Land of Pagodas: Results


Dan Margulis
 

I’ve posted the results of the Burma exercise, the eighth in a series of 11 case studies.

Reviewing: This is a dawn photo by Vincent Versace, whose raw file was available. We have 33 entries. Most people also submitted a list of their steps, thanks very much. I haven’t read these, because I’d rather get a sense of who was successful and who wasn’t before investigating why.

The files don’t have people’s names on them, and were random-generator numbered from #801 to #833. As with past studies, we also have a “par” version, #834. To get it, I chose five that I thought were among the best entrants, and averaged them, each one weighted 20%. This often creates a version that is superior to most if not all of its parents.

Normally I don't comment on results for two days after they're posted. Meanwhile, if you’d like to know how your own version stacked up, download the par version and compare the two directly. Do you think you got the same kind of quality? If not, I hope you’ll find further discussion useful.

The Folder is in the group's Photos section, 2021 Case Study: Land of Pagodas,
https://groups.io/g/colortheory/album?id=261860

I also have zipped all entries and uploaded a file to our Files section,
https://groups.io/g/colortheory/files/
Search for 032221_Land-of-Pagodas_entries.zip
If you are going to study these versions I strongly encourage you to download these files. Many of these entrants vary only in a minor way and it is hard to see the impact of a change without toggling back and forth between them.

I look forward to your comments.

Dan Margulis

P.S. Our ninth case study will be posted today; look for a separate post.


Kenneth Harris
 

Once again, my par and Dan's par differ by one image. Dan's par is a lot better than mine, since I rejected an image with a lot going for it except for the greens all heading toward kelly, which is something I've been conditioned to dislike, largely following from this tendency in pre-portra kodak negative film, and the screaming greens put out by early espon printers. Strangely, I prefer my own to either blend, a first. The par doesn't give much of a sense of recession into space, ie, it's feeling a bit flattened out, which hurts the mood.

Ken Harris


Kent Sutorius
 

I found this to be the most difficult image to work on. Mine was 823. I was stymied by how the mountain range should look and became frustrated with the process. Working with channels I kept getting the mountain range washed out (lighter) but even the par has some of that. I thought the range should remain dark and more defined. That error in judgment led to an unsuccessful look. Even though I sharpened the picture, it looks very  muddy compared to others.

Kent Sutorius

On 3/22/2021 7:20 AM, Dan Margulis via groups.io wrote:
I’ve posted the results of the Burma exercise, the eighth in a series of 11 case studies.

Reviewing: This is a dawn photo by Vincent Versace, whose raw file was available. We have 33 entries. Most people also submitted a list of their steps, thanks very much. I haven’t read these, because I’d rather get a sense of who was successful and who wasn’t before investigating why.
The files don’t have people’s names on them, and were random-generator numbered from #801 to #833. As with past studies, we also have a “par” version, #834. To get it, I chose five that I thought were among the best entrants, and averaged them, each one weighted 20%. This often creates a version that is superior to most if not all of its parents.

Normally I don't comment on results for two days after they're posted. Meanwhile, if you’d like to know how your own version stacked up, download the par version and compare the two directly. Do you think you got the same kind of quality? If not, I hope you’ll find further discussion useful.
The Folder is in the group's Photos section, 2021 Case Study: Land of Pagodas,
https://groups.io/g/colortheory/album?id=261860
I also have zipped all entries and uploaded a file to our Files section,
https://groups.io/g/colortheory/files/
Search for 032221_Land-of-Pagodas_entries.zip
If you are going to study these versions I strongly encourage you to download these files. Many of these entrants vary only in a minor way and it is hard to see the impact of a change without toggling back and forth between them.
I look forward to your comments.
Dan Margulis

P.S. Our ninth case study will be posted today; look for a separate post.



john c.
 

Without a doubt the the most difficult yet Kent, and mostly because there are so many possibilities that aren't pleasing due to the time of day. I think it's just before sunrise and we're pushing it to make it look brighter than it was and should be. I hated my version 810 when I sent it to Dan and told him so and suggested that I wanted to start over, so today I went back to the drawing board using the raw file this time and in less than five minutes I had a result that was vastly superior to the one I had submitted. Sometimes we just need a fresh look to see where we want to go. Let me add that I'm not a huge fan of the par version. I like lots of things about it, particularly the sky and the detail in the mountain range through the mist, but I think it's too bright and colorful given the dawn light, the vegetation is too yellow and the pagodas too colorfully magenta/red as well. I subtracted a lot of yellow/orange saturation from it before I found it pleasing, then I darkened it a lot before I loved it. Morning light can play tricks on the eye, particularly through mist and smog, but I think this came out overstated while trying to look as if it was taken in the middle of the morning rather than sunrise.


