C and A weighting apply different values to different octaves. Can't remember for sure which is which, but one is based upon equal sound energy in each octave, the other is based upon the 'normal' human hearing curve. The slow response just keeps the meter from bouncing around too much, giving more of an average reading.