Dan Margulis Applied Color Theory

Sheetfed and Web Profiles

   Date: Wed, 3 Nov 2004 10:35:58 EST
   From: Dan Margulis
Subject: Re: sheetfed vs. web coated SWOP profiles

Before I left for Asia, Ric Cohn wrote,

I just tried this with a file of mine. It is definitely true that if I
take a file converted with a custom Photoshop CMYK (where I don’t
change the ink definitions) and then do an assign profile for Swop(V2)
that the near neutrals and highlights all preview as much bluer. I’d
also say that this blueness is not (with my limited experience) what I
would expect if I sent this file to most printers.

That’s correct. The CMYK file that was being discussed to start this thread has been printed by at least ten different printers in at least five different countries. If you view the CMYK file on *any* reasonably calibrated monitor with the v.2 profile loaded, it looks way bluer than *any* of the actual print results.

The fact that the profile is defective in this area is no great surprise. Pastel colors are extremely difficult in CMYK. Many machine-generated profiles overstate how bright they are. If you compare a file full of pastel colors with the SWOP v.2 profile vs. a default Custom CMYK one, the Custom CMYK will be more accurate.  

Of course, there are other types of image in which the SWOP v.2 profile will do better. Plus, there’s a bright side: since it overestimates light blues, it will tend to yield warmer-looking seps, which can be a good thing.

The problem, of course, is that we shouldn’t have to live with the issue if we don’t want to. If Photoshop would give us a way to edit the profile, we could leave it alone if the extra blue doesn’t bother us and eliminate it easily if it does. Until that editing capability comes, these prepackaged profiles are not an option for quality-oriented CMYK users.

Dan Margulis
___________________________________________________________________________

   Date: Wed, 03 Nov 2004 09:57:12 -0700
   From: Andrew Rodney
Subject: Re: sheetfed vs. web coated SWOP profiles

on 11/3/04 8:35 AM, Dan Margulis wrote:

The problem, of course, is that we shouldn’t have to live with the issue if
we don’t want to. If Photoshop would give us a way to edit the profile, we
could leave it alone if the extra blue doesn’t bother us and eliminate it easily if it does.
 
I agree with you but I don1t see that ever happening.
 
 Until that editing capability comes, these prepackaged profiles are
 not an option for quality-oriented CMYK users.
 
There are third party products that can do this. While I1d love to see that in Photoshop, the option to edit profiles exists. Kodak is releasing it1s excellent profile editor once again that runs IN Photoshop.

Prepackages profiles that ARE created for the output in mind will work fine. Prepackages profiles that are not will work poorly. This isn1t anything earth shatteringly new. There are all kinds of areas Photoshop can1t deal with in this realm; device links, n-color profiles (profiles that have more then four channels). Be nice if Photoshop could do this, it can1t. Still no reason to throw the baby out with the bath water. There are solutions out there.

Andrew Rodney
http://digitaldog.net
___________________________________________________________________________

   Date: Wed, 3 Nov 2004 15:23:00 -0500
   From: Ric Cohn
Subject: Re: sheetfed vs. web coated SWOP profiles

Dan-

Thanks for the response. I think the issue is that many images (perhaps even all of Adobe’s original test images) would probably look fine converted with this profile.

The image I looked at was a cool silver colored piece of electronics. If the actual typical press reproduction really varies that much from how the file previews I don’t see how this can be considered a good-enough profile. Dan’s explanation helps me understand that many/most files would not look that different in the preview and that many typical images would not be adversely affected on press by some added yellow, but if I had adjusted my file to correct from the preview and then ran it on a typical yellowish-white press stock, a silvery piece of stereo equipment would look like piss!

I also think that Adobe should beef up the CMYK engine a bit. I don’t agree that CMYK output is so unusual a task for Photoshop that it’s users should be expected to go to a 3rd party product to do it correctly. I know Andrew Rodney is an expert in using these programs, but I really don’t know where to begin in evaluating the pro’s and con’s and then going through the learning curve to use the specialized solutions- some of which cost more than Photoshop!