Gerald Bakker
 

Mine is 802, probably the lightest of all. It never occurred to me that the image should keep its dark look. Even though we knew this was a shot taken at dawn, my reasoning was that the human eye always adapts to any lighting situation, so a correction can follow that. Looking through all entries, I think my approach is defendable, although probably many of you will disagree. At least, I didn't find this a very difficult exercise.

I must say, I don't like the par. It has a good sky, but everything else is way too colorful. Somehow all these saturated colors don't look right for a dawn scene.
--
Gerald Bakker
https://geraldbakker.nl


Kenneth Harris
 

Mine is 832. I processed with capture one, and was pretty close to what I wanted before the file hit photoshop. I actually found this easier than most, since the only way to make this a good picture would be a recrop, and that's not allowed, thus I didn't torture myself thinking of solutions.

Ken Harris


James Gray
 

Ken,
In my opinion, yours is quite good.  If it is not picked for the par I will be surprised.  Mine is 819, and at least I am not embarrassed by it.  Actually, I kind of like the way mine came out.

Jim Gray

On Mon, Mar 22, 2021 at 1:21 PM Kenneth Harris <reg@...> wrote:
Mine is 832.  I processed with capture one, and was pretty close to what I wanted before the file hit photoshop.  I actually found this easier than most, since the only way to make this a good picture would be a recrop, and that's not allowed, thus I didn't torture myself thinking of solutions.

Ken Harris






Bill Theis
 

Mine = 807.  This is like a math problem.  When you know the answer, you can usually find a way to get it.  Having the par version in hand pointed out a couple of things that I now dislike in mine.  Too bright and contrasty for dawn, which I could fix by blending back about 20% of the original unretouched jpg.  I had a conservative version on top but it was not nearly conservative enough.  Maybe like the checking of autotone, I should make the habit of ALWAYS adding the original and checking if I have gone way too far. 

I studied the colors in the examples given from the Atlantic Magazine and tried to get towards that orangy color that I saw on the sides of the temple.  Big giveaway that I missed was that the red roses (?) at the center bottom are too bright and just not saturated enough--although most of the other colors were.  To fix the roses, into LAB and apply the inverse A to L in overlay which further lightens the greens so I did darken blend.  Inverse B applied to L tames the yellow blooms that are too bright and colorless (think reflections on the altar in the last image).   Lastly I could have sharpened harder maybe.

I am a little concerned that #1 I didn't notice these discrepancies until the par pointed them out to me and #2 the difficulty I am having in reproducing the par.  This has been a very valuable experience and more lessons learned.


====
Bill Theis


John Furnes
 

 

Mine is 816,

 

I assumed that this being morning, I could not go on with too much luminosity, and hence not so much colour (saturation). I am rather happy with my take on the picture, and think I got OK lighting, contrast and colour. There was of course the difficulties with the halos against the morning mist. The sky might be a bit too light, but, as said, this is morning light.

The par is too ‘lively’ for my eyes, and I think there would not be so much colour in ‘real life’. However, I have come to understand that the spectator does not know anything about the picture, and is more ready to accept my personal taste if it shows what he thinks is important, or if I lead him to see what I find important.

To me the main focus is the closest pagodas, with their colours, the grass and bushes as well, but also the light of the sky and the mist below the mountains.

 

This place needs to enter my ‘bucket list’ for sure.

 

John Furnes

 

 

 

 


Doug Schafer
 

I did a good look at all and have comments and a question.
I think many got it quite right or close. But there were 2 groups that seemed to either get the colors wrong (inconsistent with landscape scenes for foreground, mountains, sky) and some where luminosity was either too bright, a few rather dark with little/lost detail, and some with brightness inconsistent with dawn; often too bright or too dark all within the same image.

And my questions for Dan is about halos; especially around the pagoda bldg shapes against the background. Some halos are obviously too much, some entries seem to have no halos, and some have some halos but not obnoxious. It seems there are some halos in the original? And are there "natural" halos to leave in place?
Dan, can you point out which images have halos that are acceptable and perhaps give some descriptive words about acceptability level of halos. Clearly no halos is best. But are they always worth spending time to fix if not real obvious to intended viewers?
And, what is a good technique for removing them? I know how to manually paint them out. But maybe there is a channel blending or blend-if or masking method that is more precise and less time consuming than painting?

