I think this was done as the "minimum" delay between IR commands is already quite long, unlike models based on the original Pronto.