Ric Cohn
___________________________________________________________________________

   Date: Wed, 03 Nov 2004 14:38:02 -0700
   From: Andrew Rodney
Subject: Re: sheetfed vs. web coated SWOP profiles

on 11/3/04 1:23 PM, Ric Cohn wrote:

 I know Andrew Rodney is an expert in using these programs,
 but I really don’t know where to begin in evaluating the pro’s and
 con’s and then going through the learning curve to use the specialized
 solutions- some of which cost more than Photoshop!
 
You shouldn1t have to. If service providers would provide service (give you the right profiles for good conversions in the first place) you wouldn1t have to jump through all these hoops. Photoshop has the capabilities of producing fine conversions. Just give it the right recipe. It1s not however going to provide a means of building that from scratch (you1d need a pretty expensive Spectrophotometer to do the job any justice). As for the old and in many cases scary Classic CMYK engine, I suspect Adobe would love to make it disappear but will not anytime soon. But don1t expect it to be beefed up either. It1s got the same capabilities it1s had since day one with plenty of potential hurt-me buttons.

Once you have a good conversion recipe, making this work is a simple as picking 3Mode Change-CMYK2. Getting that recipe is the hard part. If printers would standardize or provide recipes for their non-standard conversions, this would all be a lot less complicated for most users.

Andrew Rodney
http://digitaldog.net/
 ___________________________________________________________________________

   Date: Wed, 3 Nov 2004 17:48:02 -0500
   From: Ric Cohn
Subject: Re: sheetfed vs. web coated SWOP profiles.

I’m sure I’m showing my ignorance, but what the heck— I’ve done it before...   I believe that printer’s should either print to recognized standards (and not just say they do) or, better yet, supply profiles for their true printing conditions. However, real world, I don’t see this happening that fast for the majority of printers. I would love to just work with those that do, but reality is often otherwise.

I don’t understand why there can’t be some way to see what the expected conditions of an icc profile like the Photoshop SWOP V2 webcoated are and make small changes in the expectations. It seems to me the most frequent changes one wishes to make are in dot gain, total ink and GCR amount. As long as Photoshop can’t edit these profiles why can’t a family of icc profile’s at least be supplied with reasonable variations in these variables?

Thanks,

Ric Cohn
___________________________________________________________________________

   Date: Thu, 04 Nov 2004 16:55:04 -0000
   From: “John William Lund”
Subject: Re: sheetfed vs. web coated SWOP profiles

Ric Cohn wrote:

 It seems to me the most
 frequent changes one wishes to make are in dot gain, total ink and GCR
 amount. As long as Photoshop can’t edit these profiles why can’t a
 family of icc profile’s at least be supplied with reasonable variations
 in these variables?
 
—  Steve Upton has done just that, generated a series of output profiles based on the  TR001 data set, with varying Kgen & total inks. Look around his website, CHROMix.com,  for a link to download them.

John
___________________________________________________________________________

   Date: Fri, 5 Nov 2004 00:09:50 -0800
   From: “Darren Bernaerdt”
Subject: RE: Re: sheetfed vs. web coated SWOP profiles
 
John,

The “Profile Central” section of the chromix.com site has been down for a “server upgrade” for a long time. Is there another source for these profiles that you are aware of?

Darren Bernaerdt
___________________________________________________________________________

   Date: Fri, 05 Nov 2004 12:37:09 -0000
   From: Stephen Marsh
Subject: Profile Variations

It would be good if someone could make a range of GCR and Ink Limit profiles for public distribution from published data for common conditions such as TR001, Euro etc.

The Chromix TR001 profiles are a good example, as they offer a range of GCR and UCR and various ink limits - but no heavy GCR. My only real issue is that the perceptual intent has a contrast kick in it which departs from the original. With low key images this can be good, in an artistic sense. I like the Adobe v2 profiles in that the perceptual intent is tame in that it does not overly affect in-gamut colours with special secret sauce. There are both good and bad examples of freely available press profiles, what is considered good and bad often depends on the user and their expectations and setting.

Stephen Marsh.
___________________________________________________________________________

   Date: Fri, 05 Nov 2004 15:10:24 -0000
   From: “John William Lund”
Subject: Re: sheetfed vs. web coated SWOP profiles

—-  “Darren Bernaerdt”  wrote:

 The “Profile Central” section of the chromix.com site has been down for a
 “server upgrade” for a long time. Is there another source for these profiles
 that you are aware of?