Doug Schafer


Hector Davila
 

I understand that it looks like a sunset photo, but
since it is a sunrise photo
(and it appears the sun is trying to burst through the cloudy day clouds)
it appears to me that the sun is above the horizon.

There is daylight even before the sun appears over the horizon.

I see a strong cast of sunlight through the center.

And sunlight muffled by clouds to cast a soft daylight shadow.

The daises are lit and drawing in the sun.

If not for the clouds, how much daylight would it be?

mines is #830
Hector Davila


jorgeparraphotography
 

Just about my own take, Hector.

Following Dan's approach to the choir ( I could not participate in that challenge) the double-lighting situation appeared to me to be happening in Burma in the same fashion here, forcing some decision-making process: I visualized a background getting the yellowish morning light  (as a standard back-lit situation), while the foreground is still lit mostly by  bluish skylight, again, this as a consequence of the sun being behind the pagodas.

 Add to this that the haze in distant mountains is always bluish no matter the position of the sun. I tried ( but went overboard with #824) to retain the "blueish-ness",  while allowing some neutral tone in the background( which could have been more yellowish, I guess), so the end result
is still too blue compared to the Par, which, IMHO, and as others have said, is all too yellow, the overall hue too unified all across the image and pointing to suggest tons of sunlight hitting everywhere. I had a version approaching this yellowish overall tone and left it behind as it looked unrealistic.
One of the PPW tools ( can't remember which) did a great job at reducing de haze in the background. Gotta take notes to use it in the future. 
--
Jorge Parra 
www.jJorgeParraPhotography.com
Miami


bill_iverson_washington
 

Mine is 829.  The foreground is lighter than most (and lighter than the much abused par), but not the lightest.  That was a matter of choice for all of us, of course.   But unlike the Beach at Sunset, where there was room for debate on several fronts how light the foreground pebble beach and photographer should be, in this case it seemed to me that without attractive foreground temples, this image is a nothing burger.  The beach at sunset had a more interesting sky, and more importantly the sweep of land and the bay of water offered interesting potential.  Here, in addition to a relatively unimposing sky, there was really no way to make the background mountain ridge and distant temples more than a supporting framework for the foreground temples.

          Focusing on the foreground temples and surrounding green areas, it seemed to me that the most important thing was to present attractive, realistic greens.  Bad greens are, IMHO, much worse than reddish browns that might be somewhat off. 

          But the temple colors are also important.  The brick temples at Bagan are a reddish brown at midday.  They can be much warmer in later afternoon light, but this is a morning picture.

          How light to make the temples is, again, a matter of choice.  But I think mine are a fair compromise making an attractive picture within the zone of reasonableness, without departing too much from the reality of an early morning picture.  As others have pointed out, the human eye is skilled at making such adjustments, but wouldn’t make an early morning register as midday.

          With the colors of the greenery and the temples dialed in to taste, and a decision made how light the temples should be, my sky was a bit magenta and dark at the top.  So I adjusted that to avoid allowing the sky to be too heavy and dominant.

          I adjusted the mountain ridge to stand out reasonably and, more importantly, to increase the presence of the temples in the distance -- the huge number of temples is a principal charm of Bagan.  Arguably, my mountain ridge could be made somewhat darker and more defined, but I didn’t like that appearance as well.

          All of this is more a matter of taste than technique.

 

Bill Iverson


Dan Margulis
 

On Tue, Mar 23, 2021 at 09:58 AM, jorgeparraphotography wrote:
compared to the Par, which, IMHO, and as others have said, is all too yellow, the overall hue too unified all across the image and pointing to suggest tons of sunlight hitting everywhere.
The poor par's ears must be burning as one member after another slings arrows at it.

I have therefore provided four alternate versions, and John Castronovo a fifth. My versions are the results of the blending demos that I'll be describing when I post comments on individual images, where I'm suggesting that certain submissions that aren't at present competitive to the par could be made so if allowed to borrow from others. Since mine each start with a different parent, they come in four different styles, and John's version is a fifth.

You are welcome to comment on these, it might give us all a better feel for what the group is after, and maybe even will provoke some to cut the melancholy existing par a break.

The five new versions have been added to the Photos folder as ##835-839, but I have packaged all of them, plus the existing par for comparison, in our Files section:
032321_Land-of-Pagodas_alt-pars.zip

I look forward to any comments, and will explain the steps to produce my four versions in the fullness of time.

