Digital BW, The Print

Hi Kevin,

> Is zone system development time manipulation irrelevant when scanning
> film as opposed to traditional printing?

An excellent question.  It is ABSOLUTELY relevant.  What you are doing by
using the Zone system is placing as wide a tonal range as you can on your
film, period.  In order to do that, you use exposure compensation and
development compensation.

> Since I am not doing wet prints but am scanning the negatives, it seems
> to me that the N- or N+ development dependent on the scene brightness
> range is, mostly, irrelevant.

Well, no...see above.

> By setting the black point, setting the
> white point and scanning the negative am I not mapping the entire
> density range of the image to a numerical range of 0-256 or 0-64k?

Not quite.  The setpoints have not a thing to do with the scan.  They are
applied to the scanned data AFTER the scan occurs.  I believe that's
actually not the right question.  The question should be how to get the
highest density range on the film THEN how to take advantage of that.

I'll explain the scanning process as it related to setpoints.  The scanner
scans the film, and place the density range of the film in an N bit space.
The density occupies only a portion of that N bit space, and how much of
that it occupies is up to the density range of the film.

You can verify this by doing a raw scan, and looking at the data from that
scan with the histogram tool.  It will only occupy a small portion of the
overall range.

What you do when setpoint is "mark" the endpoints of the actual image data
within the overall N bits the scanner scanned at.  Example, your scanner is
12 bits, that gives you a range of 2**12 or 0-4095.  You scan.  The actual
image data occupies data values from 200-766 in the raw scan.  When you
setpoint this data, it takes the 567 values of image data, and spreads them
out over either 8 bits (some get combined when going to an 8 bit space), or
over 16 bits (where they get spread out).

If the number of data values is less than the space you are mapping it into
(577 values into 16 bits) then there will be "equal" (not necessarily
perfectly equal, but as close as can be done) gaps between the data values.
A simple example of that...you have a 4 bit scanner (0-15) and the data
occupies 3-10.  You want 4 bit data out of the scanner, so you set your low
setpoint at 3, and the high one at 10 and the data gets remapped as follows:

Original image values 3-10 get mapped to 0-15 as follows:

3 -> 0
4 -> 2
5 -> 4
6 -> 6
7 -> 8
8 -> 10
9 -> 12
10-> 14

note gaps in the data.  This is no big deal as far as real image data goes
(meaning 12 bit scanned data, mapped to 8 bit data to the printer or
screen).

Map that same 3->10 into a 2 bit space:

3 -> 0
4 -> 1
5 -> 1
6 -> 1
7 -> 2
8 -> 2
9 -> 3
10-> 3

> This mapping would take place independent of the absolute density any
> particular zone.

That is true.  There is no directly calibrated correlation between scene
density, film density and scanner value....unless you went out of your way
to calibrate it...which is really unnecessary, as doing so won't get you
better scans.

> As a hypothetical example lets assume a scene contains a 8 stop range
> of brightness.  Three images are similarly exposed to capture that
> brightnesses range. The three images are given different development
> times and produce density ranges of  (1.0-0.3=.7), (1.4-0.4=1.0) and
> (2.0-.5=1.5).  When the images are scanned each one will produce a full
> histogram from 0 to 255

Only after setpointing the scanned data into an 8 bit space.

> and a scene brightness at the 6th of the eight
> stops will show up at the same place in each of the histograms.

Possibly...

> Soooo, can't I simplify the zone mantra to "expose for the shadows and
> let the highlights fall where they may with normal development".

No, because you can expose to get a higher scene density range onto the
film.  The actual density range of the film is irrelevant to how much scene
density range you can record on the film.  BUT...the wider the film density
range, the more discernable data your scanner will record, because of how
the scanner works.

> Also,
> wouldn't it be better to generally use N+1 development times

It depends on if it's a high or low contrast scene.  Plus
exposure/development is EXPANDING the scene range into the film range, for
use with LOW contrast images.  Negative exposure/development is CONTRACTING
the scene range into the film range for use with HIGH contrast images.

> so that
> the numbers from the raw scan occupied more of the scanner's range?

Yes, exactly.  You want to contract/expand the SCENE range into the film
range, therefore giving you more scene tones on the film and therefore more
discerned scanned values.

If you already knew the answer, why get me to go through the whole process
;-)

I think I said all that correctly.  I'm sure I'll get a lashing if I typo'd
or missed something...

Regards,

Austin