Update on the vanishing blue.

[xpatUSA]

I am looking for the cause of the issue rather the outcome. Is the image saturated? Yes. Only two of the channels are contributing the the colours as the third channel has a zero value.

That is fine, but I am more interested in what happened to cause this. I suspect that will be more useful to Brian than simply suggesting that the issue is saturation.

My post #4 in this thread was a lot more than "simply suggesting" and Brian's response #5 tells me that he got it.

Apparently you did not, sad to say ...

[Manfred M]

Apparently you did not, sad to say ...

That is definitely an unnecessary dig Ted.

When I look at your explanation of the colours in your file, that is fine, but I'm not convinced the same thing has happened with Brian's image. Your white balance and exposure look fine, but I'm not convinced these elements are optimal in his image.

[xpatUSA]

That is definitely an unnecessary dig Ted.

Sorry, Manfred. You are quite right.

When I look at your explanation of the colours in your file, that is fine, but I'm not convinced the same thing has happened with Brian's image. Your white balance and exposure look fine, but I'm not convinced these elements are optimal in his image.

You see, the only way you could have assessed my images is that they contain elements other than the subject; just like Brian's other shots (lizards, etc) did not show the issue. On the other hand, Brian's image frame is full of the flower and nothing but the flower; that makes it harder to assess and that indeed is why the histogram looks like it does with next to no blues and that is why a slight reduction in saturation brings those blues bouncing back.

[JBW]

Brian - getting back on topic a bit.

Are the images that you posted straight out of camera and have little to no modifications in Capture One? Having a look at an image that has not been modified in any way would be useful.

Ok the images are not SOOC. But the subjects on either side of this flower show very normal color representation. When I take this shot and turn it into B&W is shows lots of blue.

This is a seriously saturated flower. The petals are deep and rich and the yellow, even in the shade hurts your eyes.

[george013]

...... When I take this shot and turn it into B&W is shows lots of blue.

This is a seriously saturated flower. The petals are deep and rich and the yellow, even in the shade hurts your eyes.

That's normal. Black,white or gray have equal values of R,G and B.

I still don't understand that problem with no blue. I must miss something.

George

[xpatUSA]

This is a seriously saturated flower. The petals are deep and rich and the yellow, even in the shade hurts your eyes.

Even so Brian your sensor itself should capture the petal colors irrespective of their color saturation - well, chromaticity, to be more precise. See figs. 5b and 6b here:

http://kronometric.org/phot/gamut/Ca...ysisGamuts.pdf

Have you tried looking at the flower in the ProPhoto working color space (aka Kodak ROMM) yet?

It sounds like the flower might be close to the edge of human vision at around 570-580nm, see also real world color gamut below.



Still, I would not expect gamut-clipping in the raw data and certainly not in ProPhoto. Others may not agree with that.

[JBW]

That's normal. Black,white or gray have equal values of R,G and B.

I still don't understand that problem with no blue. I must miss something.

George

There may be no problem but it ios the only subject that I have shot in 5 years and 50,000 clicks that has produced this result

[xpatUSA]

I still don't understand that problem with no blue. I must miss something.

George

What you miss is the fact that the subject (flower head) fills the entire frame. Therefore, the histogram shows only that part of the scene and will not look the same as that for a normal shot which has a background and/or other subjects. So if, after conversion from raw to RGB, there is a lack of blue in the petals - the histogram will show exactly that.

This is not rocket science - does nobody here have an app where they can select part of an image and show the histogram of the selection as opposed to that of the whole image ?? And, if so, would they not expect that the histogram of a selection to be different than that of the whole image ??

In Brian's subject image, cropping out everything but the petals has effectively made a SELECTION from the scene; and the histogram reflects that.

Does anybody get that?

[xpatUSA]

There may be no problem but it is the only subject that I have shot in 5 years and 50,000 clicks that has produced this result

Hello again, Brian.

If you were to step outside and shoot another evenly-lit flower head so that it fills the whole frame, I would guarantee a similar result, giving you 2 out of 50,001 clicks ...

[george013]

What you miss is the fact that the subject (flower head) fills the entire frame. Therefore, the histogram shows only that part of the scene and will not look the same as that for a normal shot which has a background and/or other subjects. So if, after conversion from raw to RGB, there is a lack of blue in the petals - the histogram will show exactly that.

