How many bytes of decoded frames a strip holds.
Each decoded frame counts its own pixels against it, so a strip drawn taller holds fewer frames under the same cap.