Are accurate colours actually accurate?

I see it came from here. I'm trying it out but you can't save a profile unless you pay for the license.

https://forum.luminous​-landscape.com/index.ph​p?topic=118075.0

This does not seem to have any audio. Is this at my end?

https://www.youtube.co​m …3jWs5vTk&featur​e=emb_logo

No audio on my end either.

sssc wrote in post #19058653
No audio on my end either.

Thanks

I found some other old links to Luminous Landscape and did other searches. I watched an Adorama video about CCP hosted by Marc Walberg. I watched a lot of his flash videos. He even pointed out how much the blues changes. If you search about CCP saturation there lots of links about this. Like Kirk said perhaps that teal/vs blue thing could have been something along the chain to mess it up but CCP does saturate. Complaints about the blues saturating to the point of losing detail, etc. Of course you fine tune a profile but if you do does it still remain accurate? What is the point of CCP then. You can tune any custom profile.

Lots of references to pleasing vs accurate. Accurate may not be what people want to see or they think they remembered seeing. I'm glad I started as I purchased for CCP. I dusted it off so now I have another profile to choose from depending on the scene.

No audio - just a screen capture of some of the interface and the "basic" workflow. There is a lot going in LR Profile Designer, and there are three different licensing options (the DNG-profile-only option is 30 Euro). What is nice about the LR Profile Designer approach and the manual that describes it is that it exposes all of the options and controls that can be written into a profile, in one application, without hacking or using the command line (like in DCamProf, the command line application that Anders wrote that does all of the stuff that the Profile Designer does, without a GUI; or some of the hue untwisting command line tools from dcpTool: http://dcptool.sourcef​orge.net/Hue%20Twists.​html).

Anyway, if you are interested in profiling your device and you want explicit control over the profile, then you might consider LR Profile Designer, especially if you are only going to use DNG profiles (DCP).

Kirk

kirkt wrote in post #19058673
No audio - just a screen capture of some of the interface and the "basic" workflow. There is a lot going in LR Profile Designer, and there are three different licensing options (the DNG-profile-only option is 30 Euro). What is nice about the LR Profile Designer approach and the manual that describes it is that it exposes all of the options and controls that can be written into a profile, in one application, without hacking or using the command line (like in DCamProf, the command line application that Anders wrote that does all of the stuff that the Profile Designer does, without a GUI; or some of the hue untwisting command line tools from dcpTool: http://dcptool.sourcef​orge.net/Hue%20Twists.​html).

Anyway, if you are interested in profiling your device and you want explicit control over the profile, then you might consider LR Profile Designer, especially if you are only going to use DNG profiles (DCP).

Kirk

Thanks very much. It is interesting and the price is not bad. Going through it I noticed all the available options. Sometimes I wind up going back to things as you need to digest it. More so these days at my age.

digital paradise wrote in post #19058671
Lots of references to pleasing vs accurate. Accurate may not be what people want to see or they think they remembered seeing. I'm glad I started as I purchased for CCP. I dusted it off so now I have another profile to choose from depending on the scene.

It seems to me that the inherent purpose of a custom profile is to ensure ACCURACY...regardless of the lighting which predominates the scene when a shot is taken. OTOH, tt also strikes me that your past disdain for using CCP was that it oversaturated blues...a fundamental observation that the captured color portrayal is NOT matching the actual saturation level of the objects in the photo. IOW, it is not 'accurate' if it is shown more saturated than in reality (if viewing photo and original object at the same time!
We as photographers do not always shoot for 'accuracy'...the shooting of Velvia for its deeply saturated colors is an example of seeking that which is not inherently 'accurate'. But there are also times in which accurate portrayal is necessary...accurate dye colors portrayed for a textile manufacturer, or accurate bridesmaid gown color for a bride who chose that color for her wedding party.

