You could receive only the active video and ignore the blanking -- ignore meaning the blanking information does not filter through to the DMA engine so the DMA bandwidth, the data movement inside the part, is saved because you’re not reading the blanking information into the part. Now you could look at only the blanking information, this might be interesting in some kind of interactive TV applications or closed captioning, etc., where you’re interested in the blanking intervals of the video frame, or you can read in the whole video frame (blanking plus active video).