—  Hmmm, you’re right. Sorry, I don’t have another source.

Maybe email CHROMix? Steve may check this list from time to time, but I know he’s quite busy trying to finish ColorThink Pro...
___________________________________________________________________________

   Date: Fri, 5 Nov 2004 17:23:47 -0800
   From: “John Feld”
Subject: Re: sheetfed vs. web coated SWOP profiles

As has been said previously, there seems to be an amazing lack of awareness of  about accurate color in the industry as a whole. Many of the nationwide printers have people who are very knowledgeable about color, but they are not in every plant, and on a more local level it is often extremely hard to find a shop that can even supply profiles, let alone having workable ones. When they are supplied, the prepress people tend to accept it as gospel, and don’t usually even give more than a cursory look at jobs as they go through. Dan talks of reducing tweaking to a minimum. In my experience tweaking is a rarity.

All the theory in the world is no good if it is not used in the trenches. And for the general design house, print buyer and the end customer this is a shadow world. Hence Newsweek can have that orange blob for a sun. How can we get even the basics to be understood and appreciated out in the real world? Previously the prepress hardware suppliers (Scitex, Hell, Agfa) included training with their RIPs and imagesetters, which included some color theory and real-world practicality. Now it seems that 99% of the people doing prepress are less well educated.

John

_________________________
John Feld
PC Graphics Report
www.pcgraphicsreport.com
___________________________________________________________________________

   Date: Tue, 09 Nov 2004 22:32:21 -0500
   From: Dan Margulis
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

John Feld writes,

How can we get even the basics to be understood and appreciated out in the
real world? Previously the prepress hardware suppliers (Scitex, Hell, Agfa)
included training with their RIPs and imagesetters, which included some
color theory and real-world practicality. Now it seems that 99% of the
people doing prepress are less well educated.

It may seem that way, but it isn’t true. It isn’t directly comparable because so many beginners have entered the field because it has become so cheap to get a quality digicam and a home office setup capable of professional-level image processing. If you exclude these people, and speak only of the kind of folks who are serious enough about the field that they would probably be involved in it even without today’s temptations, then the skill set is considerably better, IMHO, not worse.

I just finished teaching an advanced color theory class where we discussed this very point. The students were highly skilled, and as it happened many were from operations that accept input from a variety of sources. I put up for discussion the question of whether original images are better or worse than they were two years ago, and got a unanimous response, with which I concur, namely: the average original image is worse, however the average original image supplied by a *professional photographer* is better.

Similarly, in my own classes, the average attendee is considerably more sophisticated than persons of similar experience would have been only a few years ago. Those people who are serious about imaging know more about the subject than ever, and we shouldn’t blind ourselves to this fact by considering all the johnny-come-latelys.

Dan Margulis
___________________________________________________________________________

   Date: Wed, 10 Nov 2004 18:35:54 -0500
   From: Bob Johnson
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

On 11/9/04 10:32 PM, “Dan Margulis” wrote:

It may seem that way, but it isn’t true. It isn’t directly comparable because
so many beginners have entered the field because it has become so cheap to get
a quality digicam and a home office setup capable of professional-level image
processing. If you exclude these people, and speak only of the kind of folks
who are serious enough about the field that they would probably be involved in
it even without today’s temptations, then the skill set is considerably
better, IMHO, not worse.

Dan

From a different side:

I work for a large pre-press organization, and we’re finding the talent pool to be very shallow.

The “professional photographer”, and I have many friends in that trade, is certainly providing better files than you would get from all but the finest ‘chromes and drum scanner. Armed with your classes and books, they are ahead of most pre-press operators. Our department averages about 15 years experience. We have a very high end digital proofer, Gretag Spectrolino, all the toys. Two of us know how to use them and none of the other strippers, color guys, etc have a clue, or even care to know.

We are training ad agencies on Adobe CS programs, and our own guys barely can find their way thru them. First thing most of the strippers do is export to Ill 8.

