Re: Digicam vs. Scanner resolutions

[SKID Photography <skid@xxxxxxxx>]

byard pidgeon wrote:

> Supposedly, 6 megapixels is the threshold point for making "real" photos
> with a digicam...but, making a "real" photo quality neg or slide with a film
> recorder requires just under 12 megapixels (and a good FR, like a PCR II at
> least).
>
> So, it seems to me that what we really need are 12 megapixel digicams, and
> we're only halfway there.
>
> I'm defining the quality factor as being sufficient to make a neg that
> produces a photo print of at least 11X and preferably 16X that's
> indistinguishable from a camera neg.
>
> Along the same lines, my film scanner yields about 14 megapixels...doesn't
> it make sense that 12-14 megapixels is what we need in a pro quality
> digicam?

Well, sort of.  A '6 megapixel' capture chip in a digicam really has only a 1/4
to a 1/3 of those stated pixels, as the camera manufacturers are passing a huge
lie off on the public.

When a spec sheet says it has 6 megapixels, what they really mean is that there
are 6 million receptors, each receptor being an R,G,G, or B (only).  (These
cameras use 2 G receptors for every pair of R&B receptors.)  A 'pixel' is made
from an R,G, & B component, but each sensor only captures a third of a pixel's
worth of information, the rest is interpolation, which is to say:  Making up the
information.  Albeit, through a very good and sophisticated algorithm, it's not
'real' full pixel info it's capturing...Unlike what one gets from a scanner,
which are real RGB pixels.

So draw your own conclusion as to how big a chip they will need to replace
film.   Also, the new 'Foven' chips might be just the ticket, as they, at each
sensor point, capture an RG & B component (a pixel) at different wavelengths
(they focus at different depths on the chip).  Some have said that these chips
will have other shortcomings, but that remains to be seen.

Harvey Ferdschneider
partner, SKID Photography, NYC

-
Turn off HTML mail features. Keep quoted material short. Use accurate
subject lines. http://www.leben.com/lists for list instructions.