1. Welcome to NikonCafe.com—a friendly Nikon camera & photography discussion forum!

    If you are thinking of buying a camera or need help with your photos, you will find our forum members full of advice! Click here to join for free!

Ron R method?

Discussion in 'Nikon Capture and View NX' started by Allan, Aug 4, 2009.

  1. When I first got a Nikon I had Capture 4 and used Ron's method to neutralize shadows.
    I am wondering if people who switched to NX more than a year ago still use Ron's method or do you simple do a double threshold and place a black point.
    If you switched away from Ron's method - why?
  2. I still often use Ron's method with Capture NX2. The effects are subtle, but sometimes they make a significant difference.
  3. How about a little explanation of "Neutralize Shadows" for those of us that have never used Capture 4. Is it similar to what "Shadow Recovery" does in CNX2?
  4. I have used Ron's method, and still do at times. This Old DPReview thread might shed some help, and a search here at the cafe will surface some other threads as well.

    For me, I generally try Double-Threshold first, then LCH (thanks to many here including Mike and Yamo for an education) and sometimes the Ron method. As others have noted, sometimes one method works better than others, I wish I had the smarts to understand exactly what is going on with each, but I'll happily just accept the result :wink:
  5. Maybe we are talking about different things. My approach, based on Ron's teachings, is to open the "Curves" dialog box, and adjust the left end-point of the R-G-B histograms individually, and readjust the gamma point to the center on each.

    Allan, what is the approach you are referring to?
  6. That is what I was referring to.
    Nikon seems to think that if you do the double threshiold, slide the left slider to the right till you get a black mar on the image, and then place a black point there you have neutralized the shadows. This doesn't make sense to me - it seems it would just make r-g-or b black.
  7. That has never made any sense to me either for the reason that you mention. Rich actually pointed that out first.

    My guess is that "neutralize" is a vague term that implies varying degrees of success. If using that method produces an image that is reasonably close to what we envision when we press the shutter release, all is well. In that case, the precise degree of accuracy when it comes to neutralization does not matter.
  8. Allan spoke for himself, but that is exactly the approach I take as well. In the "old" Capture, it was easier to see as you could expand the histogram to a larger size.

    I follow the info in Jason's eBook, setting both the Black and White points this way.

    I agree, Mike, all these terms are too darned vague. I think if you read the thread in General regarding Thom Hogan's comment about "rules", this all fits, there just isn't "one way".

    I find, as I noted above, that at times I have to work with an image, and I have had all of these methods "fail" at times, which could easily be as much me doing it wrong, as to the method itself. I'd really like to find some understandable explanation for exactly what is going on with each of these, but so far I haven't, so I'll just live with the results.
  9. Bill - do you do the RR thing and then do a black point also? I was asking because I thought people would do one of the other - I still do RR's thing, not the blackpoint.
  10. Yeah, but if I do both the image usually sucks rocks, even more than normal :biggrin::biggrin:

    No, my normal workflow is to try the Double Threshold setting both Black and White points first. I find that if there is not a well-defined Black or White in the image, I can very easily get some very VERY ugly color-casts. If that is the case then I'll try one of the other methods, usually one of the three yields results I am happy with. If none of them work, them I'm pretty sure that the image just sucks, for real :wink:

    To be completely honest, I have yet to convince myself that overall any of the techniques is "better" than the others in all cases. A lot of it is what you get used to, and how easy it is for you to do. For example, Mike IIRC uses the LCH method, and does it very well. He has taken a lot of time, along with others, to walk me through this, but I still struggle with it. For me both the RR and Double Threshold methods just "clicked", so it might be as much my familiarity that makes it "better" for me than anything else.

    Hope that makes some sense.
  11. Bill - point me towards threads on LCH method.
    Last edited by a moderator: Aug 5, 2009
  12. I think you mean "LCH" and I am interested in this too. I do use the LCH editor sometimes to change the intensity of a particular tone, but I don't know what Bill is referring to.
  13. jfenton


    Jan 26, 2005
    Haverhill, MA
    I don't anymore.

    I actually do the majority of my color work in the SELECTIVE COLOR tool in CS3 after conversion in NX2
  14. Gaius


    Jul 1, 2008
    Everywhere along the grayscale continuum from black to white is considered neutral. I don’t know anything about R-R’s method, but if he is bringing the BPs of the individual channels to meet the left edges of their histograms, than he is forcing the darkest pixels in the image to black which is neutral. Moving the gamma slider back to the middle is simply limiting the effect of the curves to the lower half of the tonal range.

    Neutralizing the endpoints of an image is a time-honored first step for removing color casts.

    Setting a Black Control Point is going to do the same thing with the individual R, G, & B channels for the shadows, but I don’t know what NX does with the rest of the curve (if anything) to limit the effect.
  15. This is One of Them. If you do a search for LCH Mike Buckley you will find more. Also, check LCH and Holmes and LCH and Yamo.