Dan


Doug Schafer
 

Mine is # 801 and I was pleased to see mine aligned well with par #834 as I took an approach to get the best of cloudy misty foggy dawn yet be able to see the colorful foreground, against a contrasting background,  in sufficient but low light and sufficient contrast for 3D appearance of depth.
As usual this image, like many others, has a wide range of "to taste" for colors and luminance; so long as they are technically correct and/or meet the artistic desires. Artistic images or "color graded" images seem to be popular but I still like a lot of realism with a small artistic flair so an image is eye catching appealing, not bland and does not look "wrong" to me.
Again I'll bet we all learn some new stuff from this image exercise.

Doug Schafer


David Remington
 
Edited

My version is 815
 
I was thinking of this as sunset and sought to balance the warm light and the cool shade. I tried to preserve and enhance the sense of distance and scale. I reduced but kept the blue cast in the distant hills and shifted it to a more purple hue that better fits with the color of the sky and provides some aerial perspective.
 
I like my image. I can see that having more detail in the tops of the foreground temples would be an improvement, but I would not go as light as some. I find the top half of the par too light and the sky washed out. I had a lighter version and a much darker version. I liked the richness of the dark one and ended up blending it at about 20% into my final lighter version.
 
The discussion here about time of day and the appropriate the color to convey it has been interesting. I stared with the RAW file and used a "daylight" white point of around 5000k. My final version is a little warmer and more saturated but still pretty close to that color palette. So not far off from what was there at the time.
 
I checked the EXIF data after reading the time of day comments here. Creation date and time is recorded as November 11 at 15:33 (3:33PM) but the time offset is -4 making it eastern standard time. Converting that makes it about 2AM in Burma and sunrise is around 6AM in November. Long way of saying I don't think the time stamp helps.
 
Looking forward to Dan's take on this group of images.
 
David


Edward Bateman
 

Hello everyone-

Mine was number 811.

I get like this was fairly easy to make an OK image… but to go beyond that was a bigger challenge than I would have expected.

I felt that it contained two crucial balances that were hard to parse. One was luminosity… balancing the contrast differences between the sky and pagodas.

But by far, color and in particular saturation was the hardest balance for me to settle on. I think in large measure because the reds of the pagodas and greens of the foliage were mutually reinforcing… especially the wide spectrum of greens! I noticed that my eyes quickly adapted to high saturation - and only taking a break let me see how vivid my image had become.

And in evaluating my submission before sending, I noticed something else. I would occasionally toggle the “screen mode.” As I expected, it took on a very different appearance when surrounded by black. I remember posting way back when Adobe made the darker interface the norm that I was against that - at least for my own work. Typically, my images are printed on white paper, matted with white or on white walls. So I have always advocated the lighter interface. The differences between a dark and light surround can make a significant difference in how we see contrast and also saturation. And I still have a hard time getting my students to leave the dark side :)

So that lead me to wonder how many in this group are using the dark interface and if that is having a noticeable effect in how we are making and evaluating our corrections?

Thank you everyone… even though crazy busy, your comments have become an obsession with me and I really appreciate all of you!

Edward Bateman

Oh! And I have to say that I really respect Dan's writing - sometimes I will read a post several times just to enjoy his prose and the smart/articulate mind behind the words.


Christophe Potworowski
 

My image is #803, as usual it is too dark, even if I think the par is too well lit for a dawn image. But my image has also insufficient contrast both in the sky and in the greenery. Looking forward to further comments.

Christophe


James Gray
 

Maybe I can make the par feel better.  I like it and I do like the color.  In my opinion, it is better than many of the individual submissions. I also like the alt par versions and look forward to learning how they were created.  I like the color in all of them better than mine which is 819.  I like the contrast in 834, 835, 837, and 839.  As seems usual the differences between individually submitted versions seem to be large.  I have not spent as much time comparing individual versions as many have.  I do not agree with the notion that the processed version should be especially dark.  I think the image should be considered photographic art and should compete with other travel photography.

Jim Gray

On Tue, Mar 23, 2021 at 3:30 PM Dan Margulis via groups.io <dmargulis=aol.com@groups.io> wrote:
On Tue, Mar 23, 2021 at 09:58 AM, jorgeparraphotography wrote:
compared to the Par, which, IMHO, and as others have said, is all too yellow, the overall hue too unified all across the image and pointing to suggest tons of sunlight hitting everywhere.
The poor par's ears must be burning as one member after another slings arrows at it.

I have therefore provided four alternate versions, and John Castronovo a fifth. My versions are the results of the blending demos that I'll be describing when I post comments on individual images, where I'm suggesting that certain submissions that aren't at present competitive to the par could be made so if allowed to borrow from others. Since mine each start with a different parent, they come in four different styles, and John's version is a fifth.

