Dan Margulis Applied Color Theory
Trends in 2008 ACT Classes
Trends in 2008 ACT Classes
Posted by: Dan Margulis
Sun Nov 16, 2008 7:51 pm (PST)
I'm back at home briefly prior to the third and last
advanced course of the year. The results of the first two have been
interesting; when the third is completed I'll discuss what the three
classes suggested about current events and color. Meanwhile, the final
basic ACT of the year was last week. For the last two years, I've polled
every class at the outset on various topics and kept records of the
results.
A couple of items had nearly identical results. In
both 2007 and 2008, the ratio of persons claiming to work at least four
hours in Photoshop on a typical workday versus those who do not was just
about 50-50. Also, the choice of customary RGB workspace was nearly the
same both years: about 65 percent Adobe RGB, 20 percent sRGB, 10 percent
don't know, the remaining five percent or so being ColorMatch RGB and
various ultra-wide alternatives. These results track closely my similar
polls at Photoshop World. Five years ago, Adobe RGB and sRGB had nearly
equal shares.
Over the last few years the biggest growth in
attendance has come from the "super-amateurs" or whatever you
like to call them. These are people, typically successful professionals in
other fields, who are extremely serious about their photography/fine art
printing. Most have made some money from their hobby, but they do not
depend on it for a living. The three classes I've just returned from, for
example, included a dentist, a cardiac surgeon, a lawyer (who was
color-blind, yet) and an IT professional. More and more, it seems that
people are in the field for love, not money.
There was a disturbing trend last year that, to my
surprise, did not continue in 2008. I am not so rude as to ask the question
outright, but I do estimate whether each student is paying out of his own
pocket or whether a company is paying. In 2007, for the first time, my
figures said that more than half paid themselves. I was particularly
disgusted to see a big disparity between the U.S. and Canada: it seemed
that Canadian companies are more disposed to spend on training their staff
than U.S. firms are.
I expected this to continue, in light of the economy.
I thought that with times being hard, fewer companies would spend on
training, and more individuals would see the need to protect themselves by
improving their skills, at their own expense if need be. Plus, the
statistics would be further inflated by the super-amateurs, who pay for
themselves.
It was a pleasant surprise to find out otherwise. My
estimate last year was that 53 percent of attendees paid their own way;
this year it was a significantly lower 43 percent. Good for the companies
who realize the best investment they can make is in their people!
A couple of other interesting changes: I ask, who is
the final judge of whether your color work is successful? Is it you, or is
it somebody else, such as a boss or client? Historically, class attendees,
being heavily involved in marketing their work to the public, would
overwhelmingly have answered "somebody else". Last year, it was
down to 79 percent saying the final vote belongs to somebody else. This
year, a big drop to 64 percent. Partly that's due to the super-amateurs,
but maybe it also indicates a trend in favor of more artistic uses for
color correction, and not so much commercial activity.
Another question: we all have to work on substandard
originals from time to time: photos shot by amateurs, or by professionals
under adverse conditions, or just poor shots that nevertheless have to be
used. Some of us, however, work on such images much more than others.
Advertising retouchers rarely see them; those preparing newsletters for
corporate clients see them all the time. Would you say that you have to
color-correct such images often, or seldom?
Ten years ago the "seldom" vote would have
been close to unanimous. Back when color correction was expensive and
digital cameras poor if available at all, few would invest in having a
professional attempt to correct garbage. Now, with everyone his own
photographer, it's a different story. Last year, for the first time,
"often" won, with a bigger margin than Barack Obama: 55-45. This
year, an interesting, sharp move in the opposite direction: 59-41 in favor
of "seldom." All the factors discussed above contribute to why
such a change should have occurred, but the move is too big to be explained
away without more. I am guessing that as digital photography has become
mature and quality cameras cheaper, nonprofessionals are supplying us with
better originals. Which, I hope we would agree, is a good thing.
That's it from the front line. More in about ten days
when I summarize how the advanced classes are seeing the current situation.
Dan Margulis
___________________________________________________________________________
Trends in 2008 ACT Classes
Posted by: "John Ruttenberg"
Mon Nov 17, 2008 7:08 am (PST)
Dan Margulis:
Another question: we all have to work on substandard
originals from time to
time: photos shot by amateurs, or by professionals
under adverse conditions,
or just poor shots that nevertheless have to be used.
Some of us, however,
work on such images much more than others. Advertising
retouchers rarely see
them; those preparing newsletters for corporate
clients see them all the
time. Would you say that you have to color-correct
such images often, or
seldom?
I think the definition of "substandard"
probably plays a role here. The photograher/retoucher is in a very
different situation than the person who only retouches the work of others.
I'm thinking of my own work. As many of you may remember, I am a ballet
photograper at Boston Ballet, and have the privilege to photograph the
company's dress rehearsals. The professional theater lighting is very
beautiful, but ofen very challenging. About a year ago I posted a case
study featuring this shot:
http:
//colortheory.smugmug.com/photos/153552178_zNF9t-L.jpg
from the second act of Giselle. Is this a
"substandard" original? Well, if it just lands on your lap, I
guess it is. Dan used it to torture the members of his class last year
which is a good indication. On the other hand, I knew when I took it that
it would need substantial post work. And I was pretty sure that among the
photograpers present I was the only one with the equipment to get any
usable image under the circumstances (Canon 5D, 85mm f/1.2 @ ISO 1600). And
likely the only one with the patience and knowledge to make something
(other than a B&W) from what I got.
