Case Studies in Retrospect, 2021


Dan Margulis
 

Now that the ten case studies have drawn to a close, I’d like to thank everyone who wrote me privately or posted positive comments to the group about the experience. And I suppose we also must acknowledge, but without thanks, the role of the pandemic in making such an exercise possible in the first place, let alone doing it two years in a row.

We still have one weird exercise to go but it has taken longer than planned to put it together; probably I’ll post it this weekend and give two weeks rather than one to get it back. Meanwhile, I think we should have a thread summarizing the lessons of the ten as a whole. We had an excellent one of these last year:

I started it off with:

But, speaking of review, it would nice to hear from the group about what was accomplished, what worked, and what didn’t. I ask this because my experience in reading evaluations over the 25 years that I taught Applied Color Theory is that students and instructors often don’t see eye to eye with the instructors (or even themselves) about which images were the most useful. My suspicion is that people think the most useful image is the one they screwed up, rather than the one that made the best points.

Alas, we have to doubt that the ACT course can ever be held again. Still, by force of habit, I’d repeat one of the questions on the evaluation form: name three or more images that you found particularly useful. Those that, over time, don’t get their share of votes, get thrown out and replaced by others for future classes. And those that get lots of votes get kept even if I think there are better options.

A reminder that last year’s case studies are still posted. Those who didn’t participate then can go through the exercises and then see the same kind of discussion we’ve had recently. At the bottom of this message are lists of both the 2020 and 2021 case studies.

In both years, I selected a series of ten to be worked in a specific order, as certain one built on lessons from the previous ones. I’ll start the conversation off by saying I think the order this time was good. The themes in this series were channel blending, which was critical in at least four; dealing with unusual lighting conditions (five, and six if you count the smoke around Shasta as a lighting condition); and careful planning where it wasn’t obvious what the objective was.

I had seen four of the ten as part of the MIT series. My only surprises there were Bellagio at Night and Concert on the Beach, which both were deeper than I thought. I said at the outset that I thought Hotel Lobby and King of Beasts were the easiest of the ten, and still think that.

Mantillas and Choir came from my advanced ACT classes so I knew how hard they are.

Of the four I’d never seen before my only surprise was Land of Pagodas, which I had thought was little more than a counterweight for Beach at Sunset and, to a lesser extent, Choir. But it wound up being more instructive than I had expected.

Every image had at least a few good entrants, but speaking of the group as a whole, I’d say we did quite poorly on Hotel Lobby and Beach at Sunset and not very well on Bellagio. I’d say we did better than expected on Mantillas, Choir, and High Andes. 

An interesting change from last year: in 2020 the par version was almost always either the best or close to it. This year it wasn’t true, for a variety of reasons.

I look forward to additional comments from the group.

Dan

The 11 case studies from 2021, in order:
Hotel Lobby
Mantillas
Concert on the Beach
Beach at Sunset
Bellagio at Night
King of Beasts
Choir
Land of Pagodas
Shasta
High Andes


The 11 case studies from 2020, in order:
Carnival
Veiled Bride
Niagara Spray
Cinque Terre
Colosseum
Panama 1978
Monument Valley
Toast to Greece
Seated in Grass
Adirondacks
Red Rose


Doug Schafer
 

I enjoyed both 2020 and 2021 exercises. 2021 was another great learning set of images and entrants/entries and feed-backs. And special thanks to Dan for hosting, analyzing, providing reviews, and offering new/many methods as to how to approach several methods of blending images....as well as what is correct, or wrong, or left to the judgement of author or viewer.
I learned (or re-learned) these: (and re-verified as included in my workflow process checklist)
- Know what the specific "correctness" of the image should be (know the subject matter; don't guess)
- Know desired end results (client)(set specific goals)
- Appreciate there is a large range of "artistic license"; use it wisely
- Remember channel blending possibilities
- Do final image checks (example: auto-tone)
- Wait a day and review image with fresh mind and eyes before considering it is "done"
- Blending multiple results could yield something better
- . . . .  and more!

