Topics

Concert on beach: Dan's comments


Dan Margulis
 

You won't read much about Photoshop technique in this or the next post because this picture doesn't require much technical skill. Instead, it mostly depends on judgment of how to present this scene attractively, particularly granted the overcast conditions.

Some decisions are pure personal preference, some are personal but questionable, and some are outright wrong.

The pure personal preference is how much, if any, additional color to work in. Some people are quite conservative, almost austere, while others warm things up considerably. As we saw in the other thread, voters split on which approach to prefer.

Any move toward coldness, however, is a big mistake. A disturbing number of entries have cold casts. The tarmac, and especially the sand, cannot tend toward blue and/or green. Also, we know that the building is gray, not because such buildings are never any other color, but because it of the law of coincidences: it carries signs that are probably neutral themselves, but in principle are independent of the building's own color. If the signs and the building all measure the same thing (and they do), it can't be anything but gray. The best way to find out these is by setting the Info palette to read LAB values and mouse over these areas. You shouldn’t be seeing negative (cold) numbers in the A or B in any of these areas. This isn’t personal preference, either. Nobody is going to favor a cold version over a similar neutral or warm one.

In the Mantillas exercise we definitely had to find ways to improve the original or our work would be considered a failure. Here, it's more a question of not doing anything foolish. Accordingly, we had a much higher percentage of good entries than in our first two studies. In those we got three or four excellent entries but I had to select some that I felt were undeserving for the par version.

No such problem with this set. I think that around ten could reasonably have been chosen. I decided to be ecumenical, choosing Jorge's #309, which is middle-of-the-road, along with one lighter, one darker, one less colorful and one more colorful version.

I will say that I don't like the lighter versions. When detail is brought out in the stage it makes clear that there is strong floodlighting. That would hardly be needed if the day was as bright as some have it.

Which brings up my main takeaway from this exercise, something that I should have thought of while preparing my own version. The stage is at the very center of this scene. I realize that this is a beach town but I completely disagree with those who thought it unimportant. Quite the contrary, I'd play it up, to suggest how exciting the cultural activities are. It has nothing to do with the specific performers.

The question is how to do that and the answer is shockingly easy. If I was working on the Hotel Lobby and felt that your plant, or that your left-hand woman in the Mantillas, were excellent, I would have a hard time making a selection of them and even if I could it would be impossible to just paste them into my own version. There'd be a visual mismatch that was obvious to all.

But with this image I could literally cut out somebody else's stage and put it into mine. Real easy to select, too. It's got its own strong light source that's independent of the rest of the scene, so it doesn't have to match the rest of the scene for color or contrast at all, it just has to look good. So, I'd recommend setting up a separate procedure for the stage. And I would make sure that the yellow banner looked good; you can be sure the client wants to be able to admire it, but a lot of us did a poor job with it.

I'll comment on individual images in a second post, but I remind everyone that the summary of steps is important, and any statements as to your objectives are welcome. This particular discussion won't cover many of the steps because they're fairly obvious, but future exercises no doubt will.

Dan


Dan Margulis
 

Note: three of these versions come from the MIT study, where sharpening wasn’t permitted. To make them more comparable to our work, I’ve sharpened them myself, but haven’t changed them in any other way.

301 This person states that he did not wish to make a sunny day out of the scene. I believe that the client would prefer something sunnier than the actual reality, but this version is too cold anyway. First, Auto Tone should have been done at the end. This version is too flat. It also has a cold cast that can be measured. The building itself could conceivably be somewhat blue, but it’s hard to believe that the soffit, the N.H. State Park signs, and the white signs to their outside, could all measure as blue/cyan.

Even if no Info palette were available, though, the cast would still be noticeable. Recall the law of coincidences in the Mantillas exercise. Dresses like the ones the women wear there are usually black, but other colors are possible. However, the more women who wear the same color, the less likely that it could be anything other than black.

Here, too, a lot of people are wearing shirts that one would expect to be white. It is of course possible that one or two of them are slightly blue, but not as many as we see here.

302 This one doesn’t have as bad of a cold cast as #301, but it’s there anyway, measurable in the shirts and particularly in the building, which is slightly negative in the A channel. Admittedly it is OK not to make this a sunny day, but surely the client would prefer something neutral or slightly warm.

303 This came in late and at first I pooh-poohed it, the white shirts are blown out, the whole thing is too light, and the sunlight effect (done with a rendering model) at left of the stage is overexaggerated. But before closing it, I thought about that sunlight effect and I said to myself, “I bet that if I blend about 10% of this mess into my own version, it will improve things.”

I didn’t find out whether I was right, because I had a different inspiration. This is one of about seven versions where a good artistic idea was taken to such an extreme that I thought the version as a whole wasn’t good. Each of them, like this one, might be used as an auxiliary—a blend at a low opacity into another version that lacks the particular effect. So I further wondered, what would happen if I blended five of these misfits into their own par version? Sure, this sunlight effect is way overdone, but if it is taken down by four-fifths it’s likely desirable. I therefore created such a thing, and can tell you that #303, therefore was Chosen for the auxiliary par.

