Re: DSLR vs P&S a replay of Film vs Digital?

Floyd L. Davidson <floyd@xxxxxxxxxx> wrote:

We aren't looking at the same graphs. You appear to be
looking at the curves given on Norman Koren's webpage,
at

http://www.normankoren.com/digital_tonality.html#Dynamic_range

Sorry, I thought we had been talking about that one.



The DENSITY axis ranges from 0 to 4.0, and the curve is
plotted from about 0.3 to roughly 2.8. That is a
dynamic range of 2.5. That is what I am calculating it
with, though I was just ignoring the lower end, as if it
were 0.0, and used a range of 2.8.

I had thought we were talking about the range of light value
the film was capable of recording. It turns out that you were
talking about how dark the negative is.


That is the actual dynamic range of the captured image
on film, not that of the scene.

No, that's not what density is, density is a measure of how
dark the negative is. The actual dynamic range of the captured
image is from the Jones point on the exposure axis to a point
on the shoulder where highlight compression becomes obvious.
This point is never shown nowadays on H&D graphs of negative
films because it is very unlikely to be reached.


We should make it very clear too that Kodak is not
showing higher densities because the characteristic
curves do cover the range that is actually *used* when
generating photographs.

If you set your exposure meter to 400, you will not generally
have content to the right of -0.9 on the exposure axis.
(The maximum usable exposure on a 400 speed slide film would
be around -0.9)

The fact that it is possible to
get a photographically useless negative with a higher
dynamic range is not significant.

Um, it would be better to say "not normally needed or used" than
useless.


The LOG EXPOSURE range is indeed from about -2.9, to
perhaps 0 at best. None of the curves approach +0.3,
except on the chart shown on Koren's site (which and are
also for a longer processing time). But that is the
exposure range, not the dynamic range recorded by the
film.

Well the graph on Norman Koren's site has multiple curves showing
different developing times. The curve marked "8 minutes"
would correspond roughly to normal development.

The exposure range, from the Jones point to the point far
to the right where the shoulder is (never shown nowadays
on negative film because it is so far to the right) is the
dynamic range that the film can record.


There are other curves on the Kodak page, but they all
show less range than the one cited.

This is because the point on the right where the graph
cuts off is fairly arbitrary. They never show the shoulder
and the exact point where they cut off doesn't normally
matter.

It does have one interesting section:

"Well. most properly exposed, properly developed
negatives of typical subjects don't have a Dmax higher
than about 1.5 and even overexposed negatives rarely
go higher than about 2.0, so a dynamic range of 2.0
would probably take care of just about any negative
you are likely to come across. This should be well
within the capabilities of most film scanners."

That's the dynamic range needed by the scanner to scan normal
negatives. The dynamic range captured by the film is larger
than that because the overall slope of the characteristic curve
with normal negative development is around 0.6.


To be clear, that says the typical dynamic range of
properly developed negatives is between 32 and 100
or 30-40 dB. Granted that with over exposure or over
development that same film is capable of get 20-30
dB more dynamic range, but when it is used to best
effect for photography, it *doesn't*.

The output dynamic range of a negative is not relevant.
You don't look at negatives directly except when deciding
which ones to print. If you print a normal negative on 5302
print film developed to gamma of about 2.8 then you will get
a slide with the same density range as a colour reversal slide.

Which is to say, for those who want to argue that film
has more dynamic range, Tri-X and other print films are
not the direction to point. Slide film has *significantly*
higher dynamic range.

Slide film has a higher output dynamic range, but a lower
input dynamic range. The overall characteristic slope
for slide film is around 1.7. The overall characteristic
slope of negative film with normal development is around
0.6. The Goldberg criterion for viewing projected slides
requires an overall gamma of around 1.7 for the contrast
to look right.

Tri-x is a negative film. A "print film" is something like
Eastman 5302 which is used for making print transparencies.
Tri-x is capable of recording a huge range. Try doing a simple
latitude test carried over to massive overexposure. I've
done such tests and it takes a massive amount of excessive
exposure to show obvious highlight compression.

Correct negative exposure is based on the amount of shadow
detail you want to show in the final print. The headroom is
so huge that you do not normally have to worry about it.
Of course, if you really need to capture the inside of
a tire tread in the shade and bright sunlit clouds, you
are going to get a negative which is near impossible to print.

Peter.
--
pirwin@xxxxxxx

.

Relevant Pages

  • Re: How far...
    ... especially DSLRs have higher dynamic range than film. ... Dynamic Range and Transfer Functions of Digital Images ... means you didn't meter and expose correctly. ... With a digital camera you expose to not blow any highlight you don't want blown, then you generally have the signal to noise to pull the image from the shadows. ...
    (rec.photo.digital)
  • Re: DSLR vs P&S a replay of Film vs Digital?
    ... In theory a 12-bit digital camera can have a 72 dB ... The graphs for film that we've looked at in this thread ... Even the Kodak curves for Tri-X don't ... 3.0, or roughly about 631:1, for 56 dB of dynamic range. ...
    (rec.photo.digital)
  • Re: How far...
    ... especially DSLRs have higher dynamic range than film. ... Telling me the cameras would help. ... just like photographers have to learn to expose slide ...
    (rec.photo.digital)
  • Re: high dynamic range in P&S ??
    ... highlights and amplifying mid and low range signals. ... None of that will increase "apparent dynamic range". ... misunderstanding of exposure technical details. ... It works for both digital and film, ...
    (rec.photo.digital)
  • Re: DSLR vs P&S a replay of Film vs Digital?
    ... 3.0, or roughly about 631:1, for 56 dB of dynamic range. ... It does not show Tri-X at a Dmax above about 2.8. ... I was referring to the curves on Kodak's webpage, ... on film, ...
    (rec.photo.digital)