Via POTN I know of pro portrait photographers who had set up custom profiles for daylight vs. studio lighting vs home interior lighting, and how were pleased with how much better the portrait portrayals were. One of them commented on POTN (a long time ago)

"If I just use the white balance card in the ColorChecker and don't apply a profile in post, I think I get really nice color and nice skin tones. Having said that, when I then apply the profile I find that blacks look black, blue and red and purple become more saturated and in general it 'seems' like it puts color where it should be. Whether it's accurate or not, I can't say. I just know that I always love watching what happens as the profile is applied. Many people use other white balance tools and they're getting great color so there's nothing that says you have to use this expensive fold-up doodad if you like the color you get without it."

So accurate skin tones were more essential than accurate capture of the sitter's clothing color. IOW, different strokes for different folks or Purposes of the photo!

Back about 5 years ago, I used a trial copy of a RAW conversion program on a set of wedding photos that I had taken and processed with Lightroom. Even without the use at all of CCP custom profiles, the trial software created noticeably 'more electric blues' which were dominant in the color theme of the wedding...I preferred the LR colors and the colors were more similar to those seen in the hired photographer's shots. I set the other software aside (having experimented with it for a number of other shoots), as its color pallette simply struck me too much as 'not real'.


Another, perhaps distantly related, thought just came to mind...
Studio flash units usually come with UV coated xenon tubes. But some less expensive brands do not offer UV coated tubes in oall their products. The non-UV-coated tubes can make fabric brighteners phosphoresce when struck by the UV content light.

digital paradise wrote in post #19058671
I found some other old links to Luminous Landscape and did other searches. I watched an Adorama video about CCP hosted by Marc Walberg. I watched a lot of his flash videos. He even pointed out how much the blues changes. If you search about CCP saturation there lots of links about this. Like Kirk said perhaps that teal/vs blue thing could have been something along the chain to mess it up but CCP does saturate. Complaints about the blues saturating to the point of losing detail, etc. Of course you fine tune a profile but if you do does it still remain accurate? What is the point of CCP then. You can tune any custom profile.

Lots of references to pleasing vs accurate. Accurate may not be what people want to see or they think they remembered seeing. I'm glad I started as I purchased for CCP. I dusted it off so now I have another profile to choose from depending on the scene.

One thing to consider when dealing with blues and cyans is how your working color space versus the output color space handles those colors. Even if your display cannot display the blue properly, you can visualize the detail in the channels that produce the blue by examining the grayscale channel representation in PS or similar in your working color space. When you start in a large working space and then convert to a smaller space, at some point you have to manage how out-of-gamut colors get remapped (for blue, the red channel usually gets crushed, for example). When channel clipping occurs in saturated, clipped regions with detail, a mess results and color and detail are obscured and shifted. You can see this in the monochrome channel, for example, to see where the clipping is occurring, which might be helpful in addressing the channel's limitations before conversion to the smaller space for output.

Kirk

Wilt wrote in post #19058718
It seems to me that the inherent purpose of a custom profile is to ensure ACCURACY...regardless of the lighting which predominates the scene when a shot is taken. OTOH, tt also strikes me that your past disdain for using CCP was that it oversaturated blues...a fundamental observation that the captured color portrayal is NOT matching the actual saturation level of the objects in the photo. IOW, it is not 'accurate' if it is shown more saturated than in reality (if viewing photo and original object at the same time!
We as photographers do not always shoot for 'accuracy'...the shooting of Velvia for its deeply saturated colors is an example of seeking that which is not inherently 'accurate'. But there are also times in which accurate portrayal is necessary...accurate dye colors portrayed for a textile manufacturer, or accurate bridesmaid gown color for a bride who chose that color for her wedding party.

