Re: [Leica] re: digital noctilux vs 10X faster films?

["Mxsmanic" <mxsmanic@hotmail.com>]

Anything that reacts to single photos has hit the physical limit, so if film
manages that, it is absolutely certain that CCDs will never beat it.

The best refrigerated CCDs already are sensitive to individual photons.  So is
the human eye, under good conditions.  Film is still quite a distance off, as
you describe below, but the use of fomates shows some promise.  Unfortunately,
you'll note that the article to which you point is nearly two years old, and we
still do not have ultrasensitive films.  Until they are on the shelves, the
accomplishments seen in the lab don't amount to much.

Despite this, current CCDs in digicams are still inferior to film in
light-gathering ability, in part because they are still too tiny.  Digicams seem
to have ISO equivalent ratings around 80-100, whereas film that is ten times
more sensitive is readily available.  Digicams let you boost the gain for a
higher ISO equivalent in some cases, but noise rises a lot, too (particularly in
consumer gear), so that isn't really the same thing.  More importantly, you
cannot reduce the speed of a CCD and get better resolution or less noise,
whereas you can switch to a film like Tech Pan and get extraordinary resolution
with essentially no grain.

This is one reason why 36x24 CCDs are going to be a necessity.  You just cannot
continue to add more and more pixels to a very tiny CCD without adding noise and
reducing sensitivity to light.  It makes a lot more sense to just make the CCD
bigger.  A 36x24 CCD is one of my prerequisites for any switch to digital, not
only for these reasons, but also because I wish to preserve my investment in
lenses.  The 35mm format is at a disadvantage compared to MF because it is
smaller, and CCDs are at a disadvantage compared to 35mm film because _they_ are
smaller.  There isn't any way to get around that.

- ----- Original Message -----
From: "Robert Monaghan" <rmonagha@post.cis.smu.edu>
To: <leica-users@mejac.palo-alto.ca.us>
Sent: Wednesday, July 25, 2001 02:57
Subject: [Leica] re: digital noctilux vs 10X faster films?


> don't give up on film quite yet, see ten times faster film emulsions page:
> http://www.cnrs.fr/cw/en/pres/compress/emulsionsphoto.html
>
> quote: The yield is then _eff = 2 silver atoms per photon, i.e. 10 times
> higher than prior yields. The mechanism is such that the number of atoms
> is proportional to the number of photons in a very wide range of light
> intensities, including zero or low exposure, and that the sensitivity gain
> does not suffer from any accompanying fogging effect. endquote:
>
> it is also unlikely that future digicams will be able to use much smaller
> sensors due to issues like wavelengths of light and noise levels, viz:
> http://www.dpreview.com/news/0009/00091102foveon16mp.asp
>
> quote:Mr. Mead said that because of fundamental size limits in the
> wavelengths of light, it is unlikely that future digital sensors will gain
> much additional resolution. endquote (Mead developed 16 MP Foveon chip)
>
> Assuming that Agfa et. al. come out with 10X faster films as hoped and now
> evidently technically possible, with better linearity and less fogging,
> that translates into 3+ stops speed for film, and about equal light
> efficiency to current CCD sensors (unless you intend to supercool your
> chips?). Granted, digicams will be able to do software tricks, and image
> intensifying tubes or technology will be available (but also to film
> cameras?). But the technology wars aren't over yet, and they aren't so
> one-sided as many would have us believe ;-) grins bobm
>
>
> * Robert Monaghan POB752182 Dallas Tx 75275-2182 rmonagha@mail.smu.edu      *
> * Third Party 35mm Lenses: http://www.smu.edu/~rmonagha/third/index.html    *
> * Medium Format Cameras: http://www.smu.edu/~rmonagha/mf/index.html megasite*
>