Moral: The pure retoucher's substandard may be just a
reasonable midpoint in the photographer/retoucher's workflow.
I wonder whether I am a special case here, pushing the
limits in a way others don't ofen need to do. Available light makes my job
harder and more rewarding when it works out. Studio photographer never has
these sort of issues. What about photojournalism? I'm a little confused
over the rules. How much retouching is allowed by newspapers and the like?
-- John Ruttenberg
___________________________________________________________________________
Re: Trends in 2008 ACT Classes
Posted by: "Jim Bean"
Mon Nov 17, 2008 8:21 am (PST)
dan wrote:
It was a pleasant surprise to find out otherwise. My
estimate last year was that 53 percent
of attendees paid their own way; this year it was a
significantly lower 43 percent. Good for
the companies who realize the best investment they can
make is in their people!
my take on these numbers is that more people/companies
are more aware of your classes and the associated benefits... in my neck of
the woods several companies either send their graphic teams to classes in
dallas/houston/similar that are actively marketing their classes.. the
classes many times are subsets of service providers/suppliers such as
lexjet.. the classes on a 'professional level' are bread & butter
basic.. attempting to get your monitor to match your (closed loop)
output... lots of 'super amatuears' in my area.. several times a year I
will receive calls about 'which camera-which printer' to buy for the wifes'
birthday... the client calls a camera shop, drops $10k+ on hardware and
magical software and the box arrives in two days... then the wife (after
burning through several cartridges and filling the dumpster..) goes to an
out of state 'retreat' class on 'how to use lightroom' taught by the author
of several PS books, again sponsored by printer companies..
I think your students are either understanding that
the nuts/bolts are not available at these other programs and either pay
their own way or take the 'benefits' approach to their supervisors and
say... "we can do better if you send me to dm's boot camp."
note: even after the many years have passed... how
many imaging professionals do you encounter today that even know what
'channel blending' is?
best, jim bean
___________________________________________________________________________
Re: Trends in 2008 ACT Classes
Posted by: "Ric Cohn"
Mon Nov 17, 2008 4:36 pm (PST)
On Nov 16, 2008, at 8:46 PM, Dan Margulis wrote:
More and more, it seems that people are in the field
for love, not money.
I'd say it's more likely that those in the field are
less and less likely to be making enough money. I suspect there are just as
many people as before doing it for money who also love the field, they're
just not making as much doing it. Meanwhile, the cost of entry for serious
amateurs (a decent digital camera, a powerful enough computer, lots of
accessible how-to info) has dropped dramatically in recent years.
I believe Dan did predict this general trend years
ago.
Ric Cohn
___________________________________________________________________________
Re: Trends in 2008 ACT Classes
Posted by: Stephen Marsh
Mon Nov 17, 2008 9:40 pm (PST)
John Ruttenberg wrote:
I think the definition of "substandard"
probably plays a role here.
Agreed, I think that we all face our own
"unique" challenges, most of mine of late (apart from colour
matching) are resizing 250pixel or 500 pixel wide images for output at at
400-600% enlargement size (2000> pixels at final size).
Stephen Marsh
___________________________________________________________________________
Upward resizing [Was: Trends in 2008 ACT Classes]
Posted by: "John Ruttenberg"
Tue Nov 18, 2008 7:39 am (PST)
Stephen Marsh:
... most of mine of late (apart from colour matching)
are resizing 250pixel
or 500 pixel wide images for output at at 400-600%
enlargement size (2000>
pixels at final size).
That seems like an interesting problem, one that a lot
of people have. I have to crop the ballet shots all the time, sometimes a
lot more than I might have hoped.
So I'd like to read more about how you address this.
Perhaps we can get a good thread going. Maybe even a case study.
-- John Ruttenberg
___________________________________________________________________________
Re: Upward resizing [Was: Trends in 2008 ACT Classes]
Posted by: Michael Jahn
Tue Nov 18, 2008 11:34 am (PST)
Hi John,
Some of my clients work on development of image
conversion - one is developing a paper to PDF solution for converting
scanned color text book page (math books and books with maps and images)
and then descreening and enhancing - this is not strick upsampling - but
this might be interesting to review;
http://www.ioflex.com/imagecompare/Compare.asp?Image=
00007.tif&Loc=2x3
(example of upscale threshholding of text)
http://www.ioflex.com/imagecompare/Compare.asp?Image=
00007.tif&Loc=2x1
(example of desceen without edge blur)
There are many algorithms that have been developed for
use on images captured via military satellites or in the face recognition
field that can be used for "enhancing" - some results are quite
astounding - of then this is done buy finding edges and creating masks
where different contrast adjustments are applies - similar in the way
un-sharp masking works.
http://www.cs.huji.ac.il/~danix/epd/
and
http://www.celartem.com/en/products/pixellive.asp
Hope this is the sort of thing you were speaking of
and helpful.
Michael Jahn
Jahn & Associates
PDF Color Conversion Specialist
___________________________________________________________________________
Re: Trends in 2008 ACT Classes
Posted by: Dan Margulis
Tue Nov 18, 2008 7:40 am (PST)
John Ruttenberg writes,
I think the definition of "substandard"
probably plays a role here. The
photograher/retoucher is in a very different situation
than the person who
only retouches the work of others. I'm thinking of my
own work. As many of
you may remember, I am a ballet photograper at Boston
Ballet, and have the
privilege to photograph the company's dress
rehearsals. The professional
theater lighting is very beautiful, but ofen very
challenging. About a year
ago I posted a case study featuring this shot:
http:
//colortheory.smugmug.com/photos/153552178_zNF9t-L.jpg
from the second act of Giselle. Is this a
"substandard" original?
