On 22 Sep 2001, at 10:24, Mike Johnston wrote:
>
> I read some fascinating comments recently written by a guy who signs
> himself Peter iNova--I don't know his real name. He pointed out that
> the resolution cutoff for digital is relatively severe--because, in
> his words (paraphrasing), "You can't resolve half a pixel within a
> pixel." But, UP TO that resolution limit, there is NO loss of
> contrast. Where film loses contrast gradually as resolution increases
> (this is what modulation transfer function plots), digital can put a
> solid black pixel next to a solid white pixel, for theoretically
> maximum contrast up to the resolution limit of pixel size. Neither one
> is inherently better or worse (although for specific scenes or
> intentions one might be better than the other), but they ARE
> different.
>
> Plus, smooth areas of color/tone can be much smoother. As we all know,
> in the more heavily exposed areas of a single B&W negative, say a
> clear sky, the film still has to express the tones by a buildup of
> grain clumps. But with digital, a pixel has a single tone and a single
> color. Put it next to another pixel of the same tone and color, and
> there is NO boundary and NOTHING to interrupt the smoothness of the
> tone. This advantage is tempered somewhat by the size of the inkjet
> dots if you're printing in inkjet (dye sub isn't limited in this way),
> but with the current 4 picoliter dot size, I can barely see the inkjet
> dots under a low power loupe...certainly, you can't pick them out with
> the naked eye. Smooth gradations are similarly excellent, because
> putting one blue-sky pixel next to another that is just ever so
> slightly a different color or density is so far below the threshold of
> JND (just-noticeable difference) that it is invisible.
>
> Finally there's color purity--as is well known, color negative
> materials cannot depict something so simple as a pure yellow, because
> the dye layers are always at least slightly contaminated by the other
> dye layers. The best exception has traditionally been dye transfer
> printing, because you lay dyes on the paper completely discretely.
> But, as Frank McLaughlin once said, one lifetime is not long enough to
> become both a master photographer and a master dye transfer printer
> <g>. (Even Eliot Porter, who worked in dye transfer for most of his
> life, enlisted the services of a master dye transfer printer, a guy
> named Jim Bones.)
> ...
>
> --Mike
Hallo Mike,
what you say (or read) about the digital world is only the theoretical truth.
There is no 100% contrast, as all CMOS or CCD sensors can´t block
the charge of one pixel against the others. It is not only a contrast loss but
a color shift as the neighbourhood pixels are of other colors, and it is not
linear but a curve like a Diode. Dark pixels do not interference with the other pixels
but at a certain amount of light (charge) the electrons "break" through and
affect the pixels around.
Well, I won't discuss of all the limitations of the CCD amplifiers.
The sensors get better with the time and better than film I think,
but even the digital world has it's limitations.
Have a nice day
Andreas
-
This message is from the Pentax-Discuss Mail List. To unsubscribe,
go to http://www.pdml.net and follow the directions. Don't forget to
visit the Pentax Users' Gallery at http://pug.komkon.org .