304 Chosen for the par version. This is the conservative representative in the par suite, also lighter than most. Because of the lack of a cast, and the better portrayal of the stage, this one does the things that #302 does not. The person noted that during the process, the reds seemed to go wild, but he decided to keep them, a good idea when the overall image seems on the gray side.

305 Too light for my taste. Nice blue water. The notes say he tried to make it look like daytime. The stage lighting doesn’t make sense and the shirts are blown out.

306 Here’s the first one that strives for warmth. Comparing it to #305 makes this clear. I have no problem as such with this version and could well have chosen it as one of the five par parents. However, it’s not quite as good as the par itself, where the yellow banner is better defined, the audience a bit stronger, and the water a nicer blue. This person lightened the breaking surf, a good touch.

307 The darkest version of all. Nothing wrong with that concept, but it’s unnecessarily flat, as can be seen by running Auto Tone. The person was concerned about making the image too warm in view of his dark treatment, and responded to that problem well. However, with the overall scene this dark, he missed the opportunity of adding drama by isolating the stage and emphasizing the lighting of the musicians

308 This is the auxiliary par version, a straight average of the five indicated here. We gave it a full discussion in a separate thread.

309 Chosen for the par version. My personal favorite. Jorge Parra has the foreground dark enough to make us accept it’s a late afternoon scene. The blue ocean is well done, the stage is properly lit, and the yellow banner stands out. Here’s his description, with his request at the end,

3 Variants were produced in Lightroom. One has stronger color temp to make it bluer.  A neutral on and a warm one, all 3 mixed in different proportions

The stamp of this mix was treated with PPW Bigger hammer to add additional saturation and vibrance to the colors. A second stamp was created with PPW -MM . A third stamp only had contrast adjustments.

This new 3 stamps were mixed again and the resulting file got an extra contrast punch in curve adjustment layer. Little or no sharpening was added.

Total time working on the file : 8 minutes. Could have been a little less I did not embark in smoothing the color fringe in the roof of the house.

BTW, I wanted to comment that there should be a requirement or at least suggestion, for people to submit the time they worked on the files. I see reports of sooo many adjustments applied that it seems that some few worked for hours on the image , and I feel so in tune with your observations about spending the least possible time working on the adjustments to a decent working file

310 Chosen for the auxiliary par. This person did sensitive work to the stage area to emphasize the lighting, and the foreground is very appropriate for the sunset scene he envisioned (he read the time-of-day stamp in the metadata). He then, however, decided to emulate sunlight in the background, fell in love with the concept, and ruined the image. And yet the idea of yellowing up the background is a good one—enough so to make it a contributor to my alternate par version.

311 This person knows the area, as in his youth he used to work on this beach. However, he’s given us basically a grayscale image. Nobody seems to be wearing colorful clothing. And surely the sponsors want the banner to be a stronger yellow than what we see here.

In the background, the beach is magenta, I’m typically reading 7a2b or so. Those readings should be reversed. Browns as light as on this beach are always skewed toward yellow. Also, having the ocean this blue might make sense in a more colorful version, but it’s unrealistic in this one.

312 This version has apparently been resized (or had a lens correction) and so can’t be compared directly with others. Also, this is the one that, although a JPEG, opened by default in Camera Raw, meaning that I had to copy it to a new file and save a fresh copy. I ask everyone to double-click their files before submitting to make sure they open properly (if possible). Colorwise, this version has a lot in common with #322, so the same comments would apply.

313 A strong cyan cast, measured in the building, the signs on the front of it, and even the concrete. The sand in the background reads negative-A in some areas, which is impossible, it can’t be more green than it is magenta.

314 Chosen for the par version. The eight-minute workflow described in the note to #309 obviously doesn’t apply to this one, where the person writes,

After battling reds and blues for 6 days I apparently had a day of rest on the 7th day.

and when he started blending together all of his various versions,

It seemed like I got sucked into a vortex of brighter and warmer, I just couldn’t buy cooler and dimmer for an image meant to entire visitors and tourists.

Well, it is very light, and it isn’t growing on me, I think I was too easily impressed by the stage, which is presented well in terms of color but is too dark. I think it would be helped if the sand was made slightly darker and more colorful, to differentiate it from the concrete foreground. The water seems to be handled well. Another comment about this version is in #319, and a way to improve it big-time is in #329.

315 Penny wise, pound foolish. Paste this version in Color mode over the par and it arguably makes an improvement. But the lack of range kills it. All that needs to be done is, before submitting, run Auto Tone. That doesn’t make it better, it will need to be cancelled, but it certainly indicates that this job is not complete. So close, and yet so far.

