Raw & Post Processing, Printing > Colour Management
Custom Camera Profiling: A Look at the QPcard 203 book
mac_paolo:
Just tried to create six other profiles for my DSC-RX100.
Everything worked flawlessly under Qpcalibration software, but no DNG profile appeared under Lightroom after restarting it.
Can't understand why. ???
samueljohnchia:
--- Quote from: mac_paolo on February 18, 2013, 07:18:49 am ---Just tried to create six other profiles for my DSC-RX100.
Everything worked flawlessly under Qpcalibration software, but no DNG profile appeared under Lightroom after restarting it.
Can't understand why. ???
--- End quote ---
Have you seen this thread on the same issue?
mac_paolo:
--- Quote from: samueljohnchia on February 18, 2013, 07:33:00 am ---Have you seen this thread on the same issue?
--- End quote ---
I'll have to read it carefully, but my camera is definitely supported.
I created several other profiles in the last weeks and they all correctly appear under Lightroom.
So maybe that's not the case.
Ernst Dinkla:
--- Quote from: samueljohnchia on February 15, 2013, 07:24:18 pm ---
Also, I did not bother to calculate the dE for the patches. If you wish to do so you may. I may try that when I have a better comparison, but again the QPcard profile is just inaccurate enough that it's not funny. Interestingly it does not have a light blue sky patch, and for some reason that results in a too dark rendition of blues. Same for reds. Greens are too light. There is oversaturation of the reds, and when I have blue shadows, the blue channel often gets clipped when the QP profile is used. A check of the raw histogram indicates that the blue channel was not originally clipped for those captures. Yellows also get pushed towards green and renders with a neon-like hue. Not so for the ColorChecker. This is true even for fluorescent light tests that I did. The relationship of the lightness of each color hue is really out of proportion and it results in distorted rendering of images from my camera. The ColorChecker does much better in this regard.
--- End quote ---
For color rendering of landscapes, skies, foliage, flowers, insects, fishes, I would not rely much on cards like that. More chance with artificially colored subjects: cars, houses and even better furniture, paintings as originals in the studio with controlled light that is used both for creating the profile and the photo shoots. Pigment source has to have some relation to the pigments used in the subjects, color created in the subjects has to relate to the way color is made in the cards. Light preferably with a continuous spectrum. I have seen the QP card described here in that context and at that time there were positive reports.
--
Met vriendelijke groet, Ernst
http://www.pigment-print.com/spectralplots/spectrumviz_1.htm
December 2012, 500+ inkjet media white spectral plots.
samueljohnchia:
--- Quote from: Ernst Dinkla on February 18, 2013, 10:28:36 am ---For color rendering of landscapes, skies, foliage, flowers, insects, fishes, I would not rely much on cards like that.
--- End quote ---
Ernst, what would you rely on then?
--- Quote ---Light preferably with a continuous spectrum.
--- End quote ---
Actually, a developer at QP told me that a greater advantage to the QP comes when profiling for unusual lighting with spiky SPDs, like florescents/LEDs, as the ColorChecker is already excellent for daylight use. I was actually recommended to stick with that since I already owned one, if I only needed daylight profiles. But the offer came and I just had to know what the differences were for my workflow and setup.
--- Quote ---I have seen the QP card described here in that context and at that time there were positive reports.
--- End quote ---
Honestly, if I had a Passport and used the X-rite software only and then used a QP, I would have said the QP blows the ColorChecker out of the water. I just downloaded the X-rite software and tried it again with the mini ColorChecker I shoot with, and yeah, its pretty bad. But then there is the DNG PE and I am simply more pleased with its accuracy, and subjectively also the way it renders color. Also if I process into sRGB/Adobe RGB I would also prefer the QP over the CC, because of its propriety gamut compressing LUT, applied after all the processing steps in ACR, before conversion to output space. The DNG PE+CC approach would result in clipped colors.
Likewise if I was a studio photographer and I only had the QP and no ColorChecker, I would honestly be astounded with its quality - and wonder how I lived through the dark ages of generic camera profiles. That's how I felt when I built my first successful custom camera profile. It's all quite relative.
I did a test under fluorescent lighting, and while the QP was more accurate in some colors and tones, the ColorChecker won for others. Overall, I rated the CC better again because the QP rendered the scene too light and generally speaking the CC approach was more "accurate" overall, or perhaps was easier for me to render the scene to match the original. I'm sure others will disagree or agree depending on a variety of factors.
Again, something very troublesome in the midst of all this is if one is evaluating the performance of a profile applied with edits applied in ACR/Lightroom, it is hard to argue for "accuracy". Editing of any sort will throw off the profile to some degree. For instance, changing the tone curve would increase contrast and saturation where the curve has a gradient of more than 1, and reduces contrast and saturation where the curve is less than one, with some minor hue shifts either way. I use curves in all of my image edits (I really hope that a luminosity and chroma curve would come into a future version of ACR!) - there is no way a CC photo will look like the synthetic reference thereafter. But I also want to evaluate the QP against the CC together with edits (I'm not reproducing artwork), and I want something that looks good, and reasonably close to what I saw/felt/want from the image. The CC is just easier for me to get to the look I'm after.
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version