What to do with that sweet sunset light?

ruttrutt Registered Users Posts: 6,511 Major grins
edited June 29, 2009 in Grad School
Here is an interesting problem. That sweet sunsut light sure makes dramatic colors and beautiful images. But it also introduces a warm cast. It's a pleasing enough cast alright, not like tungsten, but a cast nonetheless.

I want to show two different examples of this and show some different approaches to them. Then I am going to discuss a little color theory (psychology really.)

Let me start with an image of Ben's. He was kind enough to email me the raw version of this. Here is the image, converted with "As Shot" color followed by a little highlight/shadow and L curve work, and sharpening. All of the moves after conversion are color neutral. And the birds come out pretty yellow:

34115215-L.jpg

They are yellow because the light is yellow and the birds are white which reflects the yellow light. Using the eyedropper on the birds at conversion time yields such horrible blue resutls that it is clear that the birds really are reflecting yellow light in this image.

But to my eye they look too yellow. I know these are white birds and my eye wants them to be white, just the same way it wants a wedding dress under tungsten light to be white.

As an experiment, I use LAB curves to neutralize the white of the birds while retaining the other colors. Well, as long as I was using that B curve, I helped myself to a little more blue in the water, but mosty I was a good boy:

34115228-L.jpg

34112945-S.jpg

Now the birds are mostly white. Does this look more natural? More right?
What if we printed it?

Let's take another example. Here is a shot I took the other night, converted "As Shot":

34116780-L.jpg

I love the dramatic colors here, but the color sampler shows that the fleshtones are unnaturally yellow. Very much so. I took a simalar approach here as I did with Ben's image:

34116848-L.jpg

Of course this image has no obvious neutral point, so I just pulled up the B curve in the neighborhood of the yellow flesh until the color sampler gave plausible CMYK readings (M and Y in balance).

34116857-S.jpg

I said there was no obvious neutral point in this image. But what about the subject's grey hair? That should be pretty reliably neutral, no? So I did a second raw conversion and used the custom white balance tool on a strand of grey hair. Here is the result:

34116721-L.jpg

To me, the subject looks a lot better. Too bad about the sunset colors, though. I have a vivid memory of that sky being very colorful, more like in the first conversion.

So I tried a blend. We want more saturation in the sky and reflections in the water, but not in the face and hair. Fortunately in this image, we can make this happen with a blend-if which takes lighter colors from the As-Shot conversion and darker colors from the custom conversion:

34116953-L.jpg

34116970-M.jpg

DiscussionWell, that was fun, wan't it. But I think it touches a very interesting underlying issue. We all have a very complex psychological mechanism that processes what we see. The camera short circuits this and photographs have to resort to trickery of different sorts to fool our eyss into "seeing" something reasonably like what we see in the real world. That's why we need to worry about color balance in the first place. We see things we know are neutral AS neutral under almost all light. At least I do. White wedding dresses look white under tungsten, even though we have learned over and over again that they really are some shade of orange under this light. I see asphalt as neutral under street lights. And flesh is very special. Our brains bend over backwards to make it look right under all but the strangest light. I imagine this has some important survival advantages. We need to recognize healthy/sick indivituals, make ethnic distinctions, recognize friends, family, enemies even if the light changes. (I don't know this, I'm making it up, but it sounds good, doesn't it?)

On the other hand, there is no harm in seeing saturated sunsets and their reflections in blue water.

Anyway, I have found that people have very different reatiions to color balance under unusual light. There was a beautiful pano of Las Vegas at dusk that was measurably very warm. I corrected to neutralize some of the lights and the white parts of some signs. The photographyer complained, "That's not how it looked." I had a similar interaction with Andy about a picture of San Fancisco Bay he shot at night. It was pink which looked wrong to me. I moved it toward neutral. Andy hated it. "That's how it was," he said. I think that means that it was how it looked to him, or at least how he remembered it. But how would it have looked to me? I think my brain's AWB is particularly diligent with artificial light. It needs to see something neutral. It seems this isn't true of others.

I'm interested in opinions on this topic. I think it's very subjective...
If not now, when?
«1

Comments

  • ginger_55ginger_55 Registered Users Posts: 8,416 Major grins
    edited August 30, 2005
    I remember that incident very well. Not the photo, have no memory of what the photo looked like, but I do remember you, Rutt, kept saying something like, "I know San Francisco and it just is not a "pink" town". I found that to be humorous. I mean not laughing at you/Andy, just laughing at the personification, as I saw it, of a city.

    My interest in this subject lies in the intellectual isn't this fun type discussion. And the other part of me says, hey, wait a minute, I used to just upload some photos, fiddle here and there, send them to my kids who would go "wow" and stuff. Was fun. (MY kids no longer even mention my photos which they accept as wonderful, but probably a drain on their "wonderful photo" comments resources.)

    Now I am not only faced with the subject of who do you try to please, is it subjective, or is it objective (most people know I think it is subjective).

    But my real thought, before I feed the dogs, is when I come back with beach shots, like I did today, just birds, not as many as usual, but do I really want to do all that work?

    And if I don't want to do all that work in post, much, or often, .............

    Well, I see this blossoming into more than a full time job, it already is that. I am not a pro, though I do have a pro acct.

    If I had a boss or a client, I would have no argument whatsoever with that person getting the wedding dress in purple, if that is what the person wanted. But I chose not to go that route. I have refused assignments: not many, but some. I have not actively sought to profit from this "worK". If people are looking for me to get some prints from, I often hide.

    So on an intellectual level I would love to discuss it, but I don't want to do it, not often, and not unless for a special occasion.

    Uh, I usually like my colors warm. I am pretty consistent in that. But also, my eyes/mind can adjust to just about whatever. I have corrected casts in RAW, thought the correction looked terrible and 10 minutes later, thought the correction was just right.

    ginger, gotta feed these dogs.