316 Chosen for the auxiliary par. With the emphasis on channel blending in the Hotel Lobby and Mantillas exercises it is somewhat surprising that more people didn’t try blending the red into the blue. This guy did, which accounts for the better definition on stage, the lighter fleshtones in the foreground, and the lighter banner. The sharpening is acceptable to me but many would find it heavy-handed.  This may have the best stage of anyone and the foreground is also excellent. I downgraded it because there’s something weird-looking about the banner, and because the colorizing of the water seems too obvious. But the strengths of this version make it a clear choice for the auxiliary par.

317 The foreground is pretty good but elsewhere the color gets out of hand. The building measures greenish-yellow, most unlikely, and its soffit is purple. The water is also purple, around 10a(10)b, which can hardly be correct.

318 This is perhaps the oversaturated counterpart of the grayscale-like #311. In both cases the solution is the same: make the stage more interesting. Here it’s too dark in relation to the rest of the scene. The building is blue, which is probably wrong, but it’s better than some of the others. Beet-red fleshtones suggest that colors need to be toned down in the foreground, but the beach and the water work nicely.

319 This was the best effort of the five MIT retouchers. It isn’t competitive with our versions, although there’s nothing overtly disqualifying about it. This person offers a very light presentation, making it similar to #314 and #330. But comparing these to it makes clear how important it is to darken the midtone.

320 An overly light treatment similar to that of #319, but this one has a much better presentation of the stage. It should probably have been chosen for the auxiliary par on that basis.

321 On reviewing, I like this one better than I did at first. I was troubled by the blueness of the building. This person took the position that the foreground tarmac was neutral, which may not be true in view of the sunlight. 

322 Chosen for the auxiliary par. This one makes the auxiliary team because of the strong contrast between the concrete seating area and the sand. Many people have difficulty distinguishing the two because the overcast day suppresses the tan color of the beach.

If the very gray concrete is a good thing, then why isn’t this effort satisfactory as a whole? That’s easy: the stage should not have been neutralized. The client is likely to be particularly upset about the loss of color in their yellow banner.

Fortunately the fix is equally easy. The viewer understands that the stage is harshly lit, so it doesn’t have to match the rest of the image. We can just pick up somebody else’s and paste it into this one. I did so with #335 (several others would have done as well). I did turn the opacity down to 80%, but I now think that this is among the very best versions we have.

323 An excellent version prepared in a straightforward way, mostly in the raw module. The drawback is that the person made IMHO a poor artistic decision. He did something that

…made bandstand players slightly less contrasty, since the presumed object was to market the bandstand and beach, not the band.

I say that making it clear that spotlights are shining on the stage may have the side effect of highlighting the players but its main purpose is to make the scene look real. If, for example, we put #336 on top of this one, but set to Luminosity mode so as not to pick up any of its color, this one will be improved and the main reason is the added detail on stage. Nothing about such a treatment makes me more interested in the players themselves.

324 This is the version I did in 2017 when comparing my work to the MIT folk. Selections weren’t permitted, so I could not make a major change in the water as so many of us did. Considering that its competition is #332 I’m not offering any apologies.

325 Although there’s good detail in the foreground and in the water, possibly the result of Bigger Hammer, on the whole the presentation is too gloomy. Also, the sunset effect in the water is handled nicely from the retouching POV but not so much from the common-sense one. We’re looking east into the Atlantic Ocean now. It might be portrayed accurately if this were a sunrise, but most days the sun sets in the west.

326 This person’s notes state that understood that the sun doesn’t belong where he put it, but decided to put it there anyway for aesthetic purposes. That might work as a wall hanging if we don’t happen to know where the shot was taken, but the Chamber of Commerce does know and likely won’t approve. The greenish-yellow backdrop of the stage is disagreeable, but I like the rest of the treatment.

327 Another reasonable version for those looking to portray the scene as fairly light. Gerald Bakker explains,

I prefer a lighter version and warmer colors.

He therefore started off with a golden Photo Filter, which sounds like a good idea. It may have backfired, however, because the basic problem is that the version lacks bright colors. Yet most logical solutions, such as the Color Boost action, result in the image becoming too overtly yellow.

My recommendation would be to darken the midtone for added weight, to go into Color Boost emphasizing the A channel significantly more than the B, and then cutting the boost in half for everything except the stage. See #329 for a further suggestion on how to improve it.

328 This is a night shot, almost as dark as #307. Fair enough, but we would then expect very strong lighting on the stage.

329 This person took a straightforward approach. At first it seems like it’s similar to a few others where the idea was something relatively light and without extravagant colors. And this one has the plus of an excellent treatment of the stage and performers.

However, on closer examination it has a cyan cast. It’s measurable in the building and its signs, and the beach itself measures A-negative in certain parts (reminiscent of #313). That’s disagreeable, but before hitting the Delete key, let’s consider two other versions that are also light.

In #327 Gerald added a golden tint. I thought it wasn’t colorful enough, but there’s no denying that it’s pleasant, and surely better than having a cyan cast. I thought also that it lacked sufficient weight, so we have an easy solution. Put one version on top of the other, pairing Gerald’s color with the luminosity of this one. Presto, a version better than either parent.