Doug Schafer


Hector Davila
 

I see I need to learn 'color correction by the numbers'.
So, I will try and gather all info I can find and maybe..
make my own 'color correction by the numbers for Dummies'.


https://documents.sessions.edu/eforms/courseware/coursedocuments/color_correction/lesson4.html
https://www.summitprintingpro.com/graphic-design/tutorials/skin-tone-correction.html
https://vimeo.com/53177325
https://www.youtube.com/watch?v=2PweW75-K84
https://geraldbakker.nl/psnumbers/color-correction-pt1.html

Hopefully I should be able to come up with a simple step by step process
that shouldn't take up more than a paragraph.

Hector Davila


Kent Sutorius
 

I am thankful that I could go through this experience. As a nonuser of Photoshop and PPW, I found all of the exercises difficult. But I was able to learn and practice on working in LAB, Channel Blending, color by numbers, blending other samples (versions) and questions to ask myself when editing an image. I also learned to walk away and come back at a later time to work on a project.

A very lasting and memorable experience for me was the generosity of Dan's time to teach and communicate his wisdom, knowledge and experience to all of us. Also to correct us but correct us with opportunities to learn and steps to make our images better. The group as a whole was very positive and generous in their insights.

In a world that is so divisive and coarse and unprofitable with its communication, colortheory was something I looked forward to everyday.

Kent Sutorius




On 4/27/2021 4:26 PM, Dan Margulis via groups.io wrote:
Now that the ten case studies have drawn to a close, I’d like to thank everyone who wrote me privately or posted positive comments to the group about the experience. And I suppose we also must acknowledge, but without thanks, the role of the pandemic in making such an exercise possible in the first place, let alone doing it two years in a row.

We still have one weird exercise to go but it has taken longer than planned to put it together; probably I’ll post it this weekend and give two weeks rather than one to get it back. Meanwhile, I think we should have a thread summarizing the lessons of the ten as a whole. We had an excellent one of these last year:

I started it off with:

But, speaking of review, it would nice to hear from the group about what was accomplished, what worked, and what didn’t. I ask this because my experience in reading evaluations over the 25 years that I taught Applied Color Theory is that students and instructors often don’t see eye to eye with the instructors (or even themselves) about which images were the most useful. My suspicion is that people think the most useful image is the one they screwed up, rather than the one that made the best points.

Alas, we have to doubt that the ACT course can ever be held again. Still, by force of habit, I’d repeat one of the questions on the evaluation form: name three or more images that you found particularly useful. Those that, over time, don’t get their share of votes, get thrown out and replaced by others for future classes. And those that get lots of votes get kept even if I think there are better options.

A reminder that last year’s case studies are still posted. Those who didn’t participate then can go through the exercises and then see the same kind of discussion we’ve had recently. At the bottom of this message are lists of both the 2020 and 2021 case studies.

In both years, I selected a series of ten to be worked in a specific order, as certain one built on lessons from the previous ones. I’ll start the conversation off by saying I think the order this time was good. The themes in this series were channel blending, which was critical in at least four; dealing with unusual lighting conditions (five, and six if you count the smoke around Shasta as a lighting condition); and careful planning where it wasn’t obvious what the objective was.

I had seen four of the ten as part of the MIT series. My only surprises there were Bellagio at Night and Concert on the Beach, which both were deeper than I thought. I said at the outset that I thought Hotel Lobby and King of Beasts were the easiest of the ten, and still think that.

Mantillas and Choir came from my advanced ACT classes so I knew how hard they are.

Of the four I’d never seen before my only surprise was Land of Pagodas, which I had thought was little more than a counterweight for Beach at Sunset and, to a lesser extent, Choir. But it wound up being more instructive than I had expected.

Every image had at least a few good entrants, but speaking of the group as a whole, I’d say we did quite poorly on Hotel Lobby and Beach at Sunset and not very well on Bellagio. I’d say we did better than expected on Mantillas, Choir, and High Andes. 

An interesting change from last year: in 2020 the par version was almost always either the best or close to it. This year it wasn’t true, for a variety of reasons.

I look forward to additional comments from the group.

