re:Digital vs Analog

[P3D Stephen J Hart <sjhart@xxxxxxxxx>]

Greg W. wrote:

>I don't know how to explain it any clearer.  Halftoning works because
>at some level, someone (e.g. a human applying a screen) or something
>(an algorithm) knows that you can approximate a grey level using only
>black/white "pixels" by trading off against resolution (you spread out
>your virtual, grey, pixel among some number of physical black/white only
>pixels, at the cost of a much larger "virtual" pixel).  In this
>arrangement, the real pixels that comprise a virtual pixel are related;
>a function related to the desired grey level that says which ones will
>be on and which will be off.
>
>The grain in film has no such mechanism.  Each silver salt crystal acts
>on its own without regard for what its neighbor is doing.  The film
>CANNOT be binary as you suggest because there is no mechanism instrinsic
>in the film to distribute the virtual pixel information around to the
>physical grain structure.

The grains don't have a mechanism on their own, but the light + grains does.

The Gurney-Mott theory of latent image formation says that a silver-halide=
 grain in an emulsion becomes a stable latent grain after it captures a=
 certain number of photons (typically four I think, though that's not=
 relevant to the gray-scale or resolution arguments). In the latent state a=
 silver-halide crystal will convert to a silver grain (black) when dunked in=
 developer. If it catches fewer photons during the exposure it stays as a=
 regular silver halide crystal, isn't altered by the developer, and is then=
 extracted by the fix.

Given this, latent image formation has to be seen in terms of photon=
 capture, and photons are "digital" (i.e., discrete) in this context.=
 There's no directed error-diffusion going on, but the recorded light=
 arrives in little photon-sized packets, each of which has a statistical=
 chance of latentizing a grain. The gray-scale image is then formed by the=
 net effect of many such binary grains, distributed pretty much at random in=
 three-dimensions within a clear gelatin layer. Grain sizes do vary, and=
 this does influence speed, contrast, and resolution, but the gray-scale=
 isn't normally due to grain-size or the blackness of any one grain.

>Well, all I can say is consider this:  a hologram, which is recorded on
>photographic emulsion, is the representation of the interference pattern
>produced by two coherent wavefronts.  In order for a hologram to work,
>the emulsion must be able to record with a resolution slightly greater
>than the wavelength of the light used.  That puts it down into the
>hundreds of nanometers.  There ain't no digital imagers with pixels
>that small yet.  (This is black-and-white film, and extremely slow film
>at that.  I don't believe there is a color film with anything near
>this resolution.  I doubt it would be possible to make one.)

Indeed, a good holographic film will have a grain size of around 30=
 nanometers. This allows it to faithfully record holographic fringes which=
 are typically much narrower than 1 micron. Quoted resolving powers are=
 typically in the 5,000 line-pairs per millimeter range. At this size the=
 grains are pretty insensitive! So while a 35 mm sized chunk of holographic=
 film would in theory resolve something like 10^11 (one hundred billion)=
 two-dimensional "pixels", the film speed is only about 0.1 ASA. In our=
 holographic systems we routinely record fairly coarse (still sub-micron)=
 fringes through the six-micron thickness of emulsion on a 14 x 17 inch=
 film... that'd be at least 10^12 discrete points if we could individually=
 address them (we can't).

Silver-halide isn't even the highest-resolution holographic material. There=
 are, for example, photopolymers which can have molecular levels of=
 resolution, but with photographic speeds roughly 1,000 times slower.

=46inally, color holography just needs a film with the relevant spectral=
 sensitivity. Each color is then recorded as a different fringe pattern, and=
 holographic effects produce the required color-selection on playback. There=
 are other techniques too, but color holography does not in principle=
 *require* any kind of spatially-encoded color sensitivity.

Stephen J Hart                      fax              +1 (714) 348-8665
Director of R&D                     e-mail            sjhart@xxxxxxxxx
                                    WWW URL      http://www.voxel.com/

VOXEL, 26081 Merit Circle, #117,    The opinions and "facts" expressed
Laguna Hills, CA 92653-7017, USA    herein are not necessarily correct



------------------------------