    (I think I will start calling my photographs "proofs") Then I will hope for a financial windfall and send everything out to get fixed just to my specifications: warm. Smile, big grin.) And what color is your town? Charleston is pretty much pink.
    After all is said and done, it is the sweet tea.
  • bfjrbfjr Registered Users Posts: 10,980 Major grins
    edited August 30, 2005
    Excellent read and learning, Rutt
    Thank you have copied and printed it for reference.
    Right off the top of my brain here's what is says to me:

    I like the birds with the warmer look, but I took the shot and have a personal connection with it and that rules me out as an impartial judge. Your work on the shot is however how the birds really look in real life no matter the color of the light.

    However I like your last take on your portrait shot better then the warmer look. My brain says right away that the skin tones are just perfect in your last shot and much more natural looking then the original shot.

    As far as printing goes I would not have a clue. Have never personally printed any of my work.

    As you already said very subjective subject, with probably not one correct answer?? ne_nau.gif
    In either case thank you for your time I know that I have learned much as I hope others that read this will.
  • ginger_55ginger_55 Registered Users Posts: 8,416 Major grins
    edited August 30, 2005
    Yes, Rutt, I can say that I don't like all that yellow in a face, so I do like the last one best. (I do know, from Cletus, that there has to be more yellow than magenta, so I check for that, but too much yellow bothers me.)

    Also, this is what I do, and it isn't anything anyone else would do. I don't like the heavy color casts, usually, and I know my audience doesn't. So I find a white spot on my eternally white birds. If it isn't a white bird you should watch me panic!

    I know the temp and other thing when I started, the reading there. I then correct it as "told to" by the white on the bird. Then I back up, so I have mine somewhere in between. I live with that color for awhile. No one has complained re the color correction since I started doing it that way. It isn't scientific, but it is fairly consistent after awhile. Birds with a green cast at Magnolia, not sweet light, green cast, well, I correct them totally. If it is a sweet light thing, I correct just to show I am trying, then, as I said, I back up. So it is not "as shot" and it is not as daylight, either.

    ginger
    After all is said and done, it is the sweet tea.
  • ginger_55ginger_55 Registered Users Posts: 8,416 Major grins
    edited August 30, 2005
    Interesting thing is that I was very happy with Ben's original color, but now that I see the "corrected" color, I like that better.

    I do not like heavily saturated sunsets, I mean they make my teeth chatter. I think some things some people react to stronger than others. I do like your sunset up there, Rutt.

    g
    After all is said and done, it is the sweet tea.
  • HarrybHarryb Registered Users, Retired Mod Posts: 22,708 Major grins
    edited August 30, 2005
    I have to say that I like Ben's original version better. One of the reasons to shoot in the early morning light is that you get different colors and a different view.

    I shot a group of snowy egrets in the morning light and my captures and Ben's captures accurately reflect the scene and the lighting you get at that time of the day. To wash out those colors for a homogenized white doesn't work for me. Why shoot in sweet light so you can post process out the results of that lighting? You might as well shoot in the mid-day.
    Harry
    http://behret.smugmug.com/ NANPA member
    How many photographers does it take to change a light bulb? 50. One to change the bulb, and forty-nine to say, "I could have done that better!"
  • Mike LaneMike Lane Registered Users Posts: 7,106 Major grins
    edited August 30, 2005
    Harryb wrote:
    I have to say that I like Ben's original version better. One of the reasons to shoot in the early morning light is that you get different colors and a different view.

    I shot a group of snowy egrets in the morning light and my captures and Ben's captures accurately reflect the scene and the lighting you get at that time of the day. To wash out those colors for a homogenized white doesn't work for me. Why shoot in sweet light so you can post process out the results of that lighting? You might as well shoot in the mid-day.
    15524779-Ti.gif

    Now, if I were ever to shoot in a studio or do product shots, I'd be darned sure to make the colors as accurate as possible. I'd probably have a color meter AND a GM card just to be safe. But outside? Meh, I say embrace the light you've got.
    Y'all don't want to hear me, you just want to dance.

    http://photos.mikelanestudios.com/
  • ginger_55ginger_55 Registered Users Posts: 8,416 Major grins
    edited August 30, 2005
    The message that I am getting is that there is no one "right" way, especially if one is shooting for him/herself.

    ginger
    After all is said and done, it is the sweet tea.
  • ruttrutt Registered Users Posts: 6,511 Major grins
    edited August 30, 2005
    ginger_55 wrote:
    But my real thought, before I feed the dogs, is when I come back with beach shots, like I did today, just birds, not as many as usual, but do I really want to do all that work?

    And if I don't want to do all that work in post, much, or often, .............

    One you become adept with curves, especially LAB curves, it isn't much work.
    If not now, when?
  • ruttrutt Registered Users Posts: 6,511 Major grins
    edited August 30, 2005
    Harryb wrote:
    I have to say that I like Ben's original version better. One of the reasons to shoot in the early morning light is that you get different colors and a different view.

    I shot a group of snowy egrets in the morning light and my captures and Ben's captures accurately reflect the scene and the lighting you get at that time of the day. To wash out those colors for a homogenized white doesn't work for me. Why shoot in sweet light so you can post process out the results of that lighting? You might as well shoot in the mid-day.

    The two versions I did are exterme endpoints. Totally neutralizing the white of the birds represents one reality: the bird are in fact white, no matter what light they are reflecting. Using the "As Shot" colors represents another reality: probably as close to the truth of light as possible, near to what a spectrometer would read. Our brains are wired in a way that rejects that latter under at least some circumstances. AWB "As Shot" color under tungsten look wrong to everyone (to my best knowledge.) There seems to be near universal consensus about the portrait in the sunset. I think our brains are good at color balancing human fleshtones.

    I don't love the neutral birds in the sunset light. They are truly neutral; I've measured it. But they look a bit blue. The ones directly from the sensor also don't really look right to me. The birds don't look light enough. I want to see at least one point on the birds that is is really neutral and white. Otherwise they just look dirty to me, and that isn't really how they look in the "sweet light".