We provide press profiling, but most of the clients refuse because they can’t print the same way twice and know that we will disallow their complaints when we tell them the job was made for the profile and they can’t hold their press in tolerance.

We just installed a big “application” from one of the last big pre-press companys. No one wants to use it because it’s new and we don’t have time for mistakes. Customers demand shorter turnaround and lower prices. We provide a web site for customers to view and approve their jobs. The customer calls in wanting to know when they can have delivery and we tell them “as soon as you approve the job, we’ll go ahead”. They tell us they don’t have time to look at a web site, it’s just a copy change anyway, so what’s the big deal?

Customer Service people are mere order takers. They have NO knowledge of the trade, wouldn’t know Magenta from PMS 485. We are lucky that they are good order takers, and seem to have a high tolerence for pain, as we make them call the customer to ask why stripping only has CMYK files for an 8 color job.

We have tried to hire “pre-press people” to little avail. The rare case of a ‘designer’ who wants to “get under the hood” usually fails. The young kids worked on the college newspaper, so they have a feel for Quark, but have no idea what happened after they pressed the print button.

Hate to sound so bad, but it’s pretty grim on the production side. Wages are stagnant if you’re lucky, the ‘old-timers’ have all retired, and customers think that they’re paying too much unless it’s free. One customer asked for a 2% move in the magenta highlite, and then didn’t want to pay for it because he couldn’t see much of a difference.
___________________________________________________________________________

   Date: Thu, 11 Nov 2004 10:34:33 -0500
   From: Terry Wyse
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

I’d have to agree with John Feld and Bob Johnson. As a color management consultant, I have the opportunity to visit quite a number of prepress/print shops across the country and I really do think the overall knowledge level has dropped in the past 10 years or so. I think one of the key reasons is lack of training. Back in my production days in the 80’s and 90’s, my employers were pretty good about sending me off to whatever training class (usually sponsored by dealers and vendors) I happen to have an interest in (imagine today being sent off for TWO WEEKS of scanner training!). Now a days, there’s very little training going on due in large part, I think, to the myth that the “computer” has made learning all these applications and skills so much easier. I would dare say that most folks application skill levels have stagnated to what they were doing about 3-4 versions ago and they’ve been playing catch-up ever since.

Dan, your impression of higher skill levels evidenced by the folks attending your classes is sort of a self-fulfilling prophecy. OF COURSE they have better skills because either they personally or their employers has seen that they’re getting the training/education they need. I would also be real curious what percentage of your students are in-the-trenches prepress folks as opposed to pure creative types, photographers and image editing specialists.

I don’t want you to take this wrong, but there’s not a whole lot of your techniques that can be used in a pure prepress production environment. Not because they AREN’T useful or productive, but the prepress person today is rarely allowed the opportunity to use these advanced techniques, even if they knew how to use them. Much of this has to do with today’s economics. These garage designers and photographers that think they know perfectly well how to prepare a job for a commercial printer generally aren’t willing to pay for the prepress folks to fix these jobs. So the prepress person’s “job” today is to simply shuffle these ill-prepared files through their production RIPing system and hope for the best. The tighter turn-arounds today compared to even a few years ago have made it virtually impossible to turn out a really good print job in the time required. I personally think that today’s technology (better apps, better color management tools, etc.) has raised the quality of the top few percent of printing jobs but I think the vast majority of jobs that are in the middle to lower end have indirectly suffered as a result of technology.

In the color management world that I live in, I think it’s a crime (sort of) that photographers are the ones leading the way towards adaption of an ICC-based workflow or “image handling”. Personally, I’d like to see the prepress and print shops of the world leading the way because they stand to benefit the most both for their internal production control and as a way to better serve their customers. In a way, I sort of pity photographers because their trying to push a technology to two groups that don’t seem to want to deal with it. One group, ad agencies and designers, resists because they don’t understand the technology and why they NOW need to be concerned with something they’ve never had to deal with before and the other group, prepress/print types, resists because they generally don’t have the time or resources to educate their operators given the quick turn of today’s jobs. I chalk this up mostly to ignorance and lack of vision and NOT to some sort of resounding rejection of ICC color management technology. Reminds me of my children at the dinner table when they’re SURE they won’t like something on the table when we know they’ve never even TRIED it yet. More often than not, we make them try it (we’re bad parents!) and, whatta ya know, they actually kind of like it!

