than for CT errors depending on the carrier generation/modulation mechanism used. The theoretical modeling is based on a previously presented discrete-time small signal model. An experimental measurement setup is presented along with measurement data compared to the theoretical model.