    Before I read your post I was thinking that a solution similar to the portrait would also work for the birds: layer the neutralized and yellow versions and use blend-if to get white in only the lightest highlights and smoth blending on the edges. Once this is set up, it's sort of a machine that allows basically any of a zillion options, not only by changing the opacity, but also by using "blend-if" to control where it happens. I tried and go this (only an example):

    34140114-L.jpg

    34140135-M.jpg

    I'm not sure where Ben's colors came from. No simple ACR conversion I tried ended up with the brilliant blue water he had. Of course, I could achieve this via curves and blending, but that's not the point. That color wasn't on the sensor; Ben did something to get it. Now, it should be clear that I'm not at all opposed to tweaking colors. In fact I think it's enevitiable if you want to end up with natural looking results. Without some work in post, your camra is basically a spectrometer and doesn't take all that wonderful processing power in our brains into account. Photography is illusion and our job is to make it work. I can furnish loads of proof for anyone who doesn't believe it.

    This all reminds me of the color of snow project which I undertook in the winter (http://www.dgrin.com/showthread.php?t=5586) The answer turned out to be subtle, but it was definitely worth figuring out for someone who loves to take pictures in the snow. I also love to take pictures in the dusk light. So I'm hoping to solve this puzzle.
    If not now, when?
  • HarrybHarryb Registered Users, Retired Mod Posts: 22,708 Major grins
    edited August 30, 2005
    rutt wrote:
    The two versions I did are exterme endpoints. Totally neutralizing the white of the birds represents one reality: the bird are in fact white, no matter what light they are reflecting. Using the "As Shot" colors represents another reality: probably as close to the truth of light as possible, near to what a spectrometer would read. Our brains are wired in a way that rejects that latter under at least some circumstances. AWB "As Shot" color under tungsten look wrong to everyone (to my best knowledge.) There seems to be near universal consensus about the portrait in the sunset. I think our brains are good at color balancing human fleshtones.

    I don't love the neutral birds in the sunset light. They are truly neutral; I've measured it. But they look a bit blue. The ones directly from the sensor also don't really look right to me. The birds don't look light enough. I want to see at least one point on the birds that is is really neutral and white. Otherwise they just look dirty to me, and that isn't really how they look in the "sweet light".

    Before I read your post I was thinking that a solution similar to the portrait would also work for the birds: layer the neutralized and yellow versions and use blend-if to get white in only the lightest highlights and smoth blending on the edges. Once this is set up, it's sort of a machine that allows basically any of a zillion options, not only by changing the opacity, but also by using "blend-if" to control where it happens. I tried and go this (only an example):

    I'm not sure where Ben's colors came from. No simple ACR conversion I tried ended up with the brilliant blue water he had. Of course, I could achieve this via curves and blending, but that's not the point. That color wasn't on the sensor; Ben did something to get it. Now, it should be clear that I'm not at all opposed to tweaking colors. In fact I think it's enevitiable if you want to end up with natural looking results. Without some work in post, your camra is basically a spectrometer and doesn't take all that wonderful processing power in our brains into account. Photography is illusion and our job is to make it work. I can furnish loads of proof for anyone who doesn't believe it.

    This all reminds me of the color of snow project which I undertook in the winter (http://www.dgrin.com/showthread.php?t=5586) The answer turned out to be subtle, but it was definitely worth figuring out for someone who loves to take pictures in the snow. I also love to take pictures in the dusk light. So I'm hoping to solve this puzzle.

    To a large extent, to me anyhow, you are trying to fix something that isn't broken. One of the joys of shooting in "sweet light" are the variations of color that comes from shooting in that light. He's a Great Blue Heron at sunrise at Ft. Meyer
    37166255.jpg

    same time and location, a Little Blue Heron
    37166258.jpg

    Now I know the birds don't look like that normally but thats they way they were when in that light. I could easily adjust the wb and color and have closer approximations of the way they "really" are and I could have myself a couple of hours of sleep instead of getting up before sunrise.

    You are trying to find a solution to something that doesn't need solving. Also your attempt at a solution intensified a weakness in Ben's original shot and that is the blown out areas. In the egret in the upper left there are blown out areas on the head, shoulder, lower chest and outer (left) wing. In the original shot that are very minor distractions but in your last version the blown out areas are more apparent and a bit larger.
    Harry
    http://behret.smugmug.com/ NANPA member
    How many photographers does it take to change a light bulb? 50. One to change the bulb, and forty-nine to say, "I could have done that better!"
  • jwearjwear Registered Users Posts: 8,005 Major grins
    edited August 30, 2005
    for what it is worth
    well Rutt that is why you do not shoot wildlife and why we talk about light ,you want life in black and white .Not me !! I want captures and moments not a photo shop trick, just like Harry turning a Night heron into a goat a very nice PS trick .but not a capture and not a moment .Sometimes you want to adj. a photo but to make it into what it is not.That is what I call studio portrait . You take a person with a ego and they want to look like what they are not and that is a good thing .We and I will call myself a wildlife photographer[ not a great one but I are one] do not take a picture of anything that has an ego .We --well me -- want a capture a moment not something we can go to PS and turn into a portrait . The color we get from light we like, why do you not take a sunset and turn the sky blue in PS ne_nau.gif .Just another view on the subject and I know how this will come across
    Jeff W

    “PHOTOGRAPHY IS THE ‘JAZZ’ FOR THE EYES…”

    http://jwear.smugmug.com/
  • ginger_55ginger_55 Registered Users Posts: 8,416 Major grins
    edited August 31, 2005
    Other interesting subjects on the same topic of what is "real":

    Blur, some people like it, some people don't. I like the whole object blurred, but it is late and this is what I found........
    28643830-L.jpg

    Backlight, our eyes see backlit objects much differently than the camera does.

    26590214-L.jpg


    Some of us might remember this photograph: in its two versions........

    26613097-L.jpg


    26592677-L.jpg


    I am sure there were more in my effort to get it right. Whichever the original photograph was, it was just reflected green light from all around. My eyes had adjusted............but which should have been used, that was the problem as sweet light came into the "picture", too.

    Always there was the green,

    25870295-M.jpg

    Then there was this, a landscape. It sure surprised me. I saw a white sail, etc. I was kind of disappointed with all that sweet light, but I used it.

    25783338-L.jpg


    It is an interesting subject. I could have made the boat look as I remembered it, but I finally chose not to.