Quote for today:
“You think education is expensive? Try ignorance.”

Terry
_____________________________
WyseConsul
Color Management Consulting
terry@wyseconsul.com
704.843.0858
http://www.colormanagementgroup.com
http://www.wyseconsul.com (coming soon)
___________________________________________________________________________

   Date: Thu, 11 Nov 2004 10:34:49 -0600
   From: Jim Bean
Subject: the problem is not sheetfed vs web coated

hello terry, I am one of those criminal photographers that endured dan’s classes.. from my experience.. there were more than a few  ‘real world-in the trenches” types in my groups. The association of skill sets with education hasn’t changed much over the years.. In today’s printing industry two weeks to go to any school would seem excessive/unnecessary... if you are truly interested in scanning/pre-press or whatever the topic.. you would make a reasonable effort to expand your skill set through a variety of methods..

Personally, I’d like to see the prepress and print shops of the world leading the way
because they stand to benefit the most both for their internal production
control and as a way to better serve their customers.

My personal experiences with a small cross section of the printing industry throughout the US indicates that those people in prepress couldn’t lead a horse to water.. Their own staff does understand the basics.. you are correct.. they basically drag/drop the files into a RIP and roll the dice.. the people I have found that consistently deliver in an otherwise tough environment are the front line graphics people in many of the medium sized newspapers.. those people get more than their ‘fair share’ of totally bizarre files and on a daily basis deliver... I think this ability to ‘get it done” is directly associated with ‘working in those trenches’.. graphics people/photographers/people that write many of the HOW TO books all suffer from narrow viewpoints...I have had a little experience with hiring high end graphics people.. one characteristic is:  They all know it all... why use quark when I can build it in Illustrator.. Too many people simply will not admit that they don’t know.. and that is where the problems begin.. rather than ask.. they guess.. When their files go to output.. rather than experiencing the ‘high five’ of a good file.. there is only the momentary pause that they were shot at and missed... jim bean
___________________________________________________________________________

   Date: Thu, 11 Nov 2004 15:39:08 -0500
   From: Terry Wyse
Subject: Re: the problem is not sheetfed vs web coated
 
Yo Jim,
Understand that I’m not “mad” at photographers for leading the CM charge. I’ll high-five any group that takes the ball and runs with it. Besides, I’ve done work for a few photogs (I’m a closet photographer) that, after I spent a bit of time with them talking CMYK and color management stuff, I come back later and they’ll show me something they’re doing that’s one of those “I wish I’d thought of that!” things. No, I’m just rather ticked that it’s not “my” group that’s leading the way but instead holding things back. But hey, basically the same thing happened with DTP back in the late 80s so why should I be surprised.

As to your other points: Amen.

Terry
___________________________________________________________________________

   Date: Thu, 11 Nov 2004 18:50:46 -0500
   From: Rick McCleary
Subject: Re: sheetfed vs. web coated SWOP profiles
 
In the color management world that I live in, I think it’s a crime
(sort of) that photographers are the ones leading the way towards
adaption of an ICC-based workflow or “image handling”.

Boy, is that ever true.  I am a photographer working for large corporations that spend tens of thousands on their annual report printing.  I find myself stuck in the middle - between the design folks and the printing/pre-press folks.  I’m having to teach both sides.  It isn’t easy.  And I don’t get paid for it.  The design community suffers from a combination of lack of time to learn the “new” ways and an old-school arrogance about their ability to control every aspect of the job (rooted in 20-year-old technology).  The printers suffer from the same disease.  It’s like our parents in the early 60’s (I’m dating myself...) whining about the “new math”.

Personally, I’d
like to see the prepress and print shops of the world leading the way
because they stand to benefit the most both for their internal
production control and as a way to better serve their customers.

Who wouldn’t want to work more efficiently, put out better work, and make more money?  I guarantee that the first printer in each market to invest the time to learn about techniques like Bill Aktinson’s high-density printing will make a killing.  Hell, they don’t even need to go that far.  How about employing a simple, short list of best practices regarding color management?