Far more dramatic, though: in #314 the treatment is equally light, but excellent color, and an overly dark, if colorful, presentation of the stage. So we put #314 on top of this one in Color mode. The result is the best I’ve seen. I’ve compared it to the ones named in the other thread. I clearly prefer it to #308, #309, #336, and the par.

330 Chosen for the par version. This is the darkest of the par suite. It’s somewhat comparable to #322, but the brighter colors make it preferable.

331 John Furnes, dissatisfied, writes

After seeing all entrants and the par-version, I realise that I have been too moody. The all-over impression is grey and lacklustre. There is a violet cast over it, and all the warmth is gone, even though at NH in July, the sun sets around the time this picture was take, and there should be quite some leftovers of the sunshine.
I am not happy with it.
I also think that the par-version is a bit blue.

So, what’s the cure for boredom, granted that he’s looking for something fairly conservative? Easy enough, just find a way to shoehorn more color in. Put an extremely saturated layer on top, with a black layer mask. Then paint bright colors into the yellow banner. Nobody can object to that. Also, go over the performers on stage, and find a few people with somewhat colorful shirts in the audience. That should do it. You may recall that in #304 the person reported that the reds, if unnaturally vivid, helped the scene. 
 
332 The average result from the five MIT retouchers. Rather cold, as indicated by comparison with my three-minute 2017 version (#324).

333 One of the best versions otherwise, but wrecked by unbelievably bright cyan water. Putting some coolness into the water is a good idea in principle, it sets off the foreground nicely, but this is about five times too much. 

334 I underrated this version and if I were doing it again, I’d pick it for the par. I think my initial reasoning was, although the color is correct, it’s fairly conservative and I didn’t want to have the par contain more than one such treatment. And I chose #304 as the representative of that school. Looking at the two again, I think I prefer this one for its natural-looking ocean and the lovely treatment of the yellow banner.

335 Chosen for the auxiliary par. The person worked hard on this one, selecting five different areas for individual correction. He felt that the eye should be directed toward the stage and gives us one of the best renditions of it. Taken overall, it doesn’t work because the strong yellow cast in the foreground contradicts the deep blue of the ocean. Yet these colors, in modest amounts, are effective.

336 Chosen for the par version. Along the same color lines as #335 but not as overdone. Among our par components, it’s the aggressive counterpart of the more conservative #305. We discussed it in a separate thread, so I won’t comment further here.

337 My entrant for this time. Probably most comparable to #330. I wish I had done more with the stage, and a better job with the water.

338 Another one where the sun sets in the east. There’s a magenta cast as measured in the sand and the tarmac.

339 The par version.


Kenneth Harris
 

I agree with Jorge Parra about noting the time put into working the image, and even estimating the time it will take before you start. Doing so helps you tune your method. This one went super fast for me, just over ten minutes (304); I had in mind to ape Stephen Shore, got there, and saved it. But on this image, you can do that. This week's study is a different matter – stuff is buried, and you have to excavate and see what it is, possibly going the wrong way.
Ken Harris


Adrian Thompson
 

325 Although there’s good detail in the foreground and in the water, possibly the result of Bigger Hammer, on the whole the presentation is too gloomy. Also, the sunset effect in the water is handled nicely from the retouching POV but not so much from the common-sense one. We’re looking east into the Atlantic Ocean now. It might be portrayed accurately if this were a sunrise, but most days the sun sets in the west.

 

This was mine – the last part I’m going to blame on being in the UK and not thinking to look on a map!

 

 

 

From: Dan Margulis via groups.io
Sent: 18 February 2021 19:51
To: colortheory@groups.io
Subject: Re: [colortheory] Concert on beach: Dan's comments

 

Note: three of these versions come from the MIT study, where sharpening wasn’t permitted. To make them more comparable to our work, I’ve sharpened them myself, but haven’t changed them in any other way.

 

301 This person states that he did not wish to make a sunny day out of the scene. I believe that the client would prefer something sunnier than the actual reality, but this version is too cold anyway. First, Auto Tone should have been done at the end. This version is too flat. It also has a cold cast that can be measured. The building itself could conceivably be somewhat blue, but it’s hard to believe that the soffit, the N.H. State Park signs, and the white signs to their outside, could all measure as blue/cyan.

 

Even if no Info palette were available, though, the cast would still be noticeable. Recall the law of coincidences in the Mantillas exercise. Dresses like the ones the women wear there are usually black, but other colors are possible. However, the more women who wear the same color, the less likely that it could be anything other than black.

 

Here, too, a lot of people are wearing shirts that one would expect to be white. It is of course possible that one or two of them are slightly blue, but not as many as we see here.

 

302 This one doesn’t have as bad of a cold cast as #301, but it’s there anyway, measurable in the shirts and particularly in the building, which is slightly negative in the A channel. Admittedly it is OK not to make this a sunny day, but surely the client would prefer something neutral or slightly warm.

