
The closed-captioning datastream is the series of flashing white dots & dashes at the very top of the frame, when viewed in underscan. What I'm seeing in a horizontal slice just below it is actual video from the full width of the 16-by-9 frame, video whose motion is evidently a continuation of the picture concealed by the pillar boxes that begin just underneath this full-width sliver.
The closed-captioning datastream is the series of flashing white dots & dashes at the very top of the frame, when viewed in underscan. What I'm seeing in a horizontal slice just below it is actual video from the full width of the 16-by-9 frame, video whose motion is evidently a continuation of the picture concealed by the pillar boxes that begin just underneath this full-width sliver.