In a way, I sort of pity photographers because their trying to push a
technology to two groups that don’t seem to want to deal with it. One
group, ad agencies and designers, resists because they don’t understand
the technology and why they NOW need to be concerned with something
they’ve never had to deal with before and the other group,
prepress/print types, resists because they generally don’t have the
time or resources to educate their operators given the quick turn of
today’s jobs. I chalk this up mostly to ignorance and lack of vision
and NOT to some sort of resounding rejection of ICC color management
technology.

This stuff is not brain surgery.  It’s not like trying to explain quantum physics.  The problem is a simple one to solve.

Education is key.

And to learn, one must be motivated.  The motivation is money.

______________________
Rick McCleary Photography
201 Orchard Drive
Purcellville, VA  20132
v  540-338-4895
c  540-454-7180
www.rickmccleary.com
___________________________________________________________________________

   Date: Thu, 11 Nov 2004 18:05:18 EST
   From: Dan Margulis
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

Bob Johnson writes,

I work for a large pre-press organization, and we’re finding the talent pool
to be very shallow. The “professional photographer” , and I have many friends
in that trade, is certainly providing better files than you would get from
all but the finest ‘chromes and drum scanner. Armed with your classes and books, they are ahead of most pre-press operators.

The prepress industry hasn’t quite vanished, but it’s a small fraction of its former size. Since that was where most of the knowledge about how to make good, printable images resided, it’s left quite a vacuum, as well as a big business opportunity. Some photographers have stepped up to the plate, and many small groups of designers or freelancers have, but the printing and what’s left of the prepress houses by and large haven’t, so what you say is true. Unfortunate, because only five or eight years ago the idea that a photographer would know more about the process than the folks down the line would have been pretty ridiculous.

Our department averages about 15 years experience. We have a very high end
digital proofer, Gretag Spectrolino, all the toys. Two of us know how to use
them and none of the other strippers, color guys, etc have a clue, or even
care to know.

Then, had I been running the company, certain individuals would not have acquired the 15 years of experience.

We have tried to hire “pre-press people” to little avail. The rare case of a
‘designer’ who wants to “get under the hood” usually fails. The young kids
worked on the college newspaper, so they have a feel for Quark, but have no
idea what happened after they pressed the print button.

It’s a nasty situation in comparison to the case ten years back, which is why prepress operations are doing so poorly:

1) Ten years ago, it would have been very difficult for an employee to leave the company and set up his own business in competition. Now, it’s easy.

2) Ten years ago, a very good employee could maybe turn out twice as much quality work as the next guy, and so could be kept happy with a slightly higher salary and a bit of recognition. But now, a very good employee can do six times as much work as the next guy, but the company can’t afford to pay him six times as much.

3) Ten years ago, few in-house departments would have competed with the prepress house to hire from the same small field of qualified retouchers. Now, they offer better working conditions and probably better salary and benefits as well.

Dan Margulis
___________________________________________________________________________

   Date: Fri, 12 Nov 2004 00:01:19 EST
   From: Dan Margulis
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

As a color management consultant, I have the opportunity to visit quite a
number of prepress/print shops across the country and I really do think the
overall knowledge level has dropped in the past 10 years or so. I think
one of the key reasons is lack of training. Back in my production days
in the 80’s and 90’s, my employers were pretty good about sending me
off to whatever training class (usually sponsored by dealers and
vendors) I happen to have an interest in (image today being sent off
for TWO WEEKS of scanner training!). Now a days, there’s very little
training going on due in large part, I think, to the myth that the
“computer” has made learning all these applications and skills so much
easier.

I agree with the result but not the reasoning. The company was willing to invest a considerable amount in training you to use a drum scanner because they didn’t think you were going to then quit, buy a drum scanner, a couple of imagesetters and appropriate darkroom equipment, and set up business in your spare bedroom.

Dan, your impression of higher skill levels evidenced by the folks
attending your classes is sort of a self-fulfilling prophecy. OF COURSE
they have better skills because either they personally or their
employers has seen that they’re getting the training/education they
need.