I thought I gave a definition that was pretty clear
"photos shot...by professionals under adverse conditions." I am
not a professional photographer myself and so may be presumptuous in my
assumptions, but I was bold enough to believe that when a ballerina is
performing athletic maneuvers on a nearly completely dark stage where the
major illumination is a blue spotlight, compromised by softer light of
different color from the bottom, this constitutes an "adverse
condition", and thus is likely to result in what I called a
"substandard" original. It was not intended to be an invidious
term, but rather to indicate a surmise that photographs shot in such
circumstances are likely to be harder to fix than, say, a shot of the same
ballerina in a photographic studio.
Well, if
it just lands on your lap, I guess it is. Dan used it
to torture the members
of his class last year which is a good indication.
I still do. I put it in the advanced class last year,
along with eight other new images. My selections weren't well received:
only three of the nine made the cut for the 2008 advanced classes, but
Giselle was one of them. She got favorable commentary from the 2007
evaluations and also on the first two sets from 2008, so she's almost
certainly going to be in the 2009 classes as well.
The image is interesting not just because it is
technically challenging but because it is a question of artistic
interpretation. The principles of chromatic adaptation and simultaneous
contrast are not repealed in a darkened theater. We humans strive mightily
to adapt to any lighting condition. While there's no way we would see
normal fleshtones under such extreme conditions, we *would* see them as far
more normal, and more varied, than any camera would. The question is how
far to go in catering to this known effect, and there's no one right
answer.
I wonder whether I am a special case here, pushing the
limits in a way others
don't ofen need to do. Available light makes my job
harder and more rewarding
when it works out. Studio photographer never has these
sort of issues.
I wouldn't say "never" but otherwise agree.
Many of my students rarely face such retouching issues, too. I ask the
questions of the classes so I can know whether such images are relevant
enough to use as exercises. Nowadays, they appear to be, whereas maybe ten
years ago, they wouldn't.
What about photojournalism? I'm a little confused over
the rules. How much
retouching is allowed by newspapers and the like?
That question is a lot more complicated than it
appears. It is very difficult to come up with standards. Consider the
following reasonably well-known cases:
1) Wire service picture of a city under bomb attack in
the mideast. Additional plumes of smoke added as embellishment, suggesting
more violence than actually there.
2) Newspaper photo of two politicians chatting in
front of a table. Can of Coke on the table is retouched out.
3) Newsmagazine cover of Ronald Reagan during the late
years of his administration, highlighting a story suggesting that he was
out of touch. Allegation: that the retoucher deliberately enhanced contrast
and sharpening in Reagan's face in an effort to make him look older by
emphasizing his wrinkles.
4) Newsmagazine cover of the mug shot of O.J. Simpson
upon his arrest for murder. Allegation: that the retoucher, motivated by
racism, intentionally made Simpson look "blacker" and thus more
threatening-looking to a white audience.
Nobody in their right mind endorses #1, and somebody
justifiably got fired for it. Of the other three, the last two are far more
troubling from an editorial standpoint. #2 is the deletion of an irrelevant
object of no editorial significance. The problem is that evil intent in
color correction is impossible to prove unless somebody can come up with a
smoking gun in the form of a specific written instruction, whereas the
deletion of an object is easy to prove. So, the newspaper in #2
sanctimoniously began disciplinary proceedings against the individual
responsible for removing the can, on the ground that some sacred line of
journalistic integrity had been crossed.
Police mug shots qualify IMHO as "substandard
originals", particularly when they have to be blown up as large as
Stephen Marsh apparently has to do regularly. God knows what the color of
Simpson's skin was in the actual mug shot; it wouldn't be surprising if it
were green or blue. Did the retoucher try to match an overly dark photo,
was it just an accident caused by his lack of familiarity with this type of
retouching work, was it a neutral (if incorrect) surmise as to what Simpson
actually looked like, given an obviously poor photograph, or was it, in
fact, a racist act? There was too much rhetoric being bandied about for
anybody to actually find out, IMHO.
I don't know what actually happened in that instance,
but in the Reagan one, I was there, and I heard the art director reject the
first proof of the Reagan shot with the comment "Can't you make him
look older?"
We could, and did. What kind of rule can prevent it?
Dan Margulis
___________________________________________________________________________
Re: Trends in 2008 ACT Classes
Posted by: "Bevi Chagnon"
Tue Nov 18, 2008 1:22 pm (PST)
Dan M wrote:
That question is a lot more complicated than it
appears. It
is very difficult to come up with
standards. Consider the following reasonably
well-known cases:
I frequently teach a class on this topic, "When
does Photoshoping go too far." Here are links to some of the cases Dan
mentioned.
http:
//www.tc.umn.edu/~hick0088/classes/csci_2101/false.html
scroll to the bottom to compare OJ Simpson photos.
Dartmouth has an extensive list of photos, including
several Dan referred to:
http:
//www.cs.dartmouth.edu/farid/research/digitaltampering/
And Stalin was famous for his use of falsified photos:
http://www.newseum.org/berlinwall/commissar_vanishes/
In my classes (adult & college level), student
response has been pretty consistent over the years:
-- Photoshopping art & personal photos, OK.