Dan

The 11 case studies from 2021, in order:
Hotel Lobby
Mantillas
Concert on the Beach
Beach at Sunset
Bellagio at Night
King of Beasts
Choir
Land of Pagodas
Shasta
High Andes


The 11 case studies from 2020, in order:
Carnival
Veiled Bride
Niagara Spray
Cinque Terre
Colosseum
Panama 1978
Monument Valley
Toast to Greece
Seated in Grass
Adirondacks
Red Rose



Gerald Bakker
 

On Tue, Apr 27, 2021 at 10:26 PM, Dan Margulis wrote:
In both years, I selected a series of ten to be worked in a specific order, as certain one built on lessons from the previous ones. I’ll start the conversation off by saying I think the order this time was good. The themes in this series were channel blending, which was critical in at least four; dealing with unusual lighting conditions (five, and six if you count the smoke around Shasta as a lighting condition); and careful planning where it wasn’t obvious what the objective was.
The "careful planning" is what failed me in many of the exercises, and it caused more misses than I'd wished. So this is my most important lesson: to first analyze the image and determine the best strategy, and not start processing right away. Consider image content and whatever artistic aspects, and then move ahead.

But there are of course more things: 
- It pays to do local enhancements. The PPW, designed to be a quick workflow, doesn't support this, but in many cases I got considerable improvements by picking a brush and highlighting or de-emphasizing certain image areas, or even fine-tuning color. Of course the full week that we had available per image helped in this respect.
- When making multiple versions, do not just take different paths, but more importantly, aim for completely different objectives. If the first version is pretty light, even if you like it, make another that is deliberately dark, etc. This is a lesson that I took from Dan's extensive comments, where different versions were cleverly merged to something even better. Normally, one doesn't have all these (very different) versions available, so why not create a few extremes. 
- I did a refresher on LAB color blending, inspired by Dan's comments on my King of Beasts submission.
- I got reminded that detail enhancement and sharpening is not always a good thing (especially in the Bellagio and Andes images).

In general, I think the second series was harder than the first, although my memory may deceive me (and I didn't finish the first series). But most of all, it was a highly enjoyable experience!
--
Gerald Bakker
https://geraldbakker.nl


jorgeparraphotography
 

On Tue, Apr 27, 2021 at 04:26 PM, Dan Margulis wrote:
But, speaking of review, it would nice to hear from the group about what was accomplished, what worked, and what didn’t. I ask this because my experience in reading evaluations over the 25 years that I taught Applied Color Theory is that students and instructors often don’t see eye to eye with the instructors (or even themselves) about which images were the most useful. My suspicion is that people think the most useful image is the one they screwed up, rather than the one that made the best points.
 
Alas, we have to doubt that the ACT course can ever be held again. Still, by force of habit, I’d repeat one of the questions on the evaluation form: name three or more images that you found particularly useful. Those that, over time, don’t get their share of votes, get thrown out and replaced by others for future classes. And those that get lots of votes get kept even if I think there are better options.

Dan, first off, of course, many many thanks for being so generous with your time and knowledge and this has been invaluable, even for a rookie like myself!

Even though I may not have a deep understanding of the "missing tasks in my own Photoshop menu, like channel blendings and working on LAB ( still behind in those areas, luckily I have not been as freely available as I thought I was gonna be during the lockdown and successive months to date). I got both the Photoshop and the Chevreul books and both are still to be finished, but it is something that has become like a natural  "to do", a necessary expansion of the current "visual database" in my brain, so it will happen and hopefully, I may perform better on a future challenge.

Now, putting this great knowledge hole aside, I still learned so many things and even did relatively fine in one of the challenges ( the concert at the beach), while I messed up big time in others, but what I keep thinking here is that everyone who entered their images for review was completely and fully convinced that his/her approach was the right one for the image at hand, as it was properly expressed in the emails BEFORE your evaluation hit in. This is what created soo many different variations of the same subject.  For you, it may be a simple reflection of how little we know about color techniques, but To me, this is still a clear view of how subjective values, cultural biases, likes and dislikes, will always play a role in the decision-making process when evaluating a file and making choices on how to do about it in term of workflow and subtle adjustments.