    Hope this helps.
    Last edited by a moderator: Mar 15, 2017
  16. Bill, cut that out. You're making me blush. :biggrin:

    I fear that at least a couple of posts in this thread are giving the impression that the LCH Editor is good for correcting the color cast. Just the opposite: The primary advantage of adjusting levels and curves built into the LCH Editor is that, as Jason Odell puts it, the LCH Editor purposely "separates out luminance (brightness) information from chrominance (color) information. Levels adjustments in the LCH Editor will not affect color cast at all."

    Instead of using the Double Threshold Tool, I use Shift+S and Shift+H to determine the point at which shadows and highlights, respectively, will be lost. My experience after thorough anecdotal testing is that both methods identify the same darkest and lightest parts of the image. As I make adjustments using the LCH Editor, I keep and eye on the lost shadows and highlights to make sure they don't get too lost, so to speak. I also keep an eye on the large histogram. If you're getting the idea that I don't do a lot of batch processing, you're right. :biggrin:

    As Bill suggests, I know the LCH Editor works because it works for me. If it doesn't work for you after thoroughly studying it, use the other tools.

    It is a very intuitive one-stop-shop for me that controls brightness, contrast, saturation and hue (though I rarely have use for controlling hue). It does all of that both globally and locally. Most people seem to think of the LCH Editor as a tool for making global adjustments, but TomTom taught me to always consider using it also for local adjustments.

    To be frank, I don't know enough about the technical distinction between color cast, contrast, dynamic range, and where the three of them overlap and interact to affect color cast. All I know is that they do, at least to my eye.

    My style of photography rarely involves the type of photograph that I usually see used to demonstrate correction of color cast. When I expect the color cast to be tricky or particularly critical, I'll solve the whole issue using a grey card when that's practical. Otherwise, I plan on sending the image to Bill and asking him to fix it for me. :biggrin:
    Last edited by a moderator: Aug 5, 2009
  17. yamo


    Jun 28, 2007
    Santa Cruz, CA

    Greetings. The "RR method" as I've seen it described is IMO a bit odd... because it seems to me that either it differs little from the double threshold method or you end up changing the colors and color relationships of the whole image based on the near black and near white colors in your image (what sense does that make?).

    Recall that what these methods do is to set contrast. Since for digital color there are only 256 shades of gray (one might also say 256 luminance levels), curves or setting black point/white point adjusts the mapping of the luminance levels from the original image to an output image.

    When the curve is a diagonal line, it means that input luminance level is equal to the ouput level. In curves (for instance) when you move the black point slider, the input box increases in number the more you move it to the left. Say you move it to 10. What this means is any point in the original image of luminance value 10 will be output in the modified image as 0 (or black).

    But more importantly it means that the output image will display more contrast as what was once a slightly gray to white range will be output as a full black to white range.

    To get a feel for this slide the endpoints in the curves graph along the bottom and top of the box and along the sides of the box also to make flatter and steeper linear "curves". You will see the effect on contrast.

    So setting black point and white point is a contrast adjustment.

    Setting individual color channels effects the color channel contrast in just the same way, but also changes color relationships throughout the image. This may in some cases look better, but also may add strange localized color casts.

    The concept of neutralizing shadows I don't get at all.


  18. I always keep two paragraphs posted six months ago about curves emblazoned in my mind when adjusting curves. I mention them now because it's not possible to change the black and white points without affecting the curves. The two paragraphs definitely support everything Yamo mentions in his previous post:

    "An s-curve causes the tones to be distributed unevenly... where the curve is steeper (the middle of the s) there are fewer tonal steps between darker and lighter values. Consequently, this region of the tonal spectrum from dark to light has more contrast. Importantly, the parts of the curve which are less steep (the ends of the s) these parts of the tonal spectrum have less contrast.

    For contrast manipulations, you want to place the middle of the s curve (the steeper part) in the tonal range where you want more contrast and understand parts that represent the ends of the s curve will have less contrast."

    Maybe the reason those two paragraphs support everything Yamo says is because...he wrote them. :biggrin: For me, they are probably the two most important paragraphs I have ever read in the forum.
  19. yamo


    Jun 28, 2007
    Santa Cruz, CA

    Greetings. Gosh. Thanks. I'm glad the description worked for you. I remember working on it for a while because I think curves is the most useful (yet, most understood) post-processing tool.

    By the way, getting back to the OP, there are automatic versions of both the double threshold method and the so-called RR method in the Photoshop curves options (the button on the right side of the curves control box Options...). This is using PS CS3 (Sorry I don't use NX2 for curves).

    "Options..." brings up the "Auto Color Correction Options".

    1. Start by setting both Shadows Clip: and Highlights Clip: to 0.0% (you might want to check Save as defaults).
    2. Once the clipping amount is set:

    • "Enhance Monochromatic Contrast" is (more or less) double threshold
    • "Enhance Per Channel Contrast" is (more or less) the RR method
    Click between the two and watch the curves graph and the image to see the differences.


  20. Thanks for all the comments. The thread went further than I had thought it would. I think I will stick with what I am doing for now.
    Jim, I like to do more than you do in NX because at times I go back to the original raw file and I can see what I have done. I suppose I could create a photoshop file with all layers saved but I have never done that and do not want to due to the size of the files.
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.