Underexposure Pixelation: Why can't black just be Black? (Film vs Digital)

[Phil Rhodes]

There's something to be said about the original question, too.

In theory, in modern digital video systems, black absolutely can just be black. If you feed in a pristine, digitally-generated frame full of zeroes, it's quite likely you'll get something very like that out (there are reasons why that may not always be completely true, but it'll generally be mostly true). Modern codecs are very capable of taking a block of image and deciding that the best way to compress it is as a defined area with a single colour in it.

The problem, of course, is that well-shot moving image material tends at the very least to shade off gradually into black, and sensors do not have zero noise. That noisefloor is, with some engineering opinion involved, defined to be "black." As a result what we're feeding into codecs is not entirely black and what we get back out is not entirely black.

Given that film has a maximum density and a grain structure this was always true for film, too, but I guess the issue is that it always just looks like a speckly pattern of noise as opposed to turning into squares.

P