That’s not it. People who attended two or five years ago were paid for by their employers just as they are today. And yet, with the same apparent background and experience level, they’re more sophisticated today.

I would also be real curious what percentage of your students are
in-the-trenches prepress folks as opposed to pure creative types,
photographers and image editing specialists.

If you are talking about somebody who works for a prepress or printing company that takes in work from the general public, less than 10%. If you mean somebody who works for a company’s in-house production and does what could loosely be called prepress, but only on the company’s own work, around a third.

I don’t want you to take this wrong, but there’s not a whole lot of
your techniques that can be used in a pure prepress production
environment. Not because they AREN’T useful or productive, but the
prepress person today is rarely allowed the opportunity to use these
advanced techniques, even if they knew how to use them. Much of this
has to do with today’s economics. These garage designers and
photographers that think they know perfectly well how to prepare a job
for a commercial printer generally aren’t willing to pay for the
prepress folks to fix these jobs. So the prepress person’s “job” today
is to simply shuffle these ill-prepared files through their production
RIPing system and hope for the best.

I don’t take it wrong because it’s true.

In the color management world that I live in, I think it’s a crime
(sort of) that photographers are the ones leading the way towards
adaption of an ICC-based workflow or “image handling”.

I see no such political correctness in their agenda, merely an acceptance of three obvious facts: 1) If the image prints badly, the client may not be able to figure out whose fault it was; 2) The chances of the printer being able to be helpful in image preparation are not good; 3) It is a lot easier for the client to hire a new photographer than a new printer.

On this list, it’s pretty clear that the photographers who have figured out how to take control of as much of the imaging process as possible are the ones who claim to be making money.

Personally, I’d like to see the prepress and print shops of the world
leading the way because they stand to benefit the most both for their internal
production control and as a way to better serve their customers.

Printers are actually rather good at making this decision for themselves. The technical quality of most printing is, IMHO, better than ever.  In just the last few years, we’ve seen a rapid transition to CTP, to a PDF workflow, and to a wide variety of new proofing methods. Thus, the printing industry has proven itself considerably more flexible and innovative than, say, advocates of ICC color management. This year, the printing industry has been spending a good deal of time and money accommodating itself to a world in which InDesign has suddenly become a huge factor.

By and large, however, printers have declined to take much interest in how images are and should be prepared, which is unfortunate. There are some printers who have a good idea of what their dot gain is, and some who know what their ink limit is or what kind of black generation works best in their workflow. However, they’re few and far between. Many books suggest that before making separations, you should consult with the printer as to what settings to use. Personally, I’d sooner consult a plumber—at least the plumber wouldn’t feel embarrassed at not knowing the answer, and wouldn’t answer with the first thing that came into his head.

My hat is off to the printers who *do* know the answers, of course.

Dan Margulis
___________________________________________________________________________

   Date: Sun, 14 Nov 2004 10:06:40 -0800
   From: Jono Moore
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

Rick wrote:
 
Boy, is that ever true. I am a photographer working for large
corporations that spend tens of thousands on their annual report
printing.  I find myself stuck in the middle - between the design folks
and the printing/pre-press folks.  I’m having to teach both sides.  It
isn’t easy.  And I don’t get paid for it.

Of course, you could supply CMYK and get on with life. : )

Playing devil’s advocate here (well, sort-of) ... You want an industry to change to make your life easier.

We’ll get there eventually, but the technology has to get there first. It isn’t easy, nor inexpensive, to implement colour management - in monetary and human terms. Currently things are far too convoluted; implementing colour management between operating systems, applications, RIPs, etc. is a major endeavour.

How about employing a simple, short list of best
practices regarding color management?

And where do we get this list? Is every printer supposed to come up with their own? Then we’re not any further ahead than we were before.

This stuff is not brain surgery.  It’s not like trying to explain
quantum physics.  The problem is a simple one to solve.

Given the money and resources to throw at the problem.
 
Education is key.

I thought K was key!

And to learn, one must be motivated.  The motivation is money.

Yes, the money...Do you have any idea what margins are like in the printing industry?  There’s not a lot of profit going around, and without profit it’s difficult to invest in new technologies. Maybe things are different elsewhere (I don’t think so), but out here on the left coast of Canada printing is cutthroat.