 

303 This came in late and at first I pooh-poohed it, the white shirts are blown out, the whole thing is too light, and the sunlight effect (done with a rendering model) at left of the stage is overexaggerated. But before closing it, I thought about that sunlight effect and I said to myself, “I bet that if I blend about 10% of this mess into my own version, it will improve things.”

 

I didn’t find out whether I was right, because I had a different inspiration. This is one of about seven versions where a good artistic idea was taken to such an extreme that I thought the version as a whole wasn’t good. Each of them, like this one, might be used as an auxiliary—a blend at a low opacity into another version that lacks the particular effect. So I further wondered, what would happen if I blended five of these misfits into their own par version? Sure, this sunlight effect is way overdone, but if it is taken down by four-fifths it’s likely desirable. I therefore created such a thing, and can tell you that #303, therefore was Chosen for the auxiliary par.

 

304 Chosen for the par version. This is the conservative representative in the par suite, also lighter than most. Because of the lack of a cast, and the better portrayal of the stage, this one does the things that #302 does not. The person noted that during the process, the reds seemed to go wild, but he decided to keep them, a good idea when the overall image seems on the gray side.

 

305 Too light for my taste. Nice blue water. The notes say he tried to make it look like daytime. The stage lighting doesn’t make sense and the shirts are blown out.

 

306 Here’s the first one that strives for warmth. Comparing it to #305 makes this clear. I have no problem as such with this version and could well have chosen it as one of the five par parents. However, it’s not quite as good as the par itself, where the yellow banner is better defined, the audience a bit stronger, and the water a nicer blue. This person lightened the breaking surf, a good touch.

 

307 The darkest version of all. Nothing wrong with that concept, but it’s unnecessarily flat, as can be seen by running Auto Tone. The person was concerned about making the image too warm in view of his dark treatment, and responded to that problem well. However, with the overall scene this dark, he missed the opportunity of adding drama by isolating the stage and emphasizing the lighting of the musicians

 

308 This is the auxiliary par version, a straight average of the five indicated here. We gave it a full discussion in a separate thread.

 

309 Chosen for the par version. My personal favorite. Jorge Parra has the foreground dark enough to make us accept it’s a late afternoon scene. The blue ocean is well done, the stage is properly lit, and the yellow banner stands out. Here’s his description, with his request at the end,

 

3 Variants were produced in Lightroom. One has stronger color temp to make it bluer.  A neutral on and a warm one, all 3 mixed in different proportions

 

The stamp of this mix was treated with PPW Bigger hammer to add additional saturation and vibrance to the colors. A second stamp was created with PPW -MM . A third stamp only had contrast adjustments.

 

This new 3 stamps were mixed again and the resulting file got an extra contrast punch in curve adjustment layer. Little or no sharpening was added.

 

Total time working on the file : 8 minutes. Could have been a little less I did not embark in smoothing the color fringe in the roof of the house.

 

BTW, I wanted to comment that there should be a requirement or at least suggestion, for people to submit the time they worked on the files. I see reports of sooo many adjustments applied that it seems that some few worked for hours on the image , and I feel so in tune with your observations about spending the least possible time working on the adjustments to a decent working file

 

310 Chosen for the auxiliary par. This person did sensitive work to the stage area to emphasize the lighting, and the foreground is very appropriate for the sunset scene he envisioned (he read the time-of-day stamp in the metadata). He then, however, decided to emulate sunlight in the background, fell in love with the concept, and ruined the image. And yet the idea of yellowing up the background is a good one—enough so to make it a contributor to my alternate par version.

 

311 This person knows the area, as in his youth he used to work on this beach. However, he’s given us basically a grayscale image. Nobody seems to be wearing colorful clothing. And surely the sponsors want the banner to be a stronger yellow than what we see here.

 

In the background, the beach is magenta, I’m typically reading 7a2b or so. Those readings should be reversed. Browns as light as on this beach are always skewed toward yellow. Also, having the ocean this blue might make sense in a more colorful version, but it’s unrealistic in this one.

 

312 This version has apparently been resized (or had a lens correction) and so can’t be compared directly with others. Also, this is the one that, although a JPEG, opened by default in Camera Raw, meaning that I had to copy it to a new file and save a fresh copy. I ask everyone to double-click their files before submitting to make sure they open properly (if possible). Colorwise, this version has a lot in common with #322, so the same comments would apply.

 

313 A strong cyan cast, measured in the building, the signs on the front of it, and even the concrete. The sand in the background reads negative-A in some areas, which is impossible, it can’t be more green than it is magenta.

 

314 Chosen for the par version. The eight-minute workflow described in the note to #309 obviously doesn’t apply to this one, where the person writes,

 

After battling reds and blues for 6 days I apparently had a day of rest on the 7th day.

