Wednesday Landscapes, 21 June 2017

You are invited to upload one or more of your landscape photos and, if you’d like, to accompany your image with some commentary: challenges you faced in making the image? your intent for the image? settings? post-processing decisions? why you did what you did? the place and time? or an aspect you’d like feedback on? And please feel free to ask questions of others who have posted images or to join the discussion. If you don’t feel like using words, that’s OK too—unaccompanied images (or unaccompanied words, for that matter) are also very much welcomed. As for the technicalities, the usual forum guidelines apply: files < 1 MB; image size <1000 px maximum dimension.

 

This is a 5-frame panorama of a January sunrise. I’m curious about how you handle color in sunrises and sunsets—I find it difficult to hit the right balance between reality and and what looks like reality, and I often have to dial in a negative vibrance to tone down the vividness of the original raw files. In this case, the only processing is a reduction in white point and highlights and a strong lightening of shadows, along with a moderate increase in clarity (mostly to bring out the foreground details). I probably should have used liquify to straighten the telephone poles on the left, and I should have locked in the same exposure for all frames (spot-metered on a light portion of the blue sky--not the brightest part--and remetered to approximately the same place for each frame [flakey, I know. But I was in a hurry]; 1/30 to 1/60, f/5, -2/3 EV, ISO 200, 18 mm; set to camera neutral profile).

 

You are invited to upload one or more of your landscape photos and, if you’d like, to accompany your image with some commentary: challenges you faced in making the image? your intent for the image? settings? post-processing decisions? why you did what you did? the place and time? or an aspect you’d like feedback on? And please feel free to ask questions of others who have posted images or to join the discussion. If you don’t feel like using words, that’s OK too—unaccompanied images (or unaccompanied words, for that matter) are also very much welcomed. As for the technicalities, the usual forum guidelines apply: files < 1 MB; image size <1000 px maximum dimension.

A really big sky, Leslie!

This is a 5-frame panorama of a January sunrise. I’m curious about how you handle color in sunrises and sunsets—I find it difficult to hit the right balance between reality and and what looks like reality, and I often have to dial in a negative vibrance to tone down the vividness of the original raw files. In this case, the only processing is a reduction in white point and highlights and a strong lightening of shadows, along with a moderate increase in clarity (mostly to bring out the foreground details). I probably should have used liquify to straighten the telephone poles on the left, and I should have locked in the same exposure for all frames (spot-metered on a light portion of the blue sky--not the brightest part--and remetered to approximately the same place for each frame [flakey, I know. But I was in a hurry]; 1/30 to 1/60, f/5, -2/3 EV, ISO 200, 18 mm; set to camera neutral profile).

 

[ATTACH=full]1193822[/ATTACH]

Near Highlands, North Carolina

Sand Dunes National Park

 

Sand Dunes National Park in Colorado by David Stephens, on Flickr

 

Sand Dunes National Park is full of photo-ops, but you have to vary your subject with the light and sky. This was an how before sundown, so the sun was starting to cast strong shadows on the sand. The sky had some interest, but it wasn't super at this point, so I decided to focus on texture and shadows, but include the background. I used my EF 14mm f/2.8L II and got down close to the sand. I stuck with my general setting of f/8 because I wasn't right on the sand, where I might have needed f/16. This is hand held at 1/400-sec., ISO 400, with a 5DS-R. RAW conversion was with DxO Optics Pro, with very little adjustment.

This is a 5-frame panorama of a January sunrise. I’m curious about how you handle color in sunrises and sunsets—I find it difficult to hit the right balance between reality and and what looks like reality, and I often have to dial in a negative vibrance to tone down the vividness of the original raw files. In this case, the only processing is a reduction in white point and highlights and a strong lightening of shadows, along with a moderate increase in clarity (mostly to bring out the foreground details). I probably should have used liquify to straighten the telephone poles on the left, and I should have locked in the same exposure for all frames (spot-metered on a light portion of the blue sky--not the brightest part--and remetered to approximately the same place for each frame [flakey, I know. But I was in a hurry]; 1/30 to 1/60, f/5, -2/3 EV, ISO 200, 18 mm; set to camera neutral profile).

 

I often run into the same problem your expressing, "This sky is unbelievable. People aren't going to believe it." So I'd take the shot and try to etch the colors into my brain so that in RAW conversion I can get them "right." I'm constantly asking myself, "Is that what I saw?" A few month back, I started carrying a color card that I hold up in front of the camera, with the sky behind it. This gives me an objective target to use for setting color in my digital darkroom. It's funny that my shots looks pretty much as before, but I can say that I color calibrated, using a card.