-- Photoshopping advertising photos, OK within reason.
-- Photoshopping editorial, informative, political,
government, or news photos, not OK (other than color-correcting)
-- Photoshoping tabloid photos, definitely
encouraged.makes long checkout lines at the supermarket more enjoyable!
--Bevi Chagnon
........................................................................
Bevi Chagnon | Adobe ACE: InDesign CS2 |
www.PubCom.com
___________________________________________________________________________
Re: Trends in 2008 ACT Classes
Posted by: Dan Margulis
Tue Nov 18, 2008 8:04 pm (PST)
Jim Bean writes,
in my neck of the woods several companies either send
their graphic teams to classes in dallas/houston/similar that are actively
marketing their classes.. the classes many times are subsets of service
providers/suppliers such as lexjet.. the classes on a 'professional level'
are bread & butter basic.. attempting to get your monitor to match your
(closed
loop) output...
That's the problem with a lot of training these days,
more so than in the past. The service provider doesn't want to tick off a
client, any client. So if they run this session, and three or four people
gripe that it was too hard and they found the instructor's explanation of
what a layer is too fast, this is a failure. The session therefore gets
toned down to the least common denominator.
lots of 'super amatuears' in my area.. several times a
year I will receive calls about
which camera-which printer' to buy for the wifes'
birthday... the client calls a camera
shop, drops $10k+ on hardware and magical software and
the box arrives in two days...
then the wife (after burning through several
cartridges and filling the dumpster..) goes
to an out of state 'retreat' class on 'how to use
lightroom' taught by the author of several
PS books, again sponsored by printer companies..
And again, probably not aimed at someone who is there
to learn how to think and not to follow recipes.
I think your students are either understanding that
the nuts/bolts are not available at
these other programs and either pay their own way or
take the 'benefits' approach to
their supervisors and say... "we can do better if
you send me to dm's boot camp."
That's probably slightly more true than in past years,
but maybe not that much. Unlike all these other programs, we have a lot of
warnings in place about long hours, frustrating experience, unpleasant
instructor who allows students to fall on their faces rather than giving
them the solution up front, etc. Consequently, those attending have already
selected themselves as being ready for the experience, and we don't have a
lot of interruptions with beginner questions or explanations, even though
the classes often contain ambitious beginners.
One way or another the quality of the entering
students has definitely improved in the past few year. The curriculum
introduced in 1/08 is harder than the previous one because it seemed that
the attendees could take it.
What I'm hearing from recent attendees is that there
really isn't any substitute for watching other people do a better job than
we do and using that to ramp up our own skills. You may recall a
monochromatic image that rejoiced in the name of "Washout
Skintone" from the advanced course. That has gotten more positive
votes than any other, which is why it's one of only three exercises that
have been in the course since 1999. That means that right around 200 people
have had at it, most without a whole lot of success.
Last week in Washington, the exercises come in, and
dog my cats if one joker hasn't got a much better version than any of the
preceding 200, including the three or four times I tried it myself. I know
basically how he did it but at the moment am failing to comprehend why it
should have worked as well as it did.
On Saturday, a different group will be working that
image, and I will be too, and I expect to know a good deal more about what
makes it tick than I have in the last ten years. Which probably puts me in
the same position as most students often find themselves during the first
class.
note: even after the many years have passed... how
many imaging professionals do you
encounter today that even know what 'channel blending'
is?
That, I know the answer to, because the class
evaluation forms ask about it. Five years ago, we had a lot of people
coming in who'd heard tell that this LAB stuff was very powerful and they
wanted to know more about it, and the subsequent evaluations found it very
significant. Now, we're seeing the same thing with channel blending. So
simple and so powerful, we wonder how we could have missed it for so long!
Dan Margulis
___________________________________________________________________________
Re: Trends in 2008 ACT Classes
Posted by: Stephen Marsh
Tue Nov 18, 2008 8:04 pm (PST)
Bevi Chagnon wrote: I frequently teach a class on this
topic, "When
does Photoshoping go too far." Here are links to
some of the cases Dan
mentioned.
Thanks Bevi, while on the topic - one that I just
stumbled over in today's paper is linked here from the UK (note: if bikini
clad female models offend, don't follow the link):
http:
//www.news.com.au/dailytelegraph/story/0,22049,24668374-5001026,00.html
http:
//www.dailymail.co.uk/tvshowbiz/article-1086567/Victorias-Secret-model-Karolina-Kurkova-riddle-missing-belly-button.html
I thought a careless person cloned OUT her navel, but
they usually have to clone one IN!
Stephen Marsh
___________________________________________________________________________
Trends in 2008 ACT Classes
Posted by: "Eric Basir/Photo Grafix"
Tue Nov 18, 2008 8:04 pm (PST)
John, you wrote:
Moral: The pure retoucher's substandard may be just a
reasonable midpoint in
the photographer/retoucher's workflow.
I wonder whether I am a special case here, pushing the
limits in a way
others don't ofen need to do. Available light makes my
job harder and more
rewarding when it works out. Studio photographer never
has these sort of
issues.
What about photojournalism? I'm a little confused over
the rules. How much
retouching is allowed by newspapers and the like?
I agree with you 100%. There are exceptions to your
"moral" of the story. Albeit, they are rare. I'm sorry to burst
anyone's bubble. The best photos I've seen are produced by a photographer
who works with a retoucher. Photographers can only spend so much time with
post-production and prepress before they cut into their creative time.