 

and when he started blending together all of his various versions,

 

It seemed like I got sucked into a vortex of brighter and warmer, I just couldn’t buy cooler and dimmer for an image meant to entire visitors and tourists.

 

Well, it is very light, and it isn’t growing on me, I think I was too easily impressed by the stage, which is presented well in terms of color but is too dark. I think it would be helped if the sand was made slightly darker and more colorful, to differentiate it from the concrete foreground. The water seems to be handled well. Another comment about this version is in #319, and a way to improve it big-time is in #329.

 

315 Penny wise, pound foolish. Paste this version in Color mode over the par and it arguably makes an improvement. But the lack of range kills it. All that needs to be done is, before submitting, run Auto Tone. That doesn’t make it better, it will need to be cancelled, but it certainly indicates that this job is not complete. So close, and yet so far.

 

316 Chosen for the auxiliary par. With the emphasis on channel blending in the Hotel Lobby and Mantillas exercises it is somewhat surprising that more people didn’t try blending the red into the blue. This guy did, which accounts for the better definition on stage, the lighter fleshtones in the foreground, and the lighter banner. The sharpening is acceptable to me but many would find it heavy-handed.  This may have the best stage of anyone and the foreground is also excellent. I downgraded it because there’s something weird-looking about the banner, and because the colorizing of the water seems too obvious. But the strengths of this version make it a clear choice for the auxiliary par.

 

317 The foreground is pretty good but elsewhere the color gets out of hand. The building measures greenish-yellow, most unlikely, and its soffit is purple. The water is also purple, around 10a(10)b, which can hardly be correct.

 

318 This is perhaps the oversaturated counterpart of the grayscale-like #311. In both cases the solution is the same: make the stage more interesting. Here it’s too dark in relation to the rest of the scene. The building is blue, which is probably wrong, but it’s better than some of the others. Beet-red fleshtones suggest that colors need to be toned down in the foreground, but the beach and the water work nicely.

 

319 This was the best effort of the five MIT retouchers. It isn’t competitive with our versions, although there’s nothing overtly disqualifying about it. This person offers a very light presentation, making it similar to #314 and #330. But comparing these to it makes clear how important it is to darken the midtone.

 

320 An overly light treatment similar to that of #319, but this one has a much better presentation of the stage. It should probably have been chosen for the auxiliary par on that basis.

 

321 On reviewing, I like this one better than I did at first. I was troubled by the blueness of the building. This person took the position that the foreground tarmac was neutral, which may not be true in view of the sunlight. 

 

322 Chosen for the auxiliary par. This one makes the auxiliary team because of the strong contrast between the concrete seating area and the sand. Many people have difficulty distinguishing the two because the overcast day suppresses the tan color of the beach.

 

If the very gray concrete is a good thing, then why isn’t this effort satisfactory as a whole? That’s easy: the stage should not have been neutralized. The client is likely to be particularly upset about the loss of color in their yellow banner.

 

Fortunately the fix is equally easy. The viewer understands that the stage is harshly lit, so it doesn’t have to match the rest of the image. We can just pick up somebody else’s and paste it into this one. I did so with #335 (several others would have done as well). I did turn the opacity down to 80%, but I now think that this is among the very best versions we have.

 

323 An excellent version prepared in a straightforward way, mostly in the raw module. The drawback is that the person made IMHO a poor artistic decision. He did something that

 

…made bandstand players slightly less contrasty, since the presumed object was to market the bandstand and beach, not the band.

 

I say that making it clear that spotlights are shining on the stage may have the side effect of highlighting the players but its main purpose is to make the scene look real. If, for example, we put #336 on top of this one, but set to Luminosity mode so as not to pick up any of its color, this one will be improved and the main reason is the added detail on stage. Nothing about such a treatment makes me more interested in the players themselves.

 

324 This is the version I did in 2017 when comparing my work to the MIT folk. Selections weren’t permitted, so I could not make a major change in the water as so many of us did. Considering that its competition is #332 I’m not offering any apologies.

 

325 Although there’s good detail in the foreground and in the water, possibly the result of Bigger Hammer, on the whole the presentation is too gloomy. Also, the sunset effect in the water is handled nicely from the retouching POV but not so much from the common-sense one. We’re looking east into the Atlantic Ocean now. It might be portrayed accurately if this were a sunrise, but most days the sun sets in the west.

 

326 This person’s notes state that understood that the sun doesn’t belong where he put it, but decided to put it there anyway for aesthetic purposes. That might work as a wall hanging if we don’t happen to know where the shot was taken, but the Chamber of Commerce does know and likely won’t approve. The greenish-yellow backdrop of the stage is disagreeable, but I like the rest of the treatment.

 

327 Another reasonable version for those looking to portray the scene as fairly light. Gerald Bakker explains,

 

I prefer a lighter version and warmer colors.

 

He therefore started off with a golden Photo Filter, which sounds like a good idea. It may have backfired, however, because the basic problem is that the version lacks bright colors. Yet most logical solutions, such as the Color Boost action, result in the image becoming too overtly yellow.