You are welcome to comment on these, it might give us all a better feel for what the group is after, and maybe even will provoke some to cut the melancholy existing par a break.

I look forward to any comments, and will explain the steps to produce my four versions in the fullness of time.

Dan


KENT SOUTHERS
 

Mine is 831, and lands on the darker side of things compared to many.



Jim wrote:
I think the image should be considered photographic art and should compete with other travel photography.



I think that it becomes a matter of intent with regard to where you want to land any given image.  By that, I mean there are a myriad of reasons of intent for landing an image in a given manner.
For instance:

As seen (emulation of human vision / emulation of human memory)
As shot (as recorded against a given setting / film profile) ... Antelope Canyon plays more colorful as shot vs. as seen.
Neutral (render lighting WB to neutral)
Time of Day (render color cast lighting to warm / cool)
Time of Day (render tonality to light / dark)

Emphasis on lighting
Emphasis on subject / foreground / background / sky
Emphasis on contrast

Mood

Artistic Liberty / License



These are but a few of the more common decisions that precede the undertaking of beginning the process of landing the image where you want it, to convey what it is you desire to convey.
Sometimes we edit to bring an image to merely a "recording device" rendering.  Yet, other times we have plenty of decisions to make ... which then guide our choices of how to proceed.

I'm a BIG FAN of Dan ... and the principles of color theory and color correction.  Particularly the ability to find neutral and the deductive reasoning of what is NOT plausible vs. what is plausible.  Of course, artistic license can take plausibility and laugh at it ... or, it can be artistic AND plausible.  As creators, we have our choices available to us.  Kinda depends on what our mission / goal is for a given image.  Personally, I try to study the lighting first ... and am pleased to hear others consider the lighting / time of day perspective also.  As technicians ... sometimes our mission is to be accurate / objective.  Other times it involves harnessing the subjective / emotive.

So, for mine ... it was an effort to emulate "Time of Day" and what the "presence" might be like to be there in person at that time of day, in those lighting conditions.  So, the decision is one of trying to present the conditions ... or, trying to offset / override them into something else.

Always a decision, decision, decision ... before execution.

BTW, this was a tricky one for me.  Given the choice of foreground vs. sunset colors, I favored retaining the sky (sunset / sunrise).  A case could easily be made for favoring the foreground and letting the sky go.  I particularly appreciated the comment about exif data time of capture, trying to assess time of day (time zones, etc.) to find which to emulate.  But, without clear direction ... we are left to our own derived intent, based on certain assumptions and goals for a given image.

After seeing the others, I took a stab at lifting my foreground tonality a bit more than I had.  That said, the difference between a mood / presence version vs. a bright "travel poster" shot can be worlds apart ... based on intent.



Kent Southers, CMRP
southers3@...



From: colortheory@groups.io <colortheory@groups.io> on behalf of James Gray <James@...>
Sent: Tuesday, March 23, 2021 8:41 PM
To: colortheory@groups.io <colortheory@groups.io>
Subject: Re: [colortheory] Land of Pagodas: Results
 
Maybe I can make the par feel better.  I like it and I do like the color.  In my opinion, it is better than many of the individual submissions. I also like the alt par versions and look forward to learning how they were created.  I like the color in all of them better than mine which is 819.  I like the contrast in 834, 835, 837, and 839.  As seems usual the differences between individually submitted versions seem to be large.  I have not spent as much time comparing individual versions as many have.  I do not agree with the notion that the processed version should be especially dark.  I think the image should be considered photographic art and should compete with other travel photography.

Jim Gray

On Tue, Mar 23, 2021 at 3:30 PM Dan Margulis via groups.io <dmargulis=aol.com@groups.io> wrote:
On Tue, Mar 23, 2021 at 09:58 AM, jorgeparraphotography wrote:
compared to the Par, which, IMHO, and as others have said, is all too yellow, the overall hue too unified all across the image and pointing to suggest tons of sunlight hitting everywhere.
The poor par's ears must be burning as one member after another slings arrows at it.

I have therefore provided four alternate versions, and John Castronovo a fifth. My versions are the results of the blending demos that I'll be describing when I post comments on individual images, where I'm suggesting that certain submissions that aren't at present competitive to the par could be made so if allowed to borrow from others. Since mine each start with a different parent, they come in four different styles, and John's version is a fifth.

You are welcome to comment on these, it might give us all a better feel for what the group is after, and maybe even will provoke some to cut the melancholy existing par a break.

I look forward to any comments, and will explain the steps to produce my four versions in the fullness of time.

Dan