Conversely, I've been in "assembly line" retouching operations
where we are totally disconnected from the photographer. That too, can lead
to problems.
I fall in the retoucher [only] category. I speak
wholly from experience working with a slew of different clients since 1999
(wow, it seemed like yesterday...). Retouchers are fastidious when it comes
to quality because we are responsible for whatever goes to the separators,
press or other final--or near final--destination. While photographers dream
about glorious images and compelling composites, retouchers dream about
stuff you read in Dan's books: Channel blending, densitometer readings,
TIL, layer blending, and rebuilding.
Whenever possible, photographers‹no matter how
talented they are‹should have a retoucher(s) who they know and trust
working with them. I'm not talking about an intern with a few months
experience in college either (smile). Of course there are many on this
list.
I wrote this short piece on our blog called Dao Te
Retouching:
?As the bee is to the flower, so a photo retoucher is
to the photographer
and designer. If the flower is to bear enough seed or
fruit for next season
and the bee needs plenty of pollen for its young, they
must work together.?
Sometimes it's not so much that a retoucher will
"fix" things. He or she may
be able to offer some tips or advice to overcome the
problems which lead to
the symptom.
Regarding photojournalism, it's always bad business to
retouch your pictures
(I worked for quite a few papers in the 1990s). Unless
you are scanning film
and need to remove fuzz, dust or digital anomalies, I
would recommend doing
nothing more than a two-step contrast and color cast
fix. Maybe a little
darkening around the corners to draw the viewer's eyes
in. But be careful.
Some links to help get specifics:
http://www.nppa.org/
http://www.editorandpublisher.com/
--
Eric C. M. Basir
Photo Grafix
http://www.photografix.pro
1-847-673-7043
___________________________________________________________________________
Re: Trends in 2008 ACT Classes
Posted by: Dan Margulis
Thu Nov 20, 2008 11:55 am (PST)
Ric Cohn writes,
I'd say it's more likely that those in the field are
less and less
likely to be making enough money. I suspect there are
just as many
people as before doing it for money who also love the
field, they're
just not making as much doing it. Meanwhile, the cost
of entry for
serious amateurs (a decent digital camera, a powerful
enough
computer, lots of accessible how-to info) has dropped
dramatically in
recent years.
This is an old, old story that has been discussed
several times on the list. As time has gone on, the value of taking a
photograph (and by this, I mean the act of clicking the shutter, not
something else) has decreased drastically. Sensible photographers have
responded by diversifying. Those who are prospering are usually doing so by
riding revenue from additional services not historically provided by
photographers, such as color correction, or managing the commercial
printing of their clients' work. The trend has been discussed at some
length in a couple of threads, such as
http:
//www.ledet.com/margulis/ACT_postings/DailyLife/ACT-Dans-photographer.htm
and
http:
//www.ledet.com/margulis/2007HTM/ACT07-Dans_photographer.htm
I believe Dan did predict this general trend years
ago.
Yes, from about 1996 on, at a time when photographers
were in denial. I just got a chance to read Alessandro Bernardi's Italian
translation of "An Infinite Number of Monkeys", the 2001 update
(available in English at http:
//www.ledet.com/margulis/Makeready/MA46-Infinite_Number.pdf )of the earlier
columns. Just as an infinite number of monkeys with an infinite number of
typewriters will eventually produce the poetry of Robert Burns, give the
INOM an infinite number of quality digital cameras and one of them will
produce professional-looking results. This, I pointed out, is very bad for
the true professionals, because clients are often looking for *one* good
picture, and do not care whether the person who offers it to them got it
the first time he shot it or had an infinite number of failures first.
This wasn't a hard prediction to make IMHO not just
because the equipment was becoming more readily available, but because the
field is such an enjoyable one. Few jobs in this world are so attractive
that people would work them for free or for a trivial amount of money. Our
field, as we have seen, is the exception. Couple the drastic improvement of
the equipment with a large number of people who are so enamored of
photography that they care very little about how much money they make off
it, and it's bad news for those who *do* care how much they make.
While it is certainly heartening to work in such a
compellingly interesting field, the closing comments of that 2001 column,
with an assist from Burns, still hold:
Those who can produce images effectively for a variety
of
purposes from a variety of sources will be able to
prosper, if
not as photographers, then at least as service
providers with
a strong specialty in photography.
The attack of the infinite number of monkeys is indeed
fearsome. The twenty-first century photographer has the
choice of preparing for battle, or of huddling in the
cold,
waiting for events to overtake him:
But, och! I backward cast my e'e
On prospects drear!
And forward, tho' I canna see,
I guess and fear.
Dan Margulis
___________________________________________________________________________
About a month after the above, I posted a similar
message discussing events in my advanced classes.-DM
From the 2008 advanced courses
Posted by: Dan Margulis
Thu Dec 4, 2008 6:11 am (PST)
In October and November I teach advanced courses, open
only to those who have already suffered through the three-day ACT course.
These are the most valuable classes of the year for me because of the skill
level of the students. The first class exposes a lot of of errors in
technique. In these advanced classes it's unusual to find students making
outright errors, so it's more of a comparison of techniques without
introduction of extraneous silliness. There's no better way to get a
definitive answer about how to get the best results from a certain image
than to assign it to 23 experts to try to correct it with a method of their
choosing and compare the results. That's how I've validated a lot of my own
techniques over the years--I use them in these classes before the public
knows about them, and see whether I'm doing better than all these
knowledgeable folk who don't have access to them.