    I would like to talk more about it, but I am really tired. I went to breech inlet today for a bit, two old ladies and birds on the Sullivan's Island side. Not exotic birds, though I didn't know what most of them were. The light was afternoon, so no color shifts to deal with. In fact I even shot at a lower ISO some of the time. Sorry, too tired to post the exif, will tomorrow if anyone wants me to. It was really what one would expect.

    Rutt, I appreciate your bringing this topic up, it makes me think. Better informed and having a firm basis for my artistic decisions, it makes things easier on my mind. Plus it is stimulating fun.

    Some things I like, or think I do, as I shoot them, other things I used to like as I shot them, like the green birds.............so I am not sure it is totally fixed. There is still artistic interpretation.

    Just thinking of the term, blown, I always thought there was a color of stark white. I found it pleasing, but I have learned to try to never show a pure white part of a bird. It is just not done and the term is blown. That was "fixed". A man who makes his living at nature photography. He is well known, and not just in the Carolinas. Imagine my surprise when I looked at his book with him and found all these white birds parts of which were clearly blown. He didn't know better, but he is respected, has a winter home in this area and a summer home in Vermont, earned nature photography money.

    It is an interesting subject, IMO.

    ginger
    After all is said and done, it is the sweet tea.
  • ruttrutt Registered Users Posts: 6,511 Major grins
    edited August 31, 2005
    Harry, you seem to be missing my point. I'm posing questions, not proposing answers. I'm interested in how we see photographs vs the reality they represent. It's very clear that our brains do all kind of complex things when we look at a scene in reality which it doesn't or can't do when it looks a photograph. For example, we see fantastic dynamic range which even the best film camera just can't capture. In this case, it's because we see differently than a camera does, mentally piecing the scene together from many "glimpses". Our pupils open and close as we glimpse different parts of the scene. In the end we see something much richer than the camera does. And yet, good photographers with good (digital) darkroom technique can sometimes produce results which capture some very extreme cases. (See http://www.dgrin.com/showthread.php?t=17377 for a discussion of this including questions about the technique Ansel Adams used to produce a famous image.)

    Color balance is another well documented case where our vision of photographs and in reality are very different. In fact our vision of the monitor vs prints is also very different. And different people seem to work somewhat differently in some cases. An incident I'll always remember occured soon after I joined dgrin about a year and a half ago. Lynn had a project to produce a calendar with pictures of women she knew. The pictures looked OK to her (and everybody else) until she tried to print them. Then they looked unacceptable. It took me a long time to get around to measuring the color values. When I did, I found that the fleshtones had inplausible values. (See http://www.dgrin.com/showthread.php?t=1077) Since that time, there have been many posts of images shot under tungsten without proper white balance or treatment in post. Sometimes the poster seems to know that something is wrong and sometimes not. Examples: http://www.dgrin.com/showthread.php?t=13533 http://www.dgrin.com/showthread.php?t=9943 There are lots more. Search for "tungsten" to get just the tip of the iceberg.

    I encourage everyone to read Chapter 1, "Preflight: How Can We Make This Image Better", from Dan Margulis' classic book Professional Photoshop. Here are a few quotations:
    The key to having viewers like our images, it turns out, is to understand how cameras and human eyes differ.
    ...That the camera saw blue is very interesting, but you or I would have seen neutrality, because we are human beings and that's how human beings see such scenes. Our vision is self-calibrating: we always see the ambient lighting as neutral, never blue or any other color.
    Irritatingly, we evaluate colors in context, not absolutely. The presence of a background color provokes the effect known as simultaneous contrast. We force the foreground color into a different direction to exaggerate its distinction from the background. Cameras don't do this, with the unfortunate result that any photography taken of soemthing full of similar colors will need to be corrected if it is to look natural. For example, if we find ourselves in a forest, we cheerfully and unconsciously visualize all kinds of variations in its greens and browns, much to the frustration of photographers whose best forest shots lack that reality.
    Dan goes on to list many ways in which the camera fails to capture the same reality the eye sees, but his big point is that there really is no objective reality that the camera captures and which the eye does not (or visa versa).

    Like all represtations of reality, photography is an illusion which works when it plays the right chords in the viewer's brain. And make no mistake that the brain is a complex mousetrap which uses enormous and sometime inscrutable processing to decode what it sees. Here is a little example which proves just how complicated and inscrutable this can be:

    2790910-M.jpg

    So what's all this got to do with picutres of birds in the sweet light? Quite a lot, actually. This light is lots warmer than "normal" daylight. In fact, it's almost as warm as tungsten (but more yellow). My portrait in the sunset was meant to illustrate that there are times when we aren't happy with this color balance. Almost all of us don't want our friends to be too yellow. The birds seem to be a horse of a different color, though. They absolutely are white, aren't they? But in this case, almost nobody (including me) wants them to be completely neutral in these sweet light shots. Some, like Harry and Jeff, even seem annoyed at the very prospect of changing the color balance of these shots. But that really misses my point. Better to ponder the difference in how we see the portrait vs the birds. The questions I'm asking (and not answering) have to do with the difference in how we see these two images, both shot under similar light. Greater understanding of this will inevitably lead us to better treatments of those great sweet light shots. (A similar project last winter resulted in a vast improvement of my winter photography: http://www.dgrin.com/showthread.php?t=6366)
    This is an intellectuall persuit that won't be of interest to everyone. For me, achieving this kind of understanding is just about the most satisfying thing about photography.
    If not now, when?
  • ginger_55ginger_55 Registered Users Posts: 8,416 Major grins
    edited August 31, 2005
    On a related subject, not sweet light, but focus. My simple book, Digital Wildlife Photography by Chris Weston, says on DOF,

    "There is only one point of focus. Everything in front of and behind that point is technically out of focus. However, because of the limitations of the human eye, an out of focus area of the scene can appear sharp. This area is known as the zone of acceptable sharpness..................."

    I found that "point" in the book very interesting, have planned to go back to it to see what it means to me/us in a practical sense.

    Another human eye issue.

    Very interesting questions you have raised, Rutt, IMO.

    ginger
    After all is said and done, it is the sweet tea.
  • ginger_55ginger_55 Registered Users Posts: 8,416 Major grins
    edited August 31, 2005
    Harry, I understand, somewhat, your fondness for the Golden Great Blue Heron, but do you think the Little Blue Heron has a pleasing WB? Interesting to have seen it through your eyes, in that light, as you shot it. At least the bird caught a golden fish, smile.

