On 8/29/2011 5:04 PM, usher99@xxxxxxx wrote:
> ...
> I subscribe to this list and for no extra charge get PS pointers from
> PS Moose. :-)
Thanks -- I think.
> Appreciate the discussion and the link as it clarifies the issues.
> I did a forensics on the impression I had from some saved links I had
> files and it actually did refer to LR brightness.
> It was transferred in my long term portable storage to apply to PS
> which may or may not be correct. Gotta watch that.
Nomenclature in image manipulation is notoriously sloppy. Imagine if the
nomenclature in your profession were so poor!
> I suppose that if one crunches up the highlight area w/o moving the white
> point though not actually clipping to pure white, so much info is lost as to
> render the non-clipping a matter of semantics.
First, I hope you are referring to some control other than the PS Brightness
tool - It does indeed move the white point
down or black point up, irretrievably losing data to clipping.
Second, not necessarily so. If working in 16 bit or more, there are so many
numerical values available that compressing
toward the WP without clipping will make the image appear to have lost all that
data. However, a reverse process, such
as moving the midpoint in Levels or use of the Highlight tool will usually
restore the lost tonal detail, at least to
the eye.
Consider an 8 bit image. That's enough for printing or web viewing without
visual degradation. Convert to 16 bit. The
number of steps from 127 to 255 on the top was 128. In 16 bit, it's become
32,768 steps. You can push all the original
tonal information way up in the top of that range. Look at it on screen, and it
looks all white. Pull it back down, and
nothing need be lost.
If you start with a true 16 bit image*, data will indeed be lost, but probably
not noticeable to our human vision
systems, depending on how extreme the compression is.
It's different at the other end. Convert to 16 bit and the first step, 0 to 1,
becomes just 4 values. The next step,
from 2 to 3, becomes 16 values, and so on. Compress at the bottom end, and you
can't get it back.
Moose
* I haven't kept up, but early camera D/A converters were 12 bit, then 14 bit
showed up. I haven't heard of 16 bit. So
we probably never have true 16 bit images to work with. I've sort of assumed,
for the reasons above, that the cameras
map their output into the top of the 16 bit range, leaving the bottom empty.
--
_________________________________________________________________
Options: http://lists.thomasclausen.net/mailman/listinfo/olympus
Archives: http://lists.thomasclausen.net/mailman/private/olympus/
Themed Olympus Photo Exhibition: http://www.tope.nl/
|