Once the classes they're complete (and the last one
was last week) for the past few years I've discussed what was learned both
from the standpoint of color correction but in terms of trends in our
field. People who are so committed to quality that they suffer through a
three-day session with me (and sometimes more; seven of the 23 attendees
this year had taken the advanced course in prior years) and still want more
are probably worth listening to. Also, a high percentage of the people who
took the advanced courses are members of this list and may be interested in
hearing an overall wrapup, so I'll go on at a little more length than in
past years.
CLASS MAKEUP.
Obviously the main topic of non-color correction
discussion was about the economy. Before talking about the disastrous last
couple of months, the very makeup of the classes said a lot about trends in
the last couple of years. For the first time in memory two groups were
totally missing: newspaper production people, and professional
photographers-- both, presumably, victims of the economy. Their places have
been taken by the continually growing number of "super-amateurs",
people who have all the attributes of professional photographers, including
occasionally selling their work, but who are professionally successful in
some other field (often medicine or law) and so are not constrained to
support themselves with their hobby.
We had six such people in the classes this year. Two
others had different legitimate reasons for needing to know how to
color-correct but were also ultra-serious photographers, so more than a
third of the attendees fell into the category of super-amateur.
Understanding that some folk are difficult to classify, I arranged them as
follows:
*7 people work in what might somewhat inaccurately be
termed an in-house setting. They work with specific types of image for a
specific purpose.
*6 super-amateurs, as described above, who AFAIK
rarely work on anything but their own photos.
*5 persons who do retouching for a service provider
and thus accept jobs from the general public. None were from commercial
printers, however.
*4 freelance retouchers.
*One person who falls into a unique category: a
bestselling author of cooking books who, not surprisingly, wants the food
pictured therein to look appetizing and doesn't trust anybody else to make
it happen.
THE ECONOMY.
Everybody is affected, but some people have not yet
been personally hit by it. The freelance retouchers say their clients have
advised them that things are drying up, but their own projects still are
proceeding as scheduled. The service providers similarly: no huge downturn
just yet, but the signs are not promising.
And that (along with the absence of professional
photographers and newspapers) was the basic feeling. An eerie absence of
any positive feeling and ample reason for foreboding. Even the people who
are being helped, if you can call it that, by the situation, tell the same
story. One super-amateur is a professor at a large university. Applications
from people older than typical students, he reports, are way up as people
who cannot find work decide to make use of their free time. Another is a
physician who performs life-saving surgery. When a life is at stake, the
hospital cannot ethically withhold its services whether the patient is able
to pay or not. This student works in an area hard-hit by the economy, and
he reports that a large percentage of what he does is now not paid for by
anybody, meaning that it's paid for by the hospital, and eventually by us.
And a field that we wish would not get any bigger:
some students worked in a police forensic lab. All those of us who complain
that we have to work poor-quality originals should be condemned to process
images shot at a crime scene in the middle of the night by a police officer
who wouldn't know an F-stop from a lens cap. Unfortunately, they are
expecting a rise in the number of images they need to so process, due to
the economy. On a more interesting note, we got to work on some evidence
photos, trying to enhance fingerprint detail and trying to resolve marks of
blood found on a distractingly colored background. In doing this, LAB
techniques were quite successful. Being significantly more magenta than
green, blood can be isolated in the magenta half of the A channel, which
can then produce a mask that will knock almost everything else out.
And an even more disturbing note: some people who
think they are exempt from these economic times find out the hard way that
they are not. One such student worked for a graphics division of a
well-known large company. On the first day of class, she advised us that
there had as yet been no impact. A few hours later the phone call came,
announcing that the company had laid off a thousand people.
As for me, I continue my warning that this is the
worst economic climate by far in the last 75 years, and that it's only
going to get worse in the immediate future until a panicked populace starts
to spend money. In the last few days I've had to visit my car repair shop
and my dentist, two operations that you wouldn't think would be affected,
but they are. Both report that their clients/patients are radically
reducing the number of non- emergency visits. They're waiting until the car
fails totally, or the tooth starts to hurt. My cousin has to relocate, and
he found a newly-built luxury residence in the city he's going to, but
could not afford the million dollars the builder was asking. As a lark, he
offered around 40 percent less, cash on the barrelhead--and the builder,
presumably unable to pay his own suppliers, took it.
With as much travel as I do, I can draw some
conclusions about how certain industries are being affected. In all three
of the cities I just visited, the restaurants were empty everywhere. Worse,
we're seeing desperation pricing of some services. Since my trips are
scheduled so far in advance, I usually book hotels more than a month out.
Then, before leaving, I check rates again to see if better bargains are
available.
I had made the mistake of scheduling a Washington
class over Veteran's Day, when hotels are usually full. So, I wound up
paying more than I would have liked, and in fact would have paid even more
if I had booked at the last minute. But in Chicago and Toronto, food grief.
I found hotel rooms in both at very desirable establishments for $115,
which is about half of what these two usually charge, and far less than
I've ever paid at either one before.
Tonight I have to pick up a car at the San Francisco
airport for a week. My reservation was going to cost me about $225, which
is pretty normal. A few days ago my rental car company sent me an e-mail
stating that due to my status as a preferred customer and all-around nice
guy, they were pleased to offer me $40 off the next rental I made, provided
that it occurred in December, in addition to my customary upgrade. Of
course, that caused me to cancel the existing reservation and rebook it
using my $40 discount. But when I did so, I found that the base rate itself
had been reduced to $100. So, I am about to rent a full-size car, in
California, at $60 for a week.