I spot meter and pick mid range areas -- usually half a dozen shots, one or more will hit the colors just about spot on. I chimp and adjust and check. As previously mentioned on various threads, I shoot JPEG and do little PP. If I post it, as with those from last night posted today, it was very close to what I saw...of course what I saw might not be what you would have seen! ;-)

I think that people tend to overdo saturation in sunrise/sunset photos in order to emphasize the warm colors. I know that I am often guilty of this. Another variable is color temperature. Here is a sunrise photo taken with daylight white balance (4850 kelvins). What is wrong is that we perceive objects that we know are white, such as snow, to look white , not blue despite being illuminated by a blue sky. Color temperature may be varied, in this case using Adobe Camera Raw, so that the snow looks more natural. A setting of 6500 kelvins was used in the following image. The result leaves the sky looking undersaturated, so I increased saturation +70 in ACR. The result, I think, looks realistic, although if I had a print instantly available for comparison to the actual sunrise, I bet that it would look quite different. Human perception of photographs and the natural world are two different things, and the two can never completely correspond.

Useful, Glenn! On occasion, fooling around, I will use the "wrong" White Balance after trying several because I like the result. I hadn't got into the Kelvin adjustments since I rarely use studio lights. Something new to try. Thanks!

Thanks Sandy. If I were making a print, I would increase midtone contrast, brighten the image, and increase saturation a judicious amount without, hopefully, overdoing it.

Hello everyone. Hawaii is starting to push close to 90f and the above snow shot's are good to see. Diamond Head was shot from Kapiolani Park, which has a refreshing, cool breeze most of the time, so I am semi cool under the sun when working there. Isolette II, G filter, Weston 853, UFX100, OA & V600 scan. Aloha, Bill

Thanks Sandy. If I were making a print, I would increase midtone contrast, brighten the image, and increase saturation a judicious amount without, hopefully, overdoing it.

 

So, is this accurate WB, or what you preferred as a human?

 

In my cameras, both WB and metering are usually fooled by a snow scene. I shoot in RAW and will ETTR any snow scene, but the sky is often unbelievably colorful and dramatic here in Colorado. That's why I've resorted to carrying a color card with me, so that I've got some objective measure to help me "remember" what I saw. When your dealing with different levels of DRAMATIC the memory, at least mine, quickly clouds into uncertainty. I like being able to say, "Yes, that's really how it looked" with certainty.

Really, it comes down to a large number of variables -- how we each perceive color that varies considerably person to person. How our (various) cameras record and display color -- that issue compounded across the various camera settings. Different cameras, even different models in the same brand e.g. my DF vs. my D 750. How the monitor renders the color, and last but often problematic for me, the printer.

Glenn's method is what I have to do with my Canon PP 100.

I often think of printing direct from the camera as I did with my little Canon Seplhy back when. That would reduce the variables at least.

...

I often think of printing direct from the camera as I did with my little Canon Seplhy back when. That would reduce the variables at least.

 

That would reduce variables, but it wouldn't be closer to "real", whatever that is. (In-camera interpretations are merely the compromises agreed by a committee of Japanese engineers).

 

The more fantastic the reality, the more I feel a need to validate it objectively. That's why I've taken to carrying that color card. To keep the chain intact, I've calibrated my monitor and I do test prints for every new paper on my PRO 100 printer.

 

I think that it's become a presumption that most of "us" (serious photographers) manipulate our images to a high degree. I like to be able to confirm that what I'm showing really does happen.

I do very little manipulation, but really can't get into carrying a card. The only value it would have would to indicate that manipulation was needed -- Post Processing best accomplished by shooting RAW, which does not interest me at all. Post Processing itself not something that appeals to me much, though it can be handy. I'll just have to muddle along. Thanks for the thought.

Jodhpur

dc said, "So, is this accurate WB, or what you preferred as a human?"

 

We humans perceive snow to be white, so I adjusted white balance in my second and third examples to remove the blue cast. (Take a white sheet of paper outdoors in shade where the illumination is from blue sky, and compare your visual color impression to the same white sheet indoors under warm tungsten or fluorescent lighting. Doesn't white still look white? ).