In my personal and humble opinions, many images from each challenge would be "acceptable" to a client, in spite of having obvious technical problems which could be improved.  Very interestingly though, most of the images would not belong to a commercial project, the idea of adjusting amateur images brings an additional layer of complexity, starting with errors in the capture, a defective color rendering by consumer cameras, poor optics, etc. In the world of clients and commercial work, many of the defects of the original images would just not be there, so getting a good final file takes less work and the use of not so many techniques to achieve it. The challenges NEED to be done on mostly amateur images so we can get deep into adjustments.

Another outstanding situation was to find out how the latest images offered the opportunity of creating several PAR images, which adds even more gas to the fire: If there is not one but several reference files, which one is the most adequate one to evaluate all contestants? We fall again into the issue of subjectivity, even when counting with the best files and the best possible mix(es) to build a PAR or reference file.

On the idea of the course not being able to be run again in person, and as an educator myself, I wonder: what would be the impediment to build it online?

How difficult it would be to create a paid site, expanding on what the book explains, include some basic, mostly unmonitored challenges per each section or set of processes, and when participants get approved, you then grant them private ( and paid) access to one big challenge like the ones we have done this and last year? This online training would inevitably raise the level of the participants BEFORE entering any challenge, giving you ample chance to enjoy good quality images sent for each project and making a REAL mark in the impact of the participant's knowledge on the matter of Color.

I know this is a long-term project, as you need coders for such a website, but once it is done, you would be having a system that pays its own costs, and gives you an additional income, which is never unwelcome!

Basically, This training deserves a permanent place on the web!.

Well, at the very least, that is how I see it, and hope you take it as positive feedback since you asked for it. 

Jorge Parra


 

 
--
Jorge Parra 
www.jJorgeParraPhotography.com
Miami


John Furnes
 

I think these series of 2020 and 2021 have been very valuable to me. I think of the opportunity to get criticized and evaluated by Dan and the Group as winning the Lottery. My starting point is not photography as a profession or even extreme interest, but at some point I had the need to get the ‘right’ colours. Luckily, I came upon Dan Margulis. And was immediately ‘converted’. Since, my interest for colours has become more than just ‘getting it right’.

 

In these courses, I have not regarded myself as in competition with Group members – it would be something like  competing at the Olympics without knowing how to run, but I have found it most instructive and rewarding.

So what have I learned?

Analysing the image - Finding the most interesting part in the picture – the real subject or focus.

Find out what can be done to improve this or emphasize it.

Check numbers in L*a*b for obvious flaws, and find colourcast / and what can be done to adjust these.

Skin colour has been one of my obvious problems. Or reds, to be more to the point.

How much colour saturation can a certain amount of light allow for.

And the techniques – BlendIf in L*a*b; Adjustment curves on channels; selective sharpening; Apply image and many more.

 

Reading the comments from group members has been instructive, and following the discussions. Of course Dan’s explanations and in-depth comments on various problems have given much ‘flesh on the bones’.

 

The most rewarding exercises have been

  • Pagodas - with the issues of lighting and saturation, and what would be a natural sky
  • Andes – with all the red colours, skin and ‘the real focus’
  • Choir – also with reds, light, saturation and general presentation (focus)

 

The last one the Courthouse Wash – is the most challenging of all.

 

However, I find all exercises challenging and instructive, and thanks to Dan for helping us with this and letting us learn from his immense knowledge. (To many of us, analysing a picture takes a lot of work, not to mention how to resolve issues concerning what we find; to Dan it is second nature)

 

 

John Furnes

Denmark

www.johnfurnes.com

 

 

 

 


Paco
 

Dan,

I only participated up to King of Beasts so I'll limit this comment to just those. I found them all to be helpful for me in one way or another and I learned from having tried my hand at all. The lobby shot, although called the easiest one, is an example of how a mask made from one of the channels can make retouching work easier so I would encourage you to keep it even if just for a quick demonstration of that technique.

Thanks for your help Dan!

Paco


Dan Margulis
 



On Apr 28, 2021, at 2:59 PM, Gerald Bakker <gc.bakker@...> wrote:

- When making multiple versions, do not just take different paths, but more importantly, aim for completely different objectives. If the first version is pretty light, even if you like it, make another that is deliberately dark, etc. This is a lesson that I took from Dan's extensive comments, where different versions were cleverly merged to something even better. Normally, one doesn't have all these (very different) versions available, so why not create a few extremes. 