THE STATE OF THE FIELD.
Each year I ask a series of questions as to whether
things have gotten better or worse in certain areas, with a view to seeing
whether there was any consensus. Here's what the advanced class attendees
said this year.
*It happened that only three attendees deal with
multiple commercial printers. Therefore, one can't draw too much from the
responses to my question as to whether printers are today more quality
conscious and/or more knowledgeable than they were too years ago,
especially since one person answered yes, another said they are worse, and
the third said they are about the same.
*A lot of people receive files from photographers,
however, and here there *was* a consensus that the quality is considerably
higher than it was two years ago. The speculation was that this is largely
the impact of better equipment being available to more people.
*A surprise: two retouchers said that a majority of
their originals came in in raw format. I had been under the impression that
few photographers would agree to this, on the grounds that it makes their
work too much of a commodity and takes them out of the creative process. I
surmise that the agencies are telling them raw or nothing.
*Those who have purchased Photoshop CS4 have reacted
negatively. Everyone notes the speed differential and the new stability of
Bridge. This is not seen as compensation for having to relearn adjustment
layers, particularly given that the most important adjustment for this
group, curves, is considerably less efficient than in previous versions.
*As to clients, the group confirmed most of what has
been said by advanced classes in the last three years: expectations for
turnaround times continue to decrease. This time, unlike past groups, we
did not hear about pressure to reduce prices. OTOH, this group repeated the
contention of last year's that clients often have no idea how difficult
certain Photoshop maneuvering is and believe that almost anything can be
"Photoshopped out" at no charge.
COLOR CORRECTION.
The advanced classes not only work more images than a
basic ACT class does, the images change more frequently from year to year.
Since 1999, advanced classes have always worked 26 images, but some get
discarded each year and replaced by new ones that I think may be
instructive. This year I replaced 10 of 26, which is a new record.
Nevertheless, with 16 holdovers from previous years, where I had saved the
best results, it's possible to make some comparisons.
Last year's group was outstandingly strong, full of
professional retouchers. This year's did not have that experience, although
there were a couple of stars. By and large it did not do as well in
exercises that were technically difficult and required imaginative
treatment. However, it seemed to do better than previous years in examples
that had no serious defect, such as scenics, or where the allocation of
color as opposed to contrast was more critical. This IMHO was due to the
influence of the picture-postcard workflow.
The difficulty this year was that the advanced class
attendees fell into three categories: those who had taken ACT prior to
mid-2007, and had thus not seen the PPW as such; those who attended in late
2007 and therefore had some exposure to it, and 2008 attendees who went
through an ACT that had been overhauled to emphasize that workflow. In the
first half of the class the later students did better than those that had
taken the course earlier, indicating the efficacy of the new moves.
I noted the following:
*Everybody is seeing the value of an extended
luminosity step while in RGB: a consideration of channel blending and/or
contrast-enhancing curves on a luminosity layer as opposed to the
traditional method of one correction pass that improves both contrast and
color simultaneously.
*In consequence, the color often looks tired after the
luminosity step. In traditional workflows, the very act of establishing
range improves color. Now, we're doing a better job of establishing range,
but omitting the desirable color improvement. When an image has no problem
other than overall unimpressive color (and that's where we usually are,
after the luminosity step), LAB has a decisive advantage over either RGB or
CMYK. So I would say that people found themselves in LAB at some point in
about 90 percent of all images. This is a much greater use of LAB than in
previous years. Interestingly, we are now in LAB almost always to exploit
the A and B--while the L channel is useful, it's now being reserved for
slight adjustments only.
*Continuing, but intensifying, a trend in workflow
that we've seen in the last year: now that most files eventually get to
LAB, we're finding it more useful to make the image *too* colorful and then
figure out an intelligent way to back off, rather than the traditional way
of trying to accent colors that aren't vivid enough. In almost every
exercise now, I've got some layer on top that's WAY more colorful than I
finally intend the image to be.
*In those images where the above workflow was
appropriate, those who followed it murdered those who did not. The problem
was in identifying the images in which it *wasn't* appropriate.
*There was much more use of the Man from Mars Method
this year than last, owing to the emergence of new techniques to control
undesirable changes, often by partial blends of the original A and B
channels to prevent the image from getting too cold.
*We had better results in noise reduction than
previous years. I am now recommending that shadow noise in underexposed
pictures be reduced by the use of Surface Blur at an early stage of the
process.
*We had better results with hiraloam sharpening than
in the past. I now recommend that immediately after this type of sharpening
we blend the original into the sharpened version at 50% Darken mode. This
corresponds to my practice with conventional, low-radius sharpening. More,
however, it now seems clear that hiraloam is not very effective in the
three-quartertone to shadow range. I therefore suggest doing the hiraloam
on a layer with what seems to be an absurdly high Amount--say 200% Amount
and 30 Radius--but loading a layer mask from a luminosity-based channel to
reduce the impact in darker areas, before finalizing the move by reducing
layer opacity.
*In comparison to past years, the 2008 classes did
best in images calling for aesthetic judgment, such as Lake & Fountain
and Over the Colorado. They did worst in images requiring imaginative
solutions to specific technical problems, such as Giselle and
Mother-in-Law. The only time the group actually crashed and burned was on
one such image, Purple Skintone. We got 23 unsatisfactory solutions to a
problem that wasn't all that difficult to resolve.