 

My recommendation would be to darken the midtone for added weight, to go into Color Boost emphasizing the A channel significantly more than the B, and then cutting the boost in half for everything except the stage. See #329 for a further suggestion on how to improve it.

 

328 This is a night shot, almost as dark as #307. Fair enough, but we would then expect very strong lighting on the stage.

 

329 This person took a straightforward approach. At first it seems like it’s similar to a few others where the idea was something relatively light and without extravagant colors. And this one has the plus of an excellent treatment of the stage and performers.

 

However, on closer examination it has a cyan cast. It’s measurable in the building and its signs, and the beach itself measures A-negative in certain parts (reminiscent of #313). That’s disagreeable, but before hitting the Delete key, let’s consider two other versions that are also light.

 

In #327 Gerald added a golden tint. I thought it wasn’t colorful enough, but there’s no denying that it’s pleasant, and surely better than having a cyan cast. I thought also that it lacked sufficient weight, so we have an easy solution. Put one version on top of the other, pairing Gerald’s color with the luminosity of this one. Presto, a version better than either parent.

 

Far more dramatic, though: in #314 the treatment is equally light, but excellent color, and an overly dark, if colorful, presentation of the stage. So we put #314 on top of this one in Color mode. The result is the best I’ve seen. I’ve compared it to the ones named in the other thread. I clearly prefer it to #308, #309, #336, and the par.

 

330 Chosen for the par version. This is the darkest of the par suite. It’s somewhat comparable to #322, but the brighter colors make it preferable.

 

331 John Furnes, dissatisfied, writes

 

After seeing all entrants and the par-version, I realise that I have been too moody. The all-over impression is grey and lacklustre. There is a violet cast over it, and all the warmth is gone, even though at NH in July, the sun sets around the time this picture was take, and there should be quite some leftovers of the sunshine.

I am not happy with it.

I also think that the par-version is a bit blue.

 

So, what’s the cure for boredom, granted that he’s looking for something fairly conservative? Easy enough, just find a way to shoehorn more color in. Put an extremely saturated layer on top, with a black layer mask. Then paint bright colors into the yellow banner. Nobody can object to that. Also, go over the performers on stage, and find a few people with somewhat colorful shirts in the audience. That should do it. You may recall that in #304 the person reported that the reds, if unnaturally vivid, helped the scene. 

 

332 The average result from the five MIT retouchers. Rather cold, as indicated by comparison with my three-minute 2017 version (#324).

 

333 One of the best versions otherwise, but wrecked by unbelievably bright cyan water. Putting some coolness into the water is a good idea in principle, it sets off the foreground nicely, but this is about five times too much. 

 

334 I underrated this version and if I were doing it again, I’d pick it for the par. I think my initial reasoning was, although the color is correct, it’s fairly conservative and I didn’t want to have the par contain more than one such treatment. And I chose #304 as the representative of that school. Looking at the two again, I think I prefer this one for its natural-looking ocean and the lovely treatment of the yellow banner.

 

335 Chosen for the auxiliary par. The person worked hard on this one, selecting five different areas for individual correction. He felt that the eye should be directed toward the stage and gives us one of the best renditions of it. Taken overall, it doesn’t work because the strong yellow cast in the foreground contradicts the deep blue of the ocean. Yet these colors, in modest amounts, are effective.

 

336 Chosen for the par version. Along the same color lines as #335 but not as overdone. Among our par components, it’s the aggressive counterpart of the more conservative #305. We discussed it in a separate thread, so I won’t comment further here.

 

337 My entrant for this time. Probably most comparable to #330. I wish I had done more with the stage, and a better job with the water.

 

338 Another one where the sun sets in the east. There’s a magenta cast as measured in the sand and the tarmac.

 

339 The par version.    

 

 


john c.
 

Well... there's time spent looking at it and considering what it should and could look like, time spent deciding what's possible, and then there's time actually doing it, and finally there's time to evaluate if what was done was the right thing to do after all. It all adds up. The time to actually do the work is probably the least. So are we talking about the total or the time we spend doing the work?

john castronovo

-----Original Message-----
From: Kenneth Harris
Sent: Thursday, February 18, 2021 3:09 PM
To: colortheory@groups.io
Subject: Re: [colortheory] Concert on beach: Dan's comments

I agree with Jorge Parra about noting the time put into working the image, and even estimating the time it will take before you start. Doing so helps you tune your method. This one went super fast for me, just over ten minutes (304); I had in mind to ape Stephen Shore, got there, and saved it. But on this image, you can do that. This week's study is a different matter – stuff is buried, and you have to excavate and see what it is, possibly going the wrong way.
Ken Harris


Kenneth Harris
 

I'd say working time. It's an open book test. I usually come up with my goal and my plan on any project within ten minutes, jotting down my notes which techniques I plan to use. Key for me is to verbalize what's good and bad about the picture or group. Your business likely allows for more rumination.