Why not, indeed. This series has certainly demonstrated the utility of extreme auxiliary versions. Yet people are afraid to construct them. I used a few of them, but neglected to do so in the Shasta exercise. I decided that as long as I created a strong mountain all would be well, and went too far, getting a version that wasn’t to my liking when compared to the others. I should have made a more conservative version, plus an extreme version with more contrast in the mountain than anyone would possibly want. That would have let me make a more intelligent decision about how to blend them.

I suspect that we all have an instinctive fear of the ugly, even when the ugliness is created in the name of beauty. That is why, IMHO, some are afraid of acquiring a very flat version in the raw module, or of making a grossly too colorful version with Color Boost—or making a ridiculously contrasty auxiliary version. 

Dan


James Gray
 

I was going to say I disagree with this.  However, I really am skeptical that it will work very well for many of us.  So to be specific I do not have reservations about creating auxiliary versions and have no real fear of constructing them.  As I have tried to study the comments from Dan and the other participants, I have often found myself saying to myself, "I never thought of that."  To be clear, I work on my photographs and almost never have access to versions made by someone else.  I could create ugly versions in the interest of beauty.  Often I blend versions together.  You mention the Shasta exercise.   I blended 4 different versions together to create my entry to that exercise.  It still left a lot to be desired because I did think of the additional version with a better sky.  I do think my mountain was better than most.  Please understand I am not really trying to discuss again the Shasta exercise.  I am really interested in the issue of blending.  I think too many of the excellent par versions and some of the secondary par versions were good because they were created by different persons, not simply because auxiliary versions were created.  At least for me, the emphasis on blending was not entirely what I needed to improve my work.  I did appreciate all of the other details about what was right and wrong about individual versions and details about how to achieve certain goals.  I have thought about going through all of the specific steps for improving versions and compiling them into one combined document.

I will just say a few words about which exercises I preferred.  I liked the beach concert, the little girl in Peru, and the stupas (aka pagodas).  Less useful were the Niagra Falls and Beach at sunset.

Jim Gray


On Sat, May 8, 2021 at 2:26 PM Dan Margulis via groups.io <dmargulis=aol.com@groups.io> wrote:


On Apr 28, 2021, at 2:59 PM, Gerald Bakker <gc.bakker@...> wrote:

- When making multiple versions, ... so why not create a few extremes. 

Why not, indeed. This series has certainly demonstrated the utility of extreme auxiliary versions. Yet people are afraid to construct them. I used a few of them, but neglected to do so in the Shasta exercise. I decided that as long as I created a strong mountain all would be well, and went too far, getting a version that wasn’t to my liking when compared to the others. I should have made a more conservative version, plus an extreme version with more contrast in the mountain than anyone would possibly want. That would have let me make a more intelligent decision about how to blend them.

I suspect that we all have an instinctive fear of the ugly, even when the ugliness is created in the name of beauty. That is why, IMHO, some are afraid of acquiring a very flat version in the raw module, or of making a grossly too colorful version with Color Boost—or making a ridiculously contrasty auxiliary version. 

Dan


KENT SOUTHERS
 

I too found the issue of blending to be a technique that comes with a mixed bag of pro / con issues that come along for the ride.  But, about half-way through the exercises ... it donned on me that (photoshop at least) blending modes afford a means for selectively using portions of an alternate image.  Basically, a mathematical mask, rather than an opaque one.  So, through the use of blending modes and alternate versions (pushing the envelope at times), coupled with opacity adjustments etc. we can harness some of what we want, and not what we don't.

Ironically, I tend to push certain things, then when I'm "finished" with an image, I'll place the original back on top, and use a blending mode / opacity adjustment to pull it back if I think I overdid something.  Basically, a reversal application of what the exercises were promoting as a cross-check / adjustment technique - tool.  Not a tell all, do all ... but, another tool for the toolbox.  And, like all tools, no single tool works every time, for everything, for everyone. 

How well we learn to use (any) the tool, will determine how frequently we use it ... or, is that vice versa. 😉
That, and if our other tools are working well for us, then we may have choices for which tools we prefer.


Kent Southers, CMRP
southers3@...



