At lunchtime I went through the local taco takeout (trots on the run?
Get your runs at the border? Chuck-a-Chimi?) and headed over to a park
to do a little B&W testing.
I tested three cameras: Olympus E-1, Minolta A1 and the Panasonic L1.
You think I like the number one?
On the E-1 and L1 the 50mm F1.4 was mounted and the aperture set to
F8. With an ISO of 100, the standard exposure for the sunny day, with
occasional cumulus clouds was F8 at 1/400. So I did a single
unfiltered shot at F8 and 1/400. Then I took exactly the same picture
with a green, yellow, orange and red filter at 1/400. Each filter
image was also taken at 1/200 and additionally the red taken at 1/100.
I repeated the test on the A1, but had to set its lens to F10 to
compensate for the true ISO 160 sensitivity of the sensor.
So, now I have a set of images from three cameras with one straight
image, and two (or three) images with each of the filters.
The end goal here is to come up with a final image of each color
filter setting. One derived through post-processing (color filtering
in the editor) and the other through converting the lens-filtered
image and monochroming it.
I'll get a round tuit sometime this evening, but do you have any
preferences as to how I go about this test? I will be doing all
editing in PWP as it supports 16-bit and has one of the better
color-to-monochrome converters around.
I'm thinking that I'll RAW convert the files straight with no color or
monochroming (which will result in green, yellow, orange and red
colored files) and save them as 16-bit tiffs. The tiff files would
then be opened in PWP and converted to monochrome. After converting to
monochrome, a linear curve adjustment will be applied to top-and-tail
the histogram so all images will be identical in contrast. Each file
will be resized to 800 width, but a detail section will be cropped to
provide a pixel-peeping oddessy for those wanting that. Files will be
converted to 8-bit and saved at 95% (2.3:1 or so compression) jpeg to
minimize jpeg artifacting. The full-color image will be converted to
monochrome using each of the same color filter settings (green,
yellow, orange, red) and then the same curves adjustments applied.
I believe that the above procedure eliminates variabilities in RAW
converters and places all the onus on the editor, which in this case
will be used to create as close to identical files (for each color) as
possible. If I do the same procedure and adjust contrast to 0%-100%
the results should give us a good baseline.
To reduce variabilities in raw converters I'm thinking of converting
all of them with the DCRAW engine in PWP, but that does result in
other artifacts, so I'd prefer to use the converter which is optimized
for each of the cameras. ACR also artifacts these files, so I'm not
comfortable using that across the boards either.
Moose? CH? Thoughts or recommendations?
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/
|