vocab.design

layout · platform-registers

Overscan safe area

also called title safe area (community), action safe area (community), overscan margin (android), TV safe area (community)

The inset a television layout keeps clear of the screen edge, about five percent, because sets crop the outer frame and anything near it disappears.

Television sets have never been trusted to show the whole picture. Cathode ray tubes were deliberately driven to scan slightly beyond the visible glass, so that ageing components and mains voltage drift produced a picture that still filled the tube rather than one with black edges creeping in, and broadcasters answered by keeping everything that mattered well inside the frame. Panels are digital now and the geometry is exact, but the habit survived in the sets themselves: many still default to a mode that scales the incoming picture up and crops the outer few percent, and a fair number ship with no way to turn it off. So a television layout still keeps a margin, and that margin is the overscan safe area.

The working vocabulary is a pair, and it is worth getting right. The action safe area is the inner ninety percent or so, and it is where movement, video and graphics belong: content there survives on any reasonable set. The title safe area is tighter, about the inner eighty percent, and it is where text, logos, scores and captions belong, because a letter with its top row of pixels shaved is not a slightly cropped letter, it is a typo. Android’s TV guidance states the same idea as a flat overscan margin, forty eight density-independent pixels across and twenty seven down on a nine hundred and sixty by five hundred and forty layout, which is five percent of each dimension.

The related terms are close enough to be confused, and the difference is what the screen shows. A phone safe area is about a screen that displays every pixel it is given, where the obstacles are notches, system bars and rounded corners sitting on top of the picture, and a display cutout is one such obstacle. Overscan is not an obstruction at all: the pixels near the edge are simply never displayed, and no inset query will tell you how many of them were lost. That is why the number is a convention rather than a measurement, and why the web platform has no env() value for it.

In practice the discipline is the same as edge to edge on a phone, run the other way round. Backgrounds, video and full-bleed artwork extend to the physical edge of the layout, because losing five percent of a gradient costs nothing. Text, buttons and anything focusable stop at the title safe line. Two extra pressures push the same way on a television: at ten foot viewing distance type is already large, so the margin is cheap, and a remote-driven interface needs room around each focusable item for the focus ring and its scale-up, which is exactly the room the safe area was going to reserve anyway.

Which word?

If you wantsay
laying out for a screen that crops its own edgesoverscan safe area
naming the hole the layout must dodgedisplay cutout

Related

See also: 10-foot UI · Letterboxing

Sources