From: colortheory@groups.io <colortheory@groups.io> on behalf of James Gray <James@...>
Sent: Saturday, May 8, 2021 8:07 PM
To: colortheory@groups.io <colortheory@groups.io>
Subject: Re: [colortheory] Case Studies in Retrospect, 2021
 
I was going to say I disagree with this.  However, I really am skeptical that it will work very well for many of us.  So to be specific I do not have reservations about creating auxiliary versions and have no real fear of constructing them.  As I have tried to study the comments from Dan and the other participants, I have often found myself saying to myself, "I never thought of that."  To be clear, I work on my photographs and almost never have access to versions made by someone else.  I could create ugly versions in the interest of beauty.  Often I blend versions together.  You mention the Shasta exercise.   I blended 4 different versions together to create my entry to that exercise.  It still left a lot to be desired because I did think of the additional version with a better sky.  I do think my mountain was better than most.  Please understand I am not really trying to discuss again the Shasta exercise.  I am really interested in the issue of blending.  I think too many of the excellent par versions and some of the secondary par versions were good because they were created by different persons, not simply because auxiliary versions were created.  At least for me, the emphasis on blending was not entirely what I needed to improve my work.  I did appreciate all of the other details about what was right and wrong about individual versions and details about how to achieve certain goals.  I have thought about going through all of the specific steps for improving versions and compiling them into one combined document.

I will just say a few words about which exercises I preferred.  I liked the beach concert, the little girl in Peru, and the stupas (aka pagodas).  Less useful were the Niagra Falls and Beach at sunset.

Jim Gray

On Sat, May 8, 2021 at 2:26 PM Dan Margulis via groups.io <dmargulis=aol.com@groups.io> wrote:


On Apr 28, 2021, at 2:59 PM, Gerald Bakker <gc.bakker@...> wrote:

- When making multiple versions, ... so why not create a few extremes. 

Why not, indeed. This series has certainly demonstrated the utility of extreme auxiliary versions. Yet people are afraid to construct them. I used a few of them, but neglected to do so in the Shasta exercise. I decided that as long as I created a strong mountain all would be well, and went too far, getting a version that wasn’t to my liking when compared to the others. I should have made a more conservative version, plus an extreme version with more contrast in the mountain than anyone would possibly want. That would have let me make a more intelligent decision about how to blend them.

I suspect that we all have an instinctive fear of the ugly, even when the ugliness is created in the name of beauty. That is why, IMHO, some are afraid of acquiring a very flat version in the raw module, or of making a grossly too colorful version with Color Boost—or making a ridiculously contrasty auxiliary version. 

Dan


Dan Margulis
 



On May 3, 2021, at 1:17 PM, jorgeparraphotography <jorgeparraphotography@...> wrote:

Dan, first off, of course, many many thanks for being so generous with your time and knowledge and this has been invaluable, even for a rookie like myself!

I am happy to read this and similar comments and hope that the two series lead to a lasting improvement in people’s work. If so, it will be one of the few positive things to come out of this pandemic, without which there wouldn’t have been enough available time to do it in such depth.

Now, putting this great knowledge hole aside, I still learned so many things and even did relatively fine in one of the challenges ( the concert at the beach), while I messed up big time in others,

Anybody who participates in several of these exercises will do better on some than on others. That much is so obvious that we can overlook an important ramification.

Participants in this group range from beginner to expert but are all interested in the subject and all reasonably competent with the tools at hand. Anyone matching that description will sometimes produce a version highly competitive with the par and sometimes not. It’s those “other” times that define the skill level. An expert’s poorest entry may be obviously worse than par but it’s unlikely to be below average for the group.

When I was hiring retouchers, therefore, I wasn’t looking at their best work so much as at their worst. Clients aren’t good at distinguishing brilliant work from merely professional-level, but they are good at detecting garbage. So, I’d much prefer to hire a person who produces consistently good results rather than someone whose work varies from awesome to unacceptable.

My suggestion to participants, therefore, is to accept their own good results as natural and expected, and then to look closely at the bad ones.

but what I keep thinking here is that everyone who entered their images for review was completely and fully convinced that his/her approach was the right one for the image at hand, as it was properly expressed in the emails BEFORE your evaluation hit in.