    Looks like it would have been a good time to pan for gold. Pan for gold? It is early hours for me, yet, but that has an interesting couple of meanings.rolleyes1.gif "Groan"
    ginger:):
    After all is said and done, it is the sweet tea.
  • pathfinderpathfinder Super Moderators Posts: 14,694 moderator
    edited August 31, 2005
    I have arrived late to the party just as the guests are finishing their last drink and putting on their coats. :D

    This discussion really sounds a lot like the "What color is Snow" thread, and John's question really revolves around why the warm sunlit egrets seem more acceptable than the candid portraiture. I think our brains know that white can be many colors depending on the ambient light, but our brains are more strict about skin tones. We don't like funny colored people. Too yellow is jaundiced and that is contagious - maybe that is why we react to skin tones too yellow ne_nau.gif But we willingly accept snow that is blue, pink, or yellow depending on the ambient sunlight. Sometimes the ocean looks pink in the fading sunlight too.

    I think this quote of rutt's from the snow thread is pertinent -"What's really odd is how the brain works. It will do it's own "auto white balance" for real situations but for photographs of them, you have to balance for it."

    This is very true - when we see things in reality our eyes autobalance and things seems "correct", but when we see the same colors in a print we find it unacceptable. Interestingly I think this mainly refers to prints and not transparencies. Think of looking at late afternoon sunlit scenes in movies - they look as they look in real life and they are very warm toned. Part of the problem with prints may be that the light illuminating the print is so different than the light the image was shot in. For example, ImagePrint has different RIPS for prints that will be lighted by daylight, tungsten or flourescent lighting....

    The eye has been described as a camera for ages, but it really DOES NOT function as a camera that takes in the entire scene sharply all at once in a brief instant. The eye+brain is more like a real time scanner that scans the scene and then goes back and examines detailed areas one at a time and slowly assembles an image in our minds eye - that is why we "see" scenes sharply all over when in fact our vision is very blurred except for the central 2 degrees where central vision takes place. I think this scanning and then assembling the image in our mind plays a role in how the color balance is evaluated also.

    As a practical matter I tend to use "as shot" in ARC if the lighting is sunlit or shade or whatever. If the image does not look right ( and if I read the white and or grey pixels as colored ) I may adjust the color slider as needed - frequently I find that a value about half way between "as shot" and the actual - shade, tungsten, cloudy whatever works pretty well.

    John, I just received Margulis "The Canyon Conundrum" a few days ago. Looks like a lot of fun stuff in it too.
    Pathfinder - www.pathfinder.smugmug.com

    Moderator of the Technique Forum and Finishing School on Dgrin
  • ruttrutt Registered Users Posts: 6,511 Major grins
    edited August 31, 2005
    I thought about Ben's and Harry's birds all day. Harry's birds don't bother me, but there is something about Ben's that was nagging at me. What? I think I know. Ben's birds are white. We know they are white. It's like having a white card in the shot. If it isn't white, well it isn't really believable (see above quote from Dan Margulis about ambient light.)

    Yet we want to show how warm that great sweet light is. I actually think the previious versions of this shot (mine and Ben's ) fail to do this. Without a white point, we (I especially) fail to register that yellow as yellow. It just looks grey. It needs a neutral point to contrast with in order to look really yellow.

    So I had this idea. I'd neutralize the most neutral point on the birds (also the lightest, most "blown", part as it turns out) and then steepen the B curve to make the more yellow parts at least as yellow as in the original. That should make them actually look yellow by contrast and satisfy my eye's craving for a neutral point. It tells my eye a believable story: the light on this bird is so yellow that even if some of it is neutral, not all can be. A side effect of steepening the B curve on the yellow side was to steepen it on the blue side as well. I could have prevented it, but I found the bluer water to be very pleasing and to invoke the way water actually looks to me in the "sweet light". I don't know if this is actually caused by some deeper, undiscovered (by me at least) truth, or just a happy coincidence, but I let it alone instead of trying to retain the gunmetal grey of the precurve version.

    This time I worked hard to start of with an original that retains as much highlight detail as possible. I was very careful to avoid any highlight clipping during my raw conversion and then was pretty aggresive with the highlight values of shadow/highlight in order to biring in all the feather detail I could. The result was a little dark, but I fixed that with the L curve. So I think I recovered a lot of the detail in the blown areas, but that really isn't the pont.

    The point is this B curve:

    34258448-S.jpg

    and the resulting image:

    34258431-L.jpg

    The rightmost point of this B curve moves the whitest points on the birds all the way to neutral. And then the leftmst point pulls the most yellow point on the feathers almost all the way back to it's original value. So now there is a white point (on the topmost bird's shoulder), but also quite a bit of yellow, which now looks yellow to my eye instead of just not white.

    I'm very interested in opinions about this version, especially from the birders.
    If not now, when?
  • ruttrutt Registered Users Posts: 6,511 Major grins
    edited August 31, 2005
    pathfinder wrote:
    I have arrived late to the party just as the guests are finishing their last drink and putting on their coats. :D

    Not too late for me. I'm still interested. I posted before I saw your post and you posted before you saw my last post. I went back and reread the color of snow posts. It seems that the lesson there is very like the one you subscribe to: compromise. That's different from the idea I just tried with the birds, but now I'm thinking that the steep B curve idea might also work with the snow, making some points relly white to contrast with the darker blue points.
    If not now, when?
  • XO-StudiosXO-Studios Registered Users Posts: 457 Major grins
    edited August 31, 2005
    One key question remains overlooked
    While it is true that the human eye, will adapt, I believe we are missing a big point in this conversation, the bigger picture.

    And I believe we all do it, be it portraits, wildlife, landscape.

    We as an audience at DGRIN, are far and above average where it comes to colors, pictures, composition, what is white, what is black, but in reality we are a very minute portion of smugmug user, and even of the overall buying audience.

