Einton Newstein wrote:

Why a Canon DSLR of only 3k x 2k @ 32-bit RGB out performs a film
scanned to the same resolution and same color depth? The grain factor?

Not with some fine grained films, and the latest in drum scanning. Also,
consider that some colours are outside the RGB gamut range of a CCD or
CMOS sensor, yet some films have little trouble capturing a wider range
(like anything approaching Yellow or Cyan).



The tradistional film has the layered color sensor, unlike the digital
camera that has planarized color array.

Layers of colour sensitive emulsion overlap in film structure. A CCD or
CMOS imaging chip is really monochromatic by comparison, with a Bayer
pattern and software to interpolate colours. I suggest reading this
article to understand a bit mo

http://www.peter-cockerell.net:8080/Bayer/bayer2.html

Intuitively the digital camera
should be inferior, quite opposite to the current observation. There
are arbuments pointed to the grain noise factor in the film.

Only a few articles have actually tested the resolution of imaging chips,
and they work out near 50 lp/mm at best. Getting that resolution, or
better, on film should not be too difficult. Colour is another issue, as
is grain or noise. Also, many who complain about grain judge it on a
computer monitor; and often the the grain that can show on a monitor will
be absent in the final print.

Another issue is aesthetic considerations. While a direct digital image
may lack fine tonal details, it can often provide a pleasing or
compelling image. Many of us like paintings, yet they are decidedly low
resolution. Interesting and compelling images have been made in the past
on film, and there should be little reason to think that should not
continue; and the same should be true for direct digital imagery.



If so, let's imitate what the digital camera does. If a film is made
like a chemical color sensors, would it make it more digital friendly?
and perhapes claim back the ground?

If I understand what you are suggesting, then placing a Bayer pattern in
front of film should allow an emulation of direct digital imaging.
Another test might be to place a micro lens array in front of the film.
My guess is that either method, or both, would degrade the film image.

Ciao!

Gordon Moat
Alliance Graphique Studio
http://www.allgstudio.com