(1) an AbsoluteDifference comparison of the two images posted above show a dynamic range change of approx 23% which when expanded to full range will show the color changes as well... looks like a good application of Remap... especially if you obtained preference results.
(2) a suggestion. Like ColorCorrect, Remap can be used with anchoring 'color pairs'. Perhaps a 7th color pair could be used as well as the previous 6 [changing the 50% tone R,G,B,C,Y,M saturations] in the vicinity of the right side clouds and even perhaps de-saturating the Remapped color [changing HSV,S to 0] so that the clouds show up a bit more...
3782966115_c480a471ee-1.JPG (41.83 KiB) Viewed 6958 times
I'm not sure how hard it is to implement, but would others find it useful to have 'Show Affected Region' switch in Remap, similarly to Colour Correction. I find it invaluable in CC for finding inadvertent discontinuities.
I also have a question. In CC there is 'pegging' analogy: if there is an unmoved control point, it protects near-colours from changing. Would choosing a colour pair and making them the same (with some large chroma and brightness radius) accomplish the same thing? I looks like it, but I can't quite convince myself.
The last question is about tightness and chroma and brightness radii. Other than the radii being individually adjustable for each colour pair, and the tightness being global, is the idea the same? Would anyone have an example when and why one would use one adjustment over the other?
A Show Affected Pixels feature would be great for Color Remap.
My take on Tightness is that it expands/contracts the Hue, Saturation, and Brightness of the colors to be changed just as the Chroma/Brightness radii expands/contracts the replace colors. Presumably, we would use it when we want to change more colors in the image than those determined by the default probe. If we had a Show Affected Pixels feature, relaxing Tightness would turn more pixels on in the Show display.
Color Remap does peg unchanged color pairs. The way I experimented with it was to make a radical change that spread widely across the image, and then start pegging parts of the image I did not want to change and watch the changed pixels contract in the Preview.
Regarding the Tightness parameter, I have been setting more color pairs if I want to expand the colors changed rather than adjust Tightness so I don't have a good example yet in practical use. Possibly more use will turn up some examples.
I was working on an example that should show the effect of relaxing Tightness, but first I was pegging areas of the image that I did not want to change and PWP blew up on about the 10th peg, so I gave up on that one. Color Correction handles the change I was aiming for w/o problem and with less pegs.
OK, here's an example where Tightness in Color Remap might be used. Take a sky with a fairly wide gradient from top to bottom. Select a point about midway with Remap and move the color to a more or less saturated blue as desired. Because Remap is precise, only a band across the gradient will be affected. Relax Tightness (and possibly adjust the radii controls) to apply the change to the whole sky.
Actually, I think Color Correction may tend to work better for wide-bandwith changes.
I'd wanted to bring out the clouds so remembered your post until I had sufficient time to figure it out. Fortunately, by then, there were other posts on this topic and I gleaned from them how to 'peg'.
Pegged pretty much on the clouds, desaturated the 'pegged' go to colour and bingo! Needs a little practice but definitely brought the clouds out. Magic! Weil, pretty smart, anyway!
The second image sampled a point above the mountain and it changed only a band of color (hope it still shows after the downsize). I set the Chroma/Brightness radii down to .02 to better show the change.
The third image shows the effect of relaxing Tightness to .18 and the change spreads across the sky.
Given Mike's example of a colour pair - the search colour being the bluish clouds and the replacement colour being more white: the range of search colour can be adjusted by 1) Chroma and brightness radii for this colour pair 2) the global tightness slider (for all colour pairs).
My question is: what's the advantage of using the global tightness slider if the 'tightness' can be adjusted more precisely for each colour pair with chroma and brightness raddi? Or is there a difference?
Here is an observation - I'm not sure if I understand why this happens:
In Colour Map, with a single colour pair, and with the Amount=100%, Tightness=1.00, Chroma Radius=0.00, and Brightness Radius=0.00, outer colour is still being replaced by the inner colour. Try for instance Mikes image - make outer colour any sky blue and the inner colour red (to see it well).