    We often forget that the ppl who will eventually buy our prints (or do the oooohs and aaaahs) are not the technicians, or perfect art critiques. Odds are Joe Blow will actually like the warm 'incorrect' colors. Tons of soccer-moms probably like their picture over-saturated, and dad could not care less if the picture has a poor white balance, as his son in the picture catches the touchdown.

    A discussion like this cannot be had w/o considering the audience; it is a different discussion for each audience, or for every use of the picture(s) there is a different post process that will create the best picture for that particular use. I have oodles of pictures that are very technically challenged (which is PC for FUBAR) yet I do not delete them, as the picture sparks a memory, or has a good story.

    The portrait on the beach, while the skin color might read incorrect, the incorrect skin color does reflect that sunset at the beach. So what is the audience, is it an Art director looking for a stock photo, by all means correct to a correct skin color, but if it is a snap shot to illustrate the great time everyone had on 'the Cape' incorrect skin color might be the way to go. The birdies, again what is the audience, is it for use on a hallmark card, or is it the National Audubon Society (SP).

    Anyway, getting off of my :soapbox

    XO,
    You can't depend on your eyes when your imagination is out of focus.
    Mark Twain


    Some times I get lucky and when that happens I show the results here: http://www.xo-studios.com
  • HarrybHarryb Registered Users, Retired Mod Posts: 22,708 Major grins
    edited September 1, 2005
    ginger_55 wrote:
    Harry, I understand, somewhat, your fondness for the Golden Great Blue Heron, but do you think the Little Blue Heron has a pleasing WB? Interesting to have seen it through your eyes, in that light, as you shot it. At least the bird caught a golden fish, smile.
    I don't think that the Little Blue gave a darn if his WB is pleasing or not. :D The simple fact is that is how he appeared when I took the shot. If I wanted to I could easily adjust the colors of that moment and make his WB more "pleasing" heck I could make him a Little Red Heron.

    When I shoot wildlife I try to capture the moment as accurately as I can. I do that by setting my camera's exposure and WB correctly. In my post work I then try to address the limitations of the camera. For example when shooting a white bird I will expose for the bird and this will often under expose the shadow areas and I will use PS to bring out the details in those areas and then use NR on those areas to decrease the resulting shadow noise.

    The color in a shot, in my experience, will usually need some minor adjustments. Usually setting a grey point will do that fine. I find that if I make significant adjustments to the color of a shot the result is unreal.

    PS is a double edged sword. If used properly (IMHO) it can improve a pic that was taken correctly. If I did my job my well the PS work is minimal. Whenever I have to do a lot of Photoshopping on a shot its usually a sign that this is a shot headed for the cull group.

    If you need PS to "save" a shot thats usually a good indicator that this is a shot that the recycle bin was created for.
    Harry
    http://behret.smugmug.com/ NANPA member
    How many photographers does it take to change a light bulb? 50. One to change the bulb, and forty-nine to say, "I could have done that better!"
  • HarrybHarryb Registered Users, Retired Mod Posts: 22,708 Major grins
    edited September 1, 2005
    34258431-L.jpg

    Looking at this shot w/o comparing it to the orignal my first thought is pretty good but it could have been so much better if the exposure had been knocked down a thrid or a half a stop.

    Comparing it to the orignal I prefer the original shot. What I like about the second version is that you have accentuated some of the feather detail more clearly than the orignal but the blown areas are so much more evident than they were in the original. The yellows of the morning light are much more yellow. Too much IMHO so that no longer appear natural. Also they eyes don't appear as clearly in the second version. The blue of the water is prettier and more vibrant but thats not necessarily better. Having done this myself a number of times it looks good initially. Then when I look at it later I'm struck by how it doesn't look real (my attempts as well as yours here). Also its so pretty it starts to compete with the subject, the egrets, for my attention when I look at the shot.
    Harry
    http://behret.smugmug.com/ NANPA member
    How many photographers does it take to change a light bulb? 50. One to change the bulb, and forty-nine to say, "I could have done that better!"
  • jwearjwear Registered Users Posts: 8,005 Major grins
    edited September 1, 2005
    another try
    As we go through life, our brains accumulate and store values for common, strongly (and consistently) coloured things – such as grass, sky blue and bananas. Our brains also have fixed ideas about the possible colours of skin, and are extremely fussy about how they are represented.

    Pastel shades are more susceptible to colour casts than stronger colours, as anyone who has tried perfecting skin tones in a photograph will attest. But fake tan or not, we can tell instantly if a face has even a slight tinge of unrealistic colour.

    Natural rendering of skin tones has long been a challenge for the film and digital imaging industry; technicians devote entire careers to rendering skin and there are countless training courses and computer programs dedicated to the task.
    And because our brains compensate for ambient light changes – after a few moments we don't 'see' the green of fluorescent tubes, for example – our colour memories remain remarkably untainted. It's only when we see through the eyes of an RGB sensor or film emulsion that colour casts are revealed as glaring 'errors'.

    The closer our emotional connection to an object and the greater its familiarity, the better we can accurately recall its colour. This is why we’re so adept at judging the colour of skin. The same applies to a photograph: the more important its content, the more we will notice any incorrect colours.

    Memory colour varies between cultures and their environments. Inuits can famously name many different shades of snow and ice, while in Japan the camera’s great challenge is to reproduce the different tones of the sea. Only natural as it surrounds the country and, traditionally, is its main source of food. In the same way, photographers in Europe and the US are more likely to be critical of inaccurate representations of harvest gold and grass green.

    Even though we can differentiate between hundreds of hues and shades, relatively few can be considered memory colours. Colour theorists put this number between six (Rudolf Arnheim) and 12 (Paul Overy).

    Because we form memory colours by combining memories from many different observations, and because we are more likely to register colours when they are bright and saturated, we tend to remember them more vividly than they usually appear. Even so, studies have shown that we more accurately recall hue than saturation.

    When it comes to tweaking digital images for reproduction, memory colour is a good guide. It's usually impossible to compare the image with the original scene.
    Jeff W

    “PHOTOGRAPHY IS THE ‘JAZZ’ FOR THE EYES…”

    http://jwear.smugmug.com/
  • pathfinderpathfinder Super Moderators Posts: 14,694 moderator
    edited September 1, 2005
    Harryb wrote:
    34258431-L.jpg

    Looking at this shot w/o comparing it to the orignal my first thought is pretty good but it could have been so much better if the exposure had been knocked down a thrid or a half a stop.