 

I am not sure what the snow would look like if I adjusted the white balance so that the colors of the sunset were accurate rather than worrying about the snow. This would be something to try with a color chart in hand at the time of exposure. And then there is the matter of color saturation. How does perception of saturation correspond to an image displayed on screen or in a print? I don't know, but I suspect that they can be quite different.

 

The sunset in the second example might or might not approximately correspond to what I saw, but my third example is what I wanted the sunset to look like without, hopefully, being too unrealistic.

I rarely do sunsets, but I couldn't resist this one on a small lake in the park I mountain bike through every day. I tried to re-create what it looked like in my mind's eye. I could easily punch up the color and contrast of the clouds, but this is more what it really looked like, more pastel.

I like the pastel colors in sjmurray's photo. Boosting saturation would, in my mind, be a mistake.

Thanks very much to all of you for the very useful discussion on the sunset issue—you’ve given me some ideas to try, and I’m thoroughly enjoying the photos you all posted.

a color card that I hold up in front of the camera

That’s an interesting point about using a color card—I use it for documentary botanical shots to get accurate color, but I had assumed it wouldn’t work for sunsets for two reasons: (1) the warmer-than-“accurate” color cast is actually perceivable in this setting, so an accurate rendering of the color card might not convey what the brain interpreted (as per Glenn’s comments), and (2) in many cases, the light source for the clouds (direct sunlight) is very different than that for the foreground (diffuse light from the sky)—it’s the same problem as getting a useful color balance for mixed sun and shade, and of necessity the color card in this case would be exposed in the shade unless the sun is above the horizon. But the next time we actually have a sunrise here (...it’s fog season, so I may need to wait until September…) I’ll definitely give it a try.

what I saw might not be what you would have seen

Good point on everyone’s color perception being different. It also turns out that corneas yellow slightly with age, which means that we’re all shifting our own color perceptions through time, too. Everything in the real world still looks the same since the brain accommodates the changes, but we might start finding that the photos we processed long ago take on perceived “color casts” that aren’t actually there.

Human perception of photographs and the natural world are two different things, and the two can never completely correspond

I’d been stressing out several years ago about the problem that the idiosyncratically programmed on-board processor located behind our eyes renders the scene in a way that doesn’t correspond to reality, then realized that this gives me a lot of freedom to interpret a scene like I want it to be—I think I arrived at pretty much the same place you have: I’m processing to achieve a result that looks like I want it to look; but within the bounds that David outlined and that you’ve achieved in your beautiful example: it has to at least be a plausibly realistic result. It’d be interesting to process a scene in a variety of ways and figure out how much of a consensus there is among viewers as to which is the most attractive rendition. BTW, I much enjoyed seeing your recommendation several months ago of Livingstone’s “Vision and Art”—I’ve relied on it for several years now to try to make some sense of the color realm I’m trying to depict, and I also heartily recommend it.

On the saturation issue: a watercolor trick for making a color appear to be more saturated than the pigment allows is to put it next to an unsaturated color, so I wonder if part of the issue with over-saturating sunset colors comes about because people tend to apply vibrance and saturation adjustments globally rather than for particular color ranges. If the indirectly lit foreground gets a saturation boost, the sunset colors would have to become even more saturated to look appropriately more saturated than the foreground. One of the ways I’ve tried dealing with this is to use the saturation adjustment instead of the vibrance adjustment in these cases, so that the saturation isn’t disproportionately enhanced in the less-saturated foreground, and to then dial back saturation in the blues using the HSL panel in Lightroom; sometimes I increase saturation and reduce vibrance to try to target a particular range. I think I’m going to have to explore the PS “blend if” option to see if there’s a potential solution in that.

How the monitor renders the color

Good point on the monitor problem. When I moved from a PC to an iMac for processing, all my previously edited photos took on distressingly over-the-top saturation levels. I calibrated both monitors and now they agree (the PC needed the biggest change, and I cringe to think of the photos I posted before calibration). David and I are using calibrated monitors—who else is?

And a final point on jpg v. raw: Sandy has an advantage in that what he sees on the camera LCD is more or less what the jpg is going to look like, so he can figure out what looks right while the show’s still in full swing. I’m using a neutral profile on the LCD in order to try to get the jpg histogram on the LCD as close to raw as I can (does this even make sense or am I wishfully thinking?). It might be interesting to shoot jpg+raw to be able to use Sandy’s strategy to get an image that you know looked right in the field, and then use the jpg as reference while processing the raw image.