*The images used in the advanced course are initially
chosen by me, but they only stay in the curriculum if they get votes on
post-class evaluations. If the classes say they're learning from them, they
stay in, whether I approve of them or not. But each year I dump some of the
low scorers and replace them with new, hopefully more interesting
challenges.
My recent selections have suffered from bipolar
disorder. Of the eight images new in 2006, six are still in, and three of
these (Mother-in-Law, Chinese New Year, and House Fire) are in the top five
of all vote-getters. By contrast, my 2007 selections fell flat. Of the nine
introduced last year, six didn't make it to the 2008 classes. John
Ruttenberg's Giselle image continues to be popular of the three remaining
ones.
The 2008 selections fared somewhere in the middle.
There were no stars, but eight of the ten new images got enough support
that they'll probably be seen next year. Of the other two, the classes
didn't find much use for the Deer in Undergrowth exercise. I disagree, and
will probably retain it. OTOH, I concur with the classes' judgment that the
Vietnamese Taxis exercise was not an inspired choice.
*Since 1999, every advanced class has been required to
correct a set of three color images with the monitor turned to grayscale.
They do not get to see what the color looks like before beginning. For
competent people this is not difficult, and the results, year after year,
aren't much worse (if at all) than if they had been able to see the color.
(Back when I was a supervisor, I used to force recalcitrant employees to
work this way if they were making foolish errors by failing to check the
numbers.)
Traditionally classes are split radically over whether
this is a worthwhile use of time. Many professionals, realizing that
working this way isn't hard, find it boring. Those who have never
considered the possibility, however, often find it an epiphany, the best
part of the class, an experience so exhilarating that they go home and
correct more images in black and white.
Over the years, I haven't put anything calling for
tough color decisions in this group. But to cater to the bored people, last
year I threw in an exercise that clearly calls for vivid color and artistic
judgment, a shot of a setting sun playing over water with a European city
in the background. That kind of shot is challenging to correct on a
grayscale monitor. Given the approval of that image last year, I added a
similar one this year: a glacial lake that the students are told is known
as one of the most lovely in the Canadian Rockies, so much so that the
photo shows several painters on the shore trying to record it. The name is
Lake Beauvert (beautiful green, in French) which puts us on notice that the
color may not be typical of most lakes.
For whatever reason, this segment of the class was
considerably more popular with this year's group than it usually is, with
many positive comments on the evaluations. And there was one rather
irritating result, in the third (Toronto) class. Of the three images being
corrected in black and white, two had also been corrected that way in
previous years, so I had successful efforts of others to display in the
event that this year's class didn't get good results themselves. As the
Lake Beauvert image was new and I wanted a "par" correction with
which to compare the class efforts, I saved a copy of the one I did when I
demonstrated the image at a recent lecture in Calgary.
When the class works in color, I usually participate
and enter my own version into the contest, not always successfully. When
somebody else gets a better result than I do, that's great, it's a learning
experience. I never expected, however, that on an image like Lake Beauvert,
where a lot of aesthetic judgment is called for, a student working on a
grayscale monitor might do better than I would with the colors visible. But
it happened.
*And how do I know that the other guy's version was
better? Because everyone in the class, including me, agreed on the point.
That color correction really is a skill, almost an art form, was confirmed
again by the voting on the results of the exercises. In both regular and
advanced ACT, eight or nine versions of the same image are compared, and
the class selects the best of the bunch. Since color is supposed to be so
subjective, students are surprised to find that the vote is usually
unanimous. Around a third of the time, there's a disagreement as to which
of two is the best. Less than ten percent of the time does the disagreement
involve more than two versions.
One might think this would not hold over into the
advanced courses. In the basic course, a lot of people knock themselves out
of the competition with outright errors, such as failing to set proper
endpoints or leaving impossible colors in critical places. So,
realistically there may only be four or five versions that the voters would
even consider. In the advanced courses, such errors are rarer, so there is
a bigger field of contenders to choose from.
Nevertheless, in advanced courses the pattern is the
same: narrowing down the field takes longer, but the final determination is
unanimous more often than not. In this field, quality *does* count--as we
found out in the second (Washington) class. The Washout Skintone exercise
(light, overly yellow professional model with various complexion problems,
easy to make the hair too colorful) has been around since 1999 and rates #3
on the votes-per-class scale of approval. We've had, therefore, around 200
different corrected versions of this image submitted over the years. It's a
tough technical exercise, and the 2008 classes didn't do expecially well
with it, with a couple of exceptions. One of these, stunningly, was better
not just than the rest of the class, but than any of the 200 previous
versions, including three or four I'd done myself. Again, this qualitative
difference is not just my opinion, but that of everyone who looked. Even
having seen the version and knowing his basic approach (it depended on a
couple of Man from Mars moves with some interesting masking) I couldn't do
as well in the third class.
But next year I will, for sure. That's the lasting
allure of getting groups as talented as this together, it forces us to keep
learning. All good things, however, must end, so I reiterate that the plan
is that the November 2009 advanced classes will be the last ones. If you
have taken ACT (or have reserved a spot in a 2009 ACT course) and would
like the advanced class invitation next month, let me know offline.
Meanwhile, thanks to all who attended for a
stimulating experience.
Dan Margulis
Note:About a month after this was written (i.e.
January 2009) it was decided not to offer advanced courses in 2009 because
of the economic circumstances. The final advanced courses will be held in
2010--DM