Judging by the time between entry closing and the posting of all the images on Mondays, seems like Dan is pretty fast at this kind of thing.

Ken Harris


James Gray
 

Mine is 316.  I believe the reason the banner looks a little weird is because of the use of ALCE 3.0.  This is from my notes:

Ran ALCE 3.0 with radii 5, 90, & 250.  Then put a mask on the neutral areas.

Clearly, I should have also masked the banner as well.

About the sea- from my notes

In my opinion, precise color is not as important as adding some snap and color variation to this image.  I think making the sea and sky a little bit orange adds some drama to the image that it needs.

I was thinking sunset-type light. Sometimes at sunset, the yellows, reds, and oranges extend into the east. I did not try to make it look like the sun was setting in front of us. I guess my change in the color of the sea did not work well enough.

I am glad my channel blending worked well especially on the stage.

James Gray


On Thu, Feb 18, 2021 at 12:51 PM Dan Margulis via groups.io <dmargulis=aol.com@groups.io> wrote:

308 This is the auxiliary par version, a straight average of the five indicated here. We gave it a full discussion in a separate thread.

316 Chosen for the auxiliary par. With the emphasis on channel blending in the Hotel Lobby and Mantillas exercises it is somewhat surprising that more people didn’t try blending the red into the blue. This guy did, which accounts for the better definition on stage, the lighter fleshtones in the foreground, and the lighter banner. The sharpening is acceptable to me but many would find it heavy-handed.  This may have the best stage of anyone and the foreground is also excellent. I downgraded it because there’s something weird-looking about the banner, and because the colorizing of the water seems too obvious. But the strengths of this version make it a clear choice for the auxiliary par.



Harvey Nagai
 

"We have met the enemy and he is me" (sic.)

I carped a bit about the par the other day, saying attention is drawn neither up to the stage 
nor down to the audience.  I don't think I need a mirror to find the culprit: 314 (mine) is
darker in the stage than the others, also lighter in the audience, so it flattened out the
difference.  A par of only the other four emphasizes the stage as their authors wished.

Dan said "I think I was too easily impressed by the stage"

The explanation for my choice was that the stage is visibly and measurably darker in the
original image, and I (over-)reasoned that the stage would have been designed with a darker
backdrop to contrast with the performers illuminated by the stage lights.  Faux-clever,
definitely wrong: googled images of the old Seashell Stage show the backdrop to be light,
as in the other par entries.

But it seems that the idea behind the contrast was effective in drawing initial attention
to the stage, which was my intention.

In retrospect, it's probably one of the reasons the rest of the image ended up so light:
the more it was lightened, the more the stage stood out.


Christophe Potworowski
 

Thanks for the comments.

My image was 301. I was afraid of overdoing the warmth to the point of leaving it on the cool side, forgetting in the process that the client is always right, and he wanted a sunny day.

Christophe

On Feb 18, 2021, at 3:07 PM, Kenneth Harris <reg@ikonyc.com> wrote:

I'd say working time. It's an open book test. I usually come up with my goal and my plan on any project within ten minutes, jotting down my notes which techniques I plan to use. Key for me is to verbalize what's good and bad about the picture or group. Your business likely allows for more rumination.

Judging by the time between entry closing and the posting of all the images on Mondays, seems like Dan is pretty fast at this kind of thing.

Ken Harris





John Furnes
 

Thanks for the advice. I just tried it – Lab and then heavy a and b curves on a new layer. Black mask and painted white at 3% flow.

It got way better. Nice ‘trick’

 

 

John Furnes

 

 


David Remington
 
Edited

Dan,
 
Thanks for the thorough and though provoking comments.
 
My version is 321. I still like it as "realism" is my style but in context I see shortcomings. I don't think we would often completely agree on photographic aesthetic but reading your comments I still find myself thinking "you know he's right", or seeing things with a fresh perspective. That is what this is all about for me.
 
I like version 314 which is not too dissimilar to mine. Lighter and cleaner with better concrete, and water. I made some blends. Even 10% of it mixed into mine helps. I also tried Match Color with some variations and got improved results. These also produce a neutral or near to it white on the signs. It's on the edge of too warm for me but good to think about.
 
I like seeing the variety that this group produces. There is an old saying that you never know if you have gone far enough until it is clear you went too far. That applies nicely to the method of producing different versions comparing and blending when it helps.
 
David


Ronny Light
 

Do you mean 321?

 

 

 

Ronny

www.RonnyLightPhoto.com

5010 B Wilkerson Dr., Nashville, TN 37211

 

 

 

From: colortheory@groups.io <colortheory@groups.io> On Behalf Of David Remington
Sent: Friday, 19 February, 2021 7:56 AM
To: colortheory@groups.io
Subject: Re: [colortheory] Concert on beach: Dan's comments

 

 

My version is 221

 

 

David

 


David Remington
 
Edited

Yes, and 314. Thanks!