    Comparing it to the orignal I prefer the original shot. What I like about the second version is that you have accentuated some of the feather detail more clearly than the orignal but the blown areas are so much more evident than they were in the original. The yellows of the morning light are much more yellow. Too much IMHO so that no longer appear natural. Also they eyes don't appear as clearly in the second version. The blue of the water is prettier and more vibrant but thats not necessarily better. Having done this myself a number of times it looks good initially. Then when I look at it later I'm struck by how it doesn't look real (my attempts as well as yours here). Also its so pretty it starts to compete with the subject, the egrets, for my attention when I look at the shot.


    Harry, if you could, would you please point out the 'blown' areas in each of rutt's three images. I am not sure I understand what areas you are referring to.

    When I run a pixel meter over the whiter areas in these images, I do not see 250 or higher in any channel of the pixels and almost all are 240, 245, 242 or less. There may not be great variation in these areas, but my understanding of blown pixels would be 255,255,255 in a lcd screen, or maybe 245,245,245 in a printed page. John is meticulous in his processing and I doubt he posts images with 250,250 250 or higher in them without specific intent.
    I tend to prefer the warm suniight of the late afternoon too, so I am not sure I prefer the postprocessing he has gone through, but I do find the topic informative, interesting, and worthwhile discussing. I will be facing precisely these issues this weekend as I photograph my niece's wedding out of doors in the late afternoon sunlight. White dress anyone?
    Pathfinder - www.pathfinder.smugmug.com

    Moderator of the Technique Forum and Finishing School on Dgrin
  • bfjrbfjr Registered Users Posts: 10,980 Major grins
    edited September 1, 2005
    this has been fun
    I must say I'm very happy with this thread. If nothing else it has helped me understand that I am right about my feelings about photography and others may not totally agree but it seems were all on the same page.
    firstly I stand behind my original reply, and the replies that followed indicate the same, "there is not one correct answer".

    Everyone here should be thankful to Rutt for his interest and his work, it helps us all. Even if we do not agree with some of his contentions or methods. I could be wrong but what I see here is Rutt's analytical approach is at odds with others artistic approach, and this in not new thing.

    I am with Harry in all that he has pointed out, except that there were blown areas in the original. I always make every effort not to approach 255, and I am pretty sure I did not in the original image. Of course I could of missed something.

    I personally am a wildlife shooter (most of the time) and having said that look eagerly for the sweet light. Actually any light real or flashed which will add another dimension/impact/detail to my work. My camera is usually is in it's bag between 11am and 3pm, unless I'm hunting raptors.

    I do have a problem with something that was said here, and that is that I somehow manipulated my original before posting in this thread:
    Hey Sweetie

    What follows here is my exact workflow before I posted the three shots and that workflow is what I normally do to all my image. If I have to do more it will usually get culled or it's somehow worth the extra work for whatever reason.

    1st my in camera settings are all at neutral and sharpening is set to none

    I download to a windows machine with a card reader

    I view and make decisions via Nikon View

    Take chosen images to Nikon Capture V4.3.1 (Harry have you downloaded upgrade, noticeably faster)

    1st apply a D2H preset (from Ron Reznick) which applies sharpening

    Check WB by using curves palette and setting a gray point, by finding a spot in image that measures 128. in the image here it really popped the blue and I had to bring it down a bit by dragging the blue channel back a bit.

    2nd check if exposure needs any compensation. This image was brought down another 1/3 of a stop

    Apply color booster by clicking auto in the proper palette

    from here the image was uploaded to Photoshop CS2 in a 16bit Tiff format

    In Photoshop I crop to my liking and I apply local contrast enhancement by using USM. Settings were 50,20,0

    Apply my EXIF script and use an action to upload to my smugmug acct. in an 8bit jpeg format with SRGB color space.

    That's it folks. Some may actually call what I do manipulation, but not I.
  • HarrybHarryb Registered Users, Retired Mod Posts: 22,708 Major grins
    edited September 1, 2005
    Ok PF here are the areas that appear blown out to me. On the egret in the upper left the area on the top of the head, the shoulder area before the wings, the outer area of the left wing thats opposite the neck area, and a small area of the lower chest thats right below the right wing. On the bird in the lower right you have the area on top of the head and small areas on the outer edge of the left wing.

    These areas are devoid of details. They may not read 255 now but at one time they probably did. I have blown out areas in my shots and then in my post adjusted the exposure or levels so that the blown out areas no longer read 255 but nothing can restore the lost details. In Ben's original shot they were they but not as clearly as they are in Rutt's version. The area around the eye of the egret in the upper left is particularly distracting to my eyes.

    Blow outs cannot always be avoided. I think Ben's original shot shows that. If he had exposed the shot to completely avoid some blow outs on the egret the overall frame would have been underexposed. I think he handled a difficult exposure really well but the there are some blown out areas. Rutt's version while improving some parts of the pic also made the blown out areas more apparent and distracting.
    Harry
    http://behret.smugmug.com/ NANPA member
    How many photographers does it take to change a light bulb? 50. One to change the bulb, and forty-nine to say, "I could have done that better!"
  • ruttrutt Registered Users Posts: 6,511 Major grins
    edited September 2, 2005
    I decided to give Ben's birds one more go. My previous attempt was a fairly quick and dirty experiment to see if having at least one neutral point would make the more yellow parts of the bird actually look warm and yellow instead of just dirty. In that I think it succeeded, but I clearly went too far and the result is harsh. This time I've done it as carefully as I could in an effort to find out if the idea can really work.

    Here is my result:

    34414269-L.jpg

    To my eye, this looks good, clearly showing the warmth of the light, recovering more highlight detail than any previous attempt, and leaving the water a plausible color.

    I captured the steps as I went along.

