> OK, I'll buy it if you give me the actual dynamic range numbers.
Without jumping to heroics, I was getting 12-14 stops of DR and the
highlights and shadows didn't go wonky. With heroics (adjusting the
scanner's exposure times" I'm seeing another two stops lurking in
there, but depending on the picture, this may or may not be overly
usable.
That's just with the digitizing process. B&W film in an analog
environment is unmatched. You can dodge and burn and not get wierd
things happening because there were too few bits representing half the
stops of range.
Honestly, with XP2, I'm not sure what the upper limit on DR is.
However, if you lean too far into the shoulder or toe, you do suffer
some lack of tonal separation because the S-curve is so flat. (Midtone
gray is already one full stop into the shoulder). XP2's DR exceeds the
ability of my scanner to capture no matter what I do. In the darkroom,
I can do things to pull stuff out that really changes things. Ilford
Pan-F and Delta 100 have different curves and don't have as much
absolute dynamic range, but they put what dynamic range they do have
to better use.
I know what the numbers say, Chuck. 12 stops should be 12 stops should
be 12 stops. When you do a straight top-and-tail of the histogram and
hit print, both technologies are going to be similar. But when you
start pulling those highlights back and lifting those shadows up, film
behaves totally differently. With the highlights, it's not uncommon to
have color shifts as well as some nifty rimming effects with things
like the setting sun or any other bright colored light source. Shadow
lifting gets nasty, hence the necessity to expose-to-the-right
whenever possible to avoid having to lift the shadows. Cameras that
dither the output of the sensor (Olympus E-1, for example) mask a lot
of those nasties, but at a price. With film, provided the scanner
(which is a digital camera) is set to adequately capture the full
dyanamic range of the film (not as easy as it sounds sometimes) will
start to reveal details in the shadows and highlights that just aren't
there with digital capture.
My best concert photos have been and remain those taken with Fujifilm
Press 800, followed closely by NPZ. Colored stage lights do horrible
things with digital capture. Film is not such affected and that
shoulder gets leaned into in a way that red lights stay red, purples
stay purple. With digital, the reds turn yellow, the purples turn
blue. Yellow lights will turn the performer into Big Bird. (I had
major Big Bird Badness last week).
When shooting things like weddings, the Portra films are unmatched.
I've NEVER had a worry, EVER, about capturing the white dress and
black tux in the most extreme of lighting conditions. Portra 400NC
(now just straight Portra 400) just nails everything dead on and the
skintones make you wonder just why we shoot digital?
So, when you compare the DR of Portra 400 with, say, a Canon 5D (which
remains a DR leader), even though the numbers are close, it's still no
contest.
Where the DR numbers really do matter is when you compare digital with
reversal films. Velvia is an obvious one which gives maybe 8 stops on
a good day, depending on several factors, Provia is about a stop
better, but isn't as picky. Astia will give another stop and isn't
picky at all. Ektachromes are all over the boards, but tended to give
a bit better range than equivalent Fujichromes. Agfachrome really set
the standard and gave a very wide range. In fact, I have one set of
pictures of an abandoned farmhouse I could never capture with my
standard Fujichromes.
B&W films are quite variable. Not only does it matter what brand and
model of film you are using, but how you expose and develop it. There
are some fantastic claims of extremely wide dynamic ranges, but those
require odd exposures and development plans. It's fascinating that in
the world of B&W, where EVERYTHING is known, that we are still
discovering new things about the films, developers and methods.
Especially so, since some information about the films has been leaking
out of Kodak. (have you hugged your APUG neighbor lately?)
I've got the spec sheets for a bunch of these films, and in actual
practice, I've found that my usability in scanned images matches these
specs very closely, but doesn't take into account how much can be
rescued from the toe and shoulder. I've been able to gain a full stop
of missing information in the shadows of Velvia. I'm able to pull
stuff out that can't even be seen by examining the slide on a light
table. With some curves adjustments, I'm able to usually get another
full stop of straight-line section from the Fujichromes. In the case
of negative films, I'm usually having to increase the contrast of the
image to reduce the effective dynamic range since the images don't
look "normal" to today's expectations.
I did do dynamic range testing of Ektar 100 compared to the 5D. I'll
have to go back and look at the images again. I didn't get around to
publishing the results of the tests since there were some procedural
things which would be easily skewered, but we were seeing two to three
stops of additional USABLE dynamic range over the 5D even when a
maximum amount of highlight and shadow recovery was applied to the 5D
image. That's not to say the 5D was bad. Far from it. It still remains
the best digital camera I've ever tested in this regard.
Oh, on an oddly related note. With the big event shoot, Lightroom 4
was able to very closely match colors between the E-1 and the DMC-L1
using Auto-Tone. The midtones and skintones almost perfectly matched
(with a slight adjustment of WB on the DMC-L1 files, but the top and
bottom edges of the dynamic range were vastly different. Lightroom got
everything in the middle to match, but sacrificed a lot on the outer
limits. My new computer (i5 with 6Gb RAM) performed like a champ, but
boy did it get HOT!!! The cooling exhaust vent was almost too hot to
touch..
AG
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|