I may not be reading this correctly but if it is saying that people needed to hear from me before concluding that their work was bad, I disagree. Plenty of people expressed disappointment in their own version as soon as the results were posted; this is why I deliberately stayed out of each thread for the first two days.

The times when the opposite occurred were when the individuals weren’t particularly happy with the par, which blinded them to certain deficiencies in their own versions. Then, yes, my comments may have changed minds because I was trying to show ways of improving their version to create an alternate par.

This is what created soo many different variations of the same subject.  For you, it may be a simple reflection of how little we know about color techniques, but To me, this is still a clear view of how subjective values, cultural biases, likes and dislikes, will always play a role in the decision-making process when evaluating a file and making choices on how to do about it in term of workflow and subtle adjustments.

That’s true but it can be overstated. Gerald and Ken often chimed in with their own choices for best, before knowing what my own were. Typically what would happen, if there were, say, 30 overall entrants, the three of us would recognize that one or two were head and shoulders above the others and that 22 were noncompetitive. This would leave five or six candidates to fill out the rest of our lists and then our personal biases would come into play.

The crucial thing seems to me to explain how the 22 noncompetitive versions got that way. Sometimes this was because of an obviously poor artistic choice, such as coolness where warmth is needed. Sometimes a crucial step was omitted, such as the channel blending needed in Hotel Lobby or in Pagodas. More frequently, though, it was a technical error. In a number of these evaluations I posted comparative LAB numbers, suggesting that certain ones were acceptable and others were wrong—not matters of personal taste at all.

In my personal and humble opinions, many images from each challenge would be "acceptable" to a client, in spite of having obvious technical problems which could be improved.  Very interestingly though, most of the images would not belong to a commercial project, the idea of adjusting amateur images brings an additional layer of complexity, starting with errors in the capture, a defective color rendering by consumer cameras, poor optics, etc. In the world of clients and commercial work, many of the defects of the original images would just not be there, so getting a good final file takes less work and the use of not so many techniques to achieve it. The challenges NEED to be done on mostly amateur images so we can get deep into adjustments.

All this is true, but again it brings up the need for more care, because without anything to compare it to, the client may accept something that isn’t very good. As an example, in all nine of the MIT studies in this round and in 2020, I posted the version that I did in 2017. Some of these still seem good, or at least good given the time constraints, but back then I had only the MIT retouchers to compare to, which made it easy to overestimate the quality of my own work. When I started a 2021 version of Bellagio I quickly realized that my 2017 version was poor. And I never did understand what was wrong with my 2017 (and 2021) Concert on the Beach until I saw the group’s versions.

Another outstanding situation was to find out how the latest images offered the opportunity of creating several PAR images, which adds even more gas to the fire: If there is not one but several reference files, which one is the most adequate one to evaluate all contestants? We fall again into the issue of subjectivity, even when counting with the best files and the best possible mix(es) to build a PAR or reference file.

Looking at a bunch of alternate pars indeed will provoke differences of opinion for the same reasons discussed earlier. When Gerald and Ken and I disagreed, we had each already eliminated a whole bunch of inferior versions and agreed on a couple of remarkably good ones. This left only a few choices and often they were questions of the lesser of the evils. But par versions don’t have serious defects, so lots of individual preferences manifest themselves.

On the idea of the course not being able to be run again in person, and as an educator myself, I wonder: what would be the impediment to build it online? 

My age.

How difficult it would be to create a paid site, expanding on what the book explains, include some basic, mostly unmonitored challenges per each section or set of processes, and when participants get approved, you then grant them private ( and paid) access to one big challenge like the ones we have done this and last year? This online training would inevitably raise the level of the participants BEFORE entering any challenge, giving you ample chance to enjoy good quality images sent for each project and making a REAL mark in the impact of the participant's knowledge on the matter of Color.

It’s do-able technically but would be a real step downward from a true classroom experience, where everything happens in real time. In an ACT class, instead of doing 10 images over 10 weeks, we did 26 over three days, and the participants become a family who support each other when discouragement sets in.


Well, at the very least, that is how I see it, and hope you take it as positive feedback since you asked for it. 

