>
> Ken, you are really bad, if our Disneyland had not changed the lighting
> setup I have to buy another E-thing now and throw away the 5D II. The two
> shots were made at different date, both straightly output from RAW:
>
Well, I know which of the two pictures I prefer. The blue haze noise in the
E-thingy shot is easily cleaned up with chroma noise reduction and a touch
of curves adjustment on the shadows.
What I recognize is the use of LED lighting. Photographing productions lit
with LED lighting is becoming a gut-wrenching nightmare. The color shifting
is so severe that we have literally no way to color match any more. If you
get the skintones matching, the wash color is totally different. If you
match the wash color, the skintones turn cadaverish.
Unfortunately, nearly all of the stage stuff I've photographed this year has
been heavily LED lit for the colors, but the spots and moving lights all use
gels or dichroics. It is really impossible to get stuff matching anymore.
This is especially a problem if the lighting uses a mixture of LED and
gelled hot lamps to create a color wash over the entire scene. To the human
eye, everything matches, but to the camera, it does anything but match.
However, I have found that Fuji Superia 400 and 800 films hold the colors
much closer to how the eye sees the scene.
Almost without exception, I get far better results shooting film than
digital in stage productions. Your pictures show one common problem with
digital. To hold some color and brightness in the background, anything lit
by follow spots gets burned out. With film, I can lean into the shoulder of
the film and still get great skintones and highlights without burning out
while still getting background lighting. (Fujicolor print films don't color
shift when overexposed). If you use highlight recovery the skintones, which
are supposed to be skintonish turn grey or cyan.
I also shoot the occasional concert and when you have stage access you get
the lights directly in the framing. With digital you get that nasty halo and
color shift in the lights and they turn into burned out blobs of bilge. With
film, the lights usually maintain their colors (red lights stay red, not
turn to white with yellow halos surrounded by a field of reddish pink). It
gets really ugly when the stage is fogged--you never know what color the fog
is going to turn into.
The contrast under stage lighting is so extreme, but film's shoulder and toe
really give you much more usable color and image. I'd have no qualms
shooting CH's Disney production with Fuji Superia 800 and if you underexpose
by one stop (effectively giving you ISO 1600), you still end up with a
decent image as long as you don't try to boost the shadows too much.
Most digital (and some films) really choke on Rosco #382 Congo Blue. To the
human eye, the scene is a deep purple, but to the camera it gets assigned a
straight blue. As mentioned in another thread it's because of the blue-red
portions. In the case of Congo Blue there is a double-peak in the
transmitted wavelengths. First wavelength is centers on about 450nm (blue)
and the second one is essentially an IR pass filter with the visible
spectrum cut occuring at 700nm and is completely nulled by almost 680nm. The
problem here is that the digital camera's IR cut filter is kicking it about
the same point and they essentially null each other out and the camera ends
up not seeing the red light. This is one area where neally all Kodak
sensored cameras have a slight advantage because the IR cut filter occurs at
a longer wavelength than most. Downside is the occasional magenta black
(which the E-1 is so blessed with too, it's not just the Leica M8).
Oh, and that deep midnight blue or indigo of gelled stage lights? To match
that for the human eye with LED lighting, you use a combination of blue and
red LEDs with the red LED's very much in the visible spectrum. Of course,
things can get really ugly if any lights use dichroic filters--which are
common in intelligent lights.
There are deep indigo gels that produce a true purple instead of the
blue-red dual-wavelengths. These gels have the distinct advantage of going
near-UV and will actually cause some florescing to occur, just like UV
lights. Most digital cameras will turn this to a straight blue with no green
or red detectors triggered. The sensor in the E-1, however, has the red
detectors sensitive to the very short purple wavelengths, so an E-1 will
actually capture a true purple as purple, albiet with a massive decrease in
sensitivity. It captures the purple, but about 2-3 stops low.
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/
|