    Raw conversion with PS/CS2 ACR

    34414310-M.jpg

    Ben used Nikon's software to do this. That at least partly explains why I couldn't reproduce the exact shade of his water with an obvious PS setting. Ben also used a custom temperature setting derived by setting a grey point and then fine tuning the settings. Because I planned to use curves later on to achieve color balance, I just converted using "Daylight" white balance which kept the birds fairly warm and the water a plausible color. I could have used almost any of the outdoors settings or "As Shot". There wasn't a huge difference.

    What I did do that is a little unconventional, is dial the "exposure" setting down a lot and dial the "brightness" setting up a lot. I did the former in order to avoid any clipping at all. I wanted to recover all the highlight detail possible and allowing any of the histogram to escape off the right will "blow" some pixels. I want it all after the conversion so I can use curves and highlight/shadow to render it visible. I dialed the brightness setting up to more evenly distribute the histogram without clipping it. Certainly, I could have used curves to do this after conversion, but this move makes my live easier in subsequent steps.

    Highlight/shadow

    34414318-M.jpg

    The goal here is to make visible the highlights which I so carefully preserved in the raw conversion. Notice that I have the shadow amount set to 0, so that I am only targeting highlight recovery. Notice also that I have a farly low tonal width but a very large amount set for highlights. The idea is to target only the brightest parts of the image (thus avoiding darkening the parts that don't need attention) and to recover quite a lot there.

    Lab curves

    34414328-S.jpg34414292-S.jpg34414313-S.jpg

    There are two things going on here.
    1. The B curve implements my idea of getting some part of the white bird to actually be neutral. This time I was very carful about picking the lightest most neutral pont to neutralize. Also, I was careful about how steep I allowed the curve to be and the effects of the steepening outside the range of yellow on the birds' feathers. So this is a much less radical implementation of my idea than my previous attempt. I still helped myself to some blueness in the water, but only a little. The A curve has a small amount of not quite symetric steepening. This is something I didn't do last time, but which makes a rather large difference fow such a subtle change. Firstly, this balances the stronger blue in the water with a little more green to keep the water color more plausible (and soft). Secondly, it erases a hint of magenta at the selected white point but not in other parts of the white feathers. It's important to my experiment to get the white point actually neutral. And I've learned (from my last try) that it's important to temper that added yellow with a touch of magenta to keep it realistly warm instaed of allowing it to turn bright yellow.
    2. Further, increasing the details in the highlights with the L curve. As Dan Margulis loves to say, "The steeper the curve, the more the contrast." Here I spent as much of my contrast budget as I could afford on those highlights at the expanse of parts of the birds that had well defined contrast and could suffer a slight decrease. I also stole come contrast from the water, which as Harrry points out, isn't really the subject there. Like all such L curve moves, this one is Robin Hood, stealing from the rich (the well exposed midtones) to give to the poor (the overexposed highlights).

    Steepen the black curve

    34414323-S.jpg

    At this point I was left without satisfying black points on the birds' legs and beaks. Also the water color wasn't as deep as Ben's and I was envious of that. I already spent my L curve budget. Any more steepening of hte right side was likely to darken parts of the shot that I didn't want to and also lose shadow detail. I've learned that this problem can often be easily addressed in CMYK with by steepening the black curve.

    Sharpen with separate light/dark layers This time I was a lot more careful with my USM parameters and the opacity levels the the lighten/darken layers (if you don't know what I'm talking about here, see: http://www.dgrin.com/showthread.php?t=9739) I needed a fairly small radious (.9) and very low threshold (2) in order to bring out the feather detail without obscuring it with neighboring halos. Then I left the darken opacity at 100% and decreased the lighten opacity to about 50%.
    If not now, when?
  • ruttrutt Registered Users Posts: 6,511 Major grins
    edited September 2, 2005
    Oh, and I do have a question for the birders. Is occured to me that maybe my basic assumption is incorrect and these birds really aren't completely neutral after all. Is that possible?
    If not now, when?
  • bfjrbfjr Registered Users Posts: 10,980 Major grins
    edited September 2, 2005
    rutt wrote:
    Oh, and I do have a question for the birders. Is occured to me that maybe my basic assumption is incorrect and these birds really aren't completely neutral after all. Is that possible?
    YES YES YES!!!

    Here I think you've hit an extremly important point that I even overlooked. No animal feathered, fured, etc. is ever one color!! Neutral as you call it.
    For instance my Cockatoo (Henry) is considered to be white, but there a tinges of yellow throughout. My B&G Macaw has varyed degrees of Blue from light to quite dark with some gray thrown in. I've shot Comorants that look mostly black but are truly varing shades from tan to yellow to black. Egrets have bright whites to gray on their feathers. Coyotes I've shot show different shades of brown to black throughout their fur.

    And all of these may or may not be visible to the eye or camera depending on the light they are shot in. Another reason wildlife shooters search for light that highlights these differences thereby of course enhancing the overall image.

    I really think you've brought up another excellent point and one that I'm sure again will have many different opinions.
  • jwearjwear Registered Users Posts: 8,005 Major grins
    edited September 2, 2005
    Ben said it well -there are many colors in white [shades ] and this is different in each bird also .the wing in this shot[ bright light ] you can see the color

    variations in the wing and his plume. the underside of the wing has even more but this is the only example i had quick.In the shot you did of Bens you can see the underside very well the color is more yellow than white
    Jeff W

    “PHOTOGRAPHY IS THE ‘JAZZ’ FOR THE EYES…”

    http://jwear.smugmug.com/
  • ruttrutt Registered Users Posts: 6,511 Major grins
    edited September 2, 2005
    Thanks for the answers, Ben and Jeff. Beautiful shot, Jeff, how did you get it? Did you have to fly an untralight?

    I think my question was a little ambigious, though, and I don't think you guys answered exactly the right question. Let me try again. Is it possible that there is no neutral white spot at all on this bird? I assumed that at least one spot would be really neutral. I guess I also assumed that that warmness was amost 100% from the light not the bird. But the effect on the rest of the shot of simple techniques (like the raw converter eyedropper) to neutralize the bird is so obviously wrong that it makes me wonder. By contrast, using the raw converter custom WB tool on the grey hair in my portrait was like the kiss of the good fairy on the man's face, which leads me to believe that I was dead right about the neutrality of that hair.
    If not now, when?
Sign In or Register to comment.