Mailinglist Archives:
Infrared
Panorama
Photo-3D
Tech-3D
Sell-3D
MF3D
|
|
Notice |
This mailinglist archive is frozen since May 2001, i.e. it will stay online but will not be updated.
|
|
re:Digital vs Analog (might be backwards!)
- From: P3D Michael Kersenbrock <michaelk@xxxxxxxxxxx>
- Subject: re:Digital vs Analog (might be backwards!)
- Date: Thu, 12 Dec 96 11:10:19 PST
> The meaning of analog is that there a property of one material analogous
> to that of another. In the case of LP, the position of the groove is
> analogous to sound pressure. Photographic film is analog: the amount of
> some chemical in the film is analogous to the amount of light used to
> expose it.
Although color films might be different, I think the silver specks that
constitute a B&W negative are digital-like. Micrographs of the
specks and their pattern that generate the image seem quite similar
to halftoning/dithering techniques used in printing or first-generation
cheapie B&W scanners.
The specs themselves at any given *point* are either there or not-there
and the light is blocked at that *point* by something opaque or not blocked
and is clear. So it's really digital-like where the smallest spec of silver
yields the approximate binary-pixel resolution.
In a digital camera, *each* pixel can be not only black or white, but
in most cases least 254 brightness values inbetween, which seems a lot
higher tonality than a singular minimum-sized silver spec. And as such,
it takes an area of the film to provide the tonal range that an equivalent
digital-camera pixel has. This is why it seems that the silver specs are
acting like dithered/halftoned binary representation of the image. One then
can conclude that silver-based film is more digital than a digital camera
because on the microscopic scale, each pixel in a digital camera *is* roughly an
analog representation, while each "pixel" of silver film is roughly binary.
This view of the film also explains and makes reasonable data I've seen
in recent news stories talking about digital images that claim that the
equivalent resolution of 35mm film is 3 million pixels, not the much larger
numbers put forth in this forum.
Each pixel in most current digital cameras represents a 24-bit value of
colors (8-bits, or 256-levels per color), where with silver images, the
"silver-pixels" are binary, only 2-levels per pixel. In other words, it
takes roughly 8 silver-film "pixels" to do what 1 digital-camera pixel does.
Perhaps those of you more familiar with the color film structures can comment
about how big of an area (on film) it takes to yield a 256-level tonal scale
*per color*. Might be different for different film structures (Kodachrome
vs E-6) as well.
I know I am handwaving and probably seem ridiculously simplistic to those
of you who can rattle off the crystal configuations and chemistry details
like child's play, but I think I'm being reasonable, and it does make the
figures in the press make sense, which is what I primarily was trying to do.
Mike K.
P.S. - I left out discussion of "addressablity" vs. "resolution" where film has
an advantage, however that advantage is kinda-roughly equivalent to the
"software resolution enhancement" that software can implement in scanners.
And in any case, I was focused only on resolution which is what articles
I've read recently mentioned.
------------------------------
|