I do, thanks.

Dan


Dan Margulis
 



On May 8, 2021, at 10:34 PM, KENT SOUTHERS <southers3@...> wrote:

Ironically, I tend to push certain things, then when I'm "finished" with an image, I'll place the original back on top, and use a blending mode / opacity adjustment to pull it back if I think I overdid something. 

There’s nothing ironic about it. Granted how the human visual system works, it’s the logical last step in PPW—a step that’s conspicuously missing in traditional workflows.

Consider the problems of Person A, who works traditionally, and Person B, who uses PPW.

Person A adds color to the original until satisfied. Person B goes overboard with the color and then backs off until satisfied.

The last thing Person A sees before making his final decision is the relatively dull original. The last thing Person B sees is the wildly oversaturated Color Boost version. These comparisons greatly influence their final choices. Person A is apt to approve something too bland, because it’s exciting next to the original. Person B will likely approve something that’s too loud, because next to the Color Boost version, it isn’t.

This is why Person B *should* now go back to the original, to see whether the “corrected” version should be toned down further. Person A has no such step available and is likely to live in ignorance of the dullness of his version.

Dan


jorgeparraphotography
 

On Sun, May 9, 2021 at 10:43 AM, Dan Margulis wrote:
On the idea of the course not being able to be run again in person, and as an educator myself, I wonder: what would be the impediment to build it online? 
My age.

How difficult it would be to create a paid site, expanding on what the book explains, include some basic, mostly unmonitored challenges per each section or set of processes, and when participants get approved, you then grant them private ( and paid) access to one big challenge like the ones we have done this and last year? This online training would inevitably raise the level of the participants BEFORE entering any challenge, giving you ample chance to enjoy good quality images sent for each project and making a REAL mark in the impact of the participant's knowledge on the matter of Color.
It’s do-able technically but would be a real step downward from a true classroom experience, where everything happens in real time. In an ACT class, instead of doing 10 images over 10 weeks, we did 26 over three days, and the participants become a family who support each other when discouragement sets in.
Dan I will add my final cent in this discussion to say one more thing: back when you ran a test for donations (and it worked ok) you were still wondering what to do with that money other than paying the server's fees, etc. I have absolutely no doubt that most of us here would be willing to contribute to cover the expense of building such a website on behalf of both the participants and of your legacy!!

It is clear that the IRL ( In real life) meetings have provided an overabundance of examples in a few days ( something that can also be overwhelming BTW).  I still feel (apologies if I am going too far), that building something that will set the basis and secure the most basic principles of your system ( work on color first, then work on contrast, etc) will definitely train an entirely new generation of retouchers, graphic designers and photographers with a more systematic approach to their work, and that can only be positive!!. 

The pandemics at the least here in the USA seems to be receding and maybe some workshops might be re-established but I have no hesitation to say that a website where you can subscribe, pay for your monthly access, develop the basic skills based on the books, and eventually participate in some online challenges would bring to life ( and keep alive) your knowledge and history as an expert in the field, showcasing a totally different workflow and approach for color processing. And of course, if the conditions are met for new in-person workshops, well, why not??

As you already checked, there is limited and incomplete literature and videos on these subjects, so, even in terms of marketing, this is an empty niche that can be filled, and you have the know-how to make it happen.

Best
 
--
Jorge Parra 
www.jJorgeParraPhotography.com
Miami


Gerald Bakker
 

On Sun, May 9, 2021 at 03:07 AM, James Gray wrote:
I was going to say I disagree with this.  However, I really am skeptical that it will work very well for many of us.  So to be specific I do not have reservations about creating auxiliary versions and have no real fear of constructing them.  As I have tried to study the comments from Dan and the other participants, I have often found myself saying to myself, "I never thought of that."  To be clear, I work on my photographs and almost never have access to versions made by someone else.  I could create ugly versions in the interest of beauty.
I too expect that in many cases constructing and blending extreme versions won't work. Also I agree with Dan when he says that many have a fear of the ugly. This is not a technique that I will use very often. But we've seen it being effective in multiple exercises, so I picked that up as a lesson. Something to add to the toolbox.
--
Gerald Bakker
https://geraldbakker.nl