RustyBug Online Upload & Sell: On
|
Moving on to the higher tonal value areas, I expect the bars to be neutral/reflecting the light that is falling on them to give me a clue for the light that is falling on our subject.
In review of them, I see some heavier Blue, with relatively equal amounts of Red & Green. So, I have made some adjustments in the Color Balance Layer to bring the RGB values all closer together. But, since my lower tonal values were already RGB neutral, I am going to target the upper range of tonal values via the highlight segment of Color Balance, hopefully not impacting the lower tones too much that I want to retain closer to neutral. However, in this case, the amount of adjustment I used in the upper values still made some impact to the lower values. So, I then applied some gradient "Blend If" to restrict things from the lower values.
Also, it should be noted that depending on which side of the bar I'm checking, I'll get different numbers. Realizing that the light is falling from one side vs. the other, I'll give more credence to the side that reflects lighting most similar to that which is falling on my subject.
While I target <5 variance R,G,B for neutral, it isn't always achievable due to subject/lighting angles/mixed lighting colors and nearby surfaces reflecting color influence. But, it still gives a clue to which direction I might want to make adjustments ... inside <10 and I'm still closer than if it is off by >20. It is an iterative objective process that winds up with the final decisions being subjective. If I've gone through the process of considering low/mid/high neutral RGB's and I don't like it ... I'll put it where I want it, but at least I know which/what/why I'm moving things around. Often times, I find that where a cast did exist, the correction to it can render things a bit differently such that I pull the saturation down just a bit (which I may have also driven slightly upward via some USM) too.
If we look at the pant leg of the three versions OP, Karen, mine and check the numbers we can see how much variance there is between the three channels. None of them are perfectly neutral, but one has more blue than the others, and one has less green (i.e. more magenta) and one is a little more balanced. Is the more balanced one "correct" ... maybe not, but the point is that we get to put things where we want them. Going from a 35-40 point spread to a 10-15 point spread @ RGB values renders the area more neutral ... but that's assuming that we actually wanted them to be more neutral.
Imo, it's tough to get color where we want via global tools, which is why I use Color Balance Layer @ low,mid,high and "blend if" to selectively adjust. But going back to the earlier comment @ neutrals looked neutral ... that is also noting that most folks are going to try to make color changes via global methods WB. Imo, there is a point at which "close enough" is pretty good because once you globally push one way, you are also globally pulling the other way. However, if you are wanting push the saturation/vibrancy of your colors more ... the differences might start to show up a bit more.
In the end, where I want it, is where I want it. These are just tools that can help us get it wherever that might be, particularly when it can be challenging to asses color by eye alone. Most folks will not go through this kind of process and it is time consuming, so there is a time/place for using it or not. But, a "quick check" of your RGB values at high/mid/low can give you a clue to compare with what your eyes/brain are telling you. Then, you can decide how much you think it needs to be changed/corrected/adjusted for ... or not at all, so you can just get back to capturing excellent images.
HTH
|