HTH.

I do understand that.
If Brian states the color are as they are in real that would be normal. As stated in a very simplified post there're over 65000 combinations without blue in 8 bits.
And if somebody states there should be blue in it then I wonder where that's based on. I'm still missing something.

Brian,
I reread your first post again. There's blue in your image. A little but nearly every pixel has some blue in it.
Yellow is a combination of red and green, additive colors. It goes from red-orange-yellow-?-green. I don't know the names. From Wiki

Your flowers are orange.

George

[Manfred M]

There may be no problem but it ios the only subject that I have shot in 5 years and 50,000 clicks that has produced this result

Brian - I rather doubt that this is correct. It's just that this was not something you had picked up on before.

[JBW]

Brian - I rather doubt that this is correct. It's just that this was not something you had picked up on before.

Obviously I've picked up some bad trumpian habits.

There may be no problem. But in 5 years of shooting and fifty give or take thousand clicks this is the first time I've noticed this lack of blue.

[ccphoto]

It would be infinite easier to draw reasonable conclusions if the original file were posted with no editing nor cropping. There are a lot of ways reflective light can affect natural saturated light. Building a zillion diagrams on color balance won't help on an already edited image.

[DanK]

George,

As stated in a very simplified post there're over 65000 combinations without blue in 8 bits.
And if somebody states there should be blue in it then I wonder where that's based on.

Nature. No one contested that it is mathematically possible to have colors in that space with no blue. What Ted explained is that this would rarely or never occur in nature, when photographing actual flowers.

I reread your first post again. There's blue in your image. A little but nearly every pixel has some blue in it.

That's not what I found when I opened the image in Photoshop. I found that most pixels I checked were B=0, and the few with blue were typically B <=3. Nothing like my own yellow flower photos, where it is hard to find pixels with B<35.

Chris,

It would be infinite easier to draw reasonable conclusions if the original file were posted with no editing

Exactly so. You beat me to it. I was in fact checking the thread to see if that had been done when your post arrived.

I don't know Capture 1, so I don't know how one would to this in Capture 1, but what we need is the most neutral rendering it will provide. It's presumably easy to set all adjustments to zero, but this will also depend on the initial rendering choice (or, at least, it does in ACT and LR, which I use).

If there is blue in the original capture, then it's clear that one should look in the editing to find the cause.

[Manfred M]

But in 5 years of shooting and fifty give or take thousand clicks this is the first time I've noticed this lack of blue.

I'm good with that statement!

[JBW]

I'm good with that statement!

and as requested: SOOC. In C1 the petals show no yellow and yellow shows no blue

[JBW]

It would be infinite easier to draw reasonable conclusions if the original file were posted with no editing nor cropping. There are a lot of ways reflective light can affect natural saturated light. Building a zillion diagrams on color balance won't help on an already edited image.

the SOOC (no cropping but I did shrink it down to 1 mb) is now posted.

[JBW]

George,



Nature. No one contested that it is mathematically possible to have colors in that space with no blue. What Ted explained is that this would rarely or never occur in nature, when photographing actual flowers.



That's not what I found when I opened the image in Photoshop. I found that most pixels I checked were B=0, and the few with blue were typically B <=3. Nothing like my own yellow flower photos, where it is hard to find pixels with B<35.

Chris,



Exactly so. You beat me to it. I was in fact checking the thread to see if that had been done when your post arrived.

I don't know Capture 1, so I don't know how one would to this in Capture 1, but what we need is the most neutral rendering it will provide. It's presumably easy to set all adjustments to zero, but this will also depend on the initial rendering choice (or, at least, it does in ACT and LR, which I use).

If there is blue in the original capture, then it's clear that one should look in the editing to find the cause.

The SOOC is posted. No yellow in the petals and no blue in the yellow.

[JBW]

Ted is absolutely correct. When I go to -20 saturation there is blue in the yellow.

[DanK]

How did you get the SOOC? Did you shoot raw, and if so, how did you produce the image you posted? I'm still trying to figure out where that degree of saturation was introduced.

BTW, although I would crop it a bit, I much prefer the uncropped to the original. The original shows a few fine details but is mostly featureless, out-of-focus yellow. The original is a much more interesting photo, IMHO.