Gaudy and Radioactive Color Rendition in Lightroom 4.1

  • 1
  • Problem
  • Updated 6 years ago
Do not like the Develop module at all. Version 4.1 RC2 just seems to over-saturate from the get-go, especially if there is skin tones involved. But even on landscapes, I am finding myself REDUCING VIBRANCE AND/OR SATURATION on RAW images. If I change the process version on an image I've already worked on, and set the sliders to zero, and then begin to re-work the image, the result is always gaudier than my version from Lr 3.6. I cannot believe that I have to reduce Vibrance to get acceptable images....not good images, barely acceptable images. If I go to process version 2012, the whole look of my photos, old and new, will change. And not for the better...at least I haven't yet figure out how to make 'em better. Wow.
Photo of Kurt Kramer

Kurt Kramer

  • 17 Posts
  • 0 Reply Likes

Posted 6 years ago

  • 1
Photo of Lee Jay

Lee Jay

  • 990 Posts
  • 134 Reply Likes
What profile are you using? The defaults for PV2010 and PV2012 have nearly identical looks with the same profile.
Photo of Dorin Nicolaescu-Musteață

Dorin Nicolaescu-Musteață, Champion

  • 703 Posts
  • 38 Reply Likes
Are you talking about PV2012 only? Do you a dramatic change in saturation between LR3 and LR4/PV2010?

When you say "Version 4.1 RC2 just seems to over-saturate", are the sliders in the Basic panel set to 0?
Photo of Kurt Kramer

Kurt Kramer

  • 17 Posts
  • 0 Reply Likes
Yes, process version 2012. Yes, I am making a virtual copy of images I "developed" in version 3.6. Then I update to process-2012 and set all sliders to zero and Tone Curve to Linear. Then I start "developing" in Ver 4 to see if I can improve on the image-work I did in Lr 3.6. I went to Kelby Training's Lr4 seminar here on Monday, so I know how the sliders work.
Lee Jay, what do you mean "profile"? The Camera Calibration profile?
Photo of Lee Jay

Lee Jay

  • 990 Posts
  • 134 Reply Likes
Yes. Was wondering if you had selected "landscape" or something by accident.
Photo of Eric Chan

Eric Chan, Camera Raw Engineer

  • 617 Posts
  • 121 Reply Likes
Would be helpful if you could provide examples.
Photo of Kurt Kramer

Kurt Kramer

  • 17 Posts
  • 0 Reply Likes
Hi Eric, You've made some good printer/paper profiles for me over the years. Thank you for that. And thank you for requesting examples. In what form do you want them and how do I get them to you? They are RAW files for the original model of the Canon 5D.
Photo of Kurt Kramer

Kurt Kramer

  • 17 Posts
  • 0 Reply Likes
Now I see the Add Image icon in the lower right. Here goes.

#1 - This is the image as I "developed" it originally in Lightroom 3.X, i.e. PV2010:


#2 - This is the image immediately after updating a virtual copy to PV2012:

Not too different.

#3 - Then my question was "could I make a better image by "developing this from scratch in the Lightroom 4.1?" So, I made virtual copy from #2 and, other than white balance (5500/10 daylight), I zeroed all of the Develop sliders and set the tone curve back to linear. I was surprised that it came out looking like this:


Without looking at the slider-settings that Lightroom set when converting the image from PV2010 to PV2012, I found it nigh impossible to get an acceptable image in this manner. I am finding that I seem to be able to get acceptable output when I import new images into LR4.1 and the new process version is applied from the outset. But if I update an image that I previously worked on in 3.X and try to start anew on it, forget it. Gaudy and Radioactive. It just seems that if I do update an older image (an I am now not seeing any reason to do that), I should only do minor tweaks from the values Lightroom assigns in the PV2012-conversion.

Any comments.
Photo of Dorin Nicolaescu-Musteață

Dorin Nicolaescu-Musteață, Champion

  • 703 Posts
  • 38 Reply Likes
What color profiles are used in cases 1, 2 and 3?

Is there any chance you modified your develop develop to include a crazu Contrast or Tone curve setting?
Photo of Kurt Kramer

Kurt Kramer

  • 17 Posts
  • 0 Reply Likes
They are all using ACR 3.3. I converted the original adjusted image to PV2012. To try something different, I then selected Adobe Standard. I went to a recent NAPP seminar and presenter Matt Kosklowski recommended that as his default starting point. That change moved the image more toward gaudy and radioactive. Look how bright red his cheeks became, under his eyes, as compared to #1 above, the adjusted in PV2010 and to the almost identical #2, converted to PV2012 without further adjustment.
#4 - Same as #2, with only the profile changed from ACR3.3 to Adobe Standard.
Photo of TK

TK

  • 531 Posts
  • 110 Reply Likes
Kurt, note that "0" slider settings in PV 2010 actually mean (close to) neutral RAW development, whereas "0" slider settings in PV 2012 mean "whatever Adobe believes is a good starting point for your RAW development".

Also note that if you see a linear tone curve in PV 2012, it means that the tone curve that is really applied corresponds to the PV 2010 medium-contrast curve. The tone curve looks linear, but it is not. I'd say you are being lied to, but of course one could argue that tone curve panel only shows (further) relative changes and that it is absolutely necessary to have a non-linear tone curve somewhere in the pipeline, etc. pp. I do not buy these arguments and would much prefer to be shown what is really going on (and would prefer not to be forced to create custom camera profiles in order to avoid arbitrary defaults), but apparently my viewpoint is carried by a minority only.

If you have used the (non-zero) Adobe developments defaults for brightness and contrast in PV 2010 then using "0" settings for PV 2012 produces something equivalent. If you have used "0" settings in PV 2010, you now need to go negative and apply an inverse S-curve in the tone curve panel (or use a custom camera profile) in PV 2012.

I am not sure whether this is of any help to you and/or related to your problem, but you may find the information useful. All the best for finding the root cause for your problem.

P.S.: I believe it was fundamentally wrong to go the "0 = some arbitrary default (rather than neutral)" route, but, again, apparently I belong to a minority.