Via POTN I know of pro portrait photographers who had set up custom profiles for daylight vs. studio lighting vs home interior lighting, and how were pleased with how much better the portrait portrayals were. One of them commented on POTN (a long time ago)

"If I just use the white balance card in the ColorChecker and don't apply a profile in post, I think I get really nice color and nice skin tones. Having said that, when I then apply the profile I find that blacks look black, blue and red and purple become more saturated and in general it 'seems' like it puts color where it should be. Whether it's accurate or not, I can't say. I just know that I always love watching what happens as the profile is applied. Many people use other white balance tools and they're getting great color so there's nothing that says you have to use this expensive fold-up doodad if you like the color you get without it."

So accurate skin tones were more essential than accurate capture of the sitter's clothing color. IOW, different strokes for different folks or Purposes of the photo!

Back about 5 years ago, I used a trial copy of a RAW conversion program on a set of wedding photos that I had taken and processed with Lightroom. Even without the use at all of CCP custom profiles, the trial software created noticeably 'more electric blues' which were dominant in the color theme of the wedding...I preferred the LR colors and the colors were more similar to those seen in the hired photographer's shots. I set the other software aside (having experimented with it for a number of other shoots), as its color pallette simply struck me too much as 'not real'.

Another, perhaps distantly related, thought just came to mind...
Studio flash units usually come with UV coated xenon tubes. But some less expensive brands do not offer UV coated tubes in oall their products. The non-UV-coated tubes can make fabric brighteners phosphoresce when struck by the UV content light.

Yes with colour we are dealing with different spectrums. Maybe that is what it is. Daylight and Tungsten being most common and that has changed. Fluorescent, etc. The nice thing about it is you can create one daylight profile which I have done and use in sun, cloud and shade. You just need to adjust the WB.

That is a good quote. Even with the teal sweater looking blue I said the skin tones looked good. I actually found a thread I posted at my favourite flash site when I was learning about that subject preparing for my first wedding. I found a few blue items and you can see the sweater on my wife's arm. She won't let me take pictures of her but offered her hand

http://neilvn.com …olor-checker-pro-passport

It is interesting how we see colour yet the models we developed using science look different. Even when I tested C1 a few weeks ago it, DXO and LR were fairly close. Are they all weighted towards pleasing the general public? Then if you need then next level you profile.

Off topic when I was doing a little research for my presentations they figured children lived in more brilliant/saturated world because their eyes were young. Not sure it that was more related to everything being new and exciting. It does make sense.

Example of blue below. The original raw file was adjusted in ACR and opened in PS with a conversion into ProPhoto, with ProPhoto as the PS working space. In PS, I simply converted to sRGB for the side-by-side comparison. While the displayed images in full color may not look too different, inspection of the channels reveals clipping and loss of detail caused by converting from the large space into the smaller space (here, the RED channel, or the channel with the lowest exposure in the original image).

For sRGB display on the web, this may not be a big deal, but for printing, this may wreak havoc. Also, note that the lack of a difference in the images displayed in PS may also be the result of my display not being able to fully render the ProPhoto blue "accurately." This was displayed on an Eizo CG246.

Kirk

And, when you examine the gamut of the image in the context of various color spaces you can see how the blue can become more and more of a challenge to manage. Shown in the diagram is the typical 2D Yxy gamut diagram with the image color in the ProPhoto rendering space depicted, as well as four color space gamuts (ProPhoto, Epson 3880 Premium Glossy, AdobeRGB, sRGB). The intense blues that ProPhoto can handle with aplomb get more and more limited by the smaller color spaces and thus, need to be mapped somehow into the smaller gamut. This mapping may cause problems with colors like these blues.

The point is, your profile might be ok, but your display or other element of your color management may not be able to render that "accurate" teal accurately, leading you to think that the profile is bad. Pinpointing the culprit would require analyzing the image color and your color workflow chain.

kirk

That is interesting.

And here is the same thing, but with my display profile (white) and sRGB superimposed on the image data in ProPhoto.

kirk

Many many years ago I was following the Digital Dog I think at LL He said to work in aRGB because you never knew what your printer could manage beyond the specs. Then covert to profile for printing. These days the odd time I open in ACR O'm on ProPhoto as well. LR only has one colour space.