Don’t get me wrong - I think that colour management is something that will be good in the long run but there’s still a lot of work that has to be done to make it user friendly.

I’ve spent a few years learning about it (in my spare time) and am at the point now that I am comfortable using profiles in PS for converting images and the like, but I’m not ready to hand control of CMYK conversions over to an automatic process - I don’t think the pictures will look as good as they could.

Those are my thoughts on the problem; from the trenches in prepress.
___________________________________________________________________________

   Date: Sun, 14 Nov 2004 10:30:38 -0800
   From: Jono Moore
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

Dan wrote:
 
By and large, however, printers have declined to take much interest in how
images are and should be prepared, which is unfortunate.

I think the problem here is that they didn’t have to before. Now everyone seems to expect the printer to handle this where not too long ago it wasn’t their problem.

I think part of the problem is the way Photoshop took over years ago - everyone ended up working to whatever PS spat out.

There are some printers
who have a good idea of what their dot gain is, and some who know what their
ink limit is or what kind of black generation works best in their workflow.

This is a tough one - I’ve tried talking to pressman about these things and get blank stares. Pressman know nothing of dots, what they know and work to is Standard Ink Density.

If I can’t talk to pressman about these things how do I find them out? Currently I’m experimenting with different black generation settings and the like but it’s tough to say whether one way is better than another.

I, as a prepress tech, don’t seem to have access to any sort of empirical data to help (unless someone can point me to some).

Part of the problem, as I mentioned in my previous post, is the lack of money to throw at the problem. I work for smaller printers and the resources just aren’t there.

I suggest that if the colour management evangelists want to get wider adoption they’re going about it the wrong way. The grassroots effort isn’t working, you’ve got to start at the top and work down. Someone has to come up with a way to sell it to the owners and managers.

   Date: Thu, 09 Dec 2004 08:10:52 -0000
   From:Stephen Marsh
Subject: Re: sheetfed vs. web coated SWOP profiles

Howard writes:

Or do you do what
Dan Margulis indicates is his preferred plan of action:  convert RGB files
to a CMYK working space that consists of Adobe's USWebCoatedSWOP.icc
profile(?), edited to have 17% dot gain instead of the default 20%, changed
to Light black generation instead of the default Medium, and changed to an
85% black limit instead of the default 100%?

This has been mistaken MANY times before.

Dan uses the LEGACY CUSTOM CMYK ENGINE which IS EDITABLE.

The Adobe v2 profiles are proper third party ICC profile software package generated ICC profiles and not the custom CMYK legacy of Photoshop.

One can't edit a proper third party ICC profile in Photoshop alone (there is a plugin available but I don't wish to confuse the issue).

Binuscans PhotoRetouchPro product, does include a ICC profile editor but I am unsure if it is RGB only or also CMYK and what other features it has.

When I make this my CMYK
working space and open the womanbluecast.psd file from Dan's Professional
Photoshop CD, the cast changes from the decided blue described by one of
Dan's readers to the more magenta-tinted blue in Dan's printed version.

For me, I think it was Japanese defualts that gave the best visual preview to the printed result when viewed in less than ideal average consumer conditions.

Stephen Marsh.
___________________________________________________________________________

   Date: Thu, 9 Dec 2004 08:01:48 -0600
   From: "drhobbes"
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

Stephen,

    Thank you for answering my question.  It sounds so clearcut in Professional Photoshop, yet turns out to be so confusing in real life.

    I'll just have to research this thing and post my findings later. There's no question that I'm not alone my quest to get some fundamental understanding of what must be done to a file before sending it to any printer.

    Through this post I would like to suggest to Dan that he really should clarify this point in his next revision.  Those of us who don't have his extensive background in prepress and printing press operation are justified in believing from his book that it's a simple matter of just changing the settings in the Adobe v2 profiles.  It is misleading at best that this seems to work just as he describes when we are dealing with monitor images. Evidently it does not work as well with files sent off to the printer.

Howard Smith
___________________________________________________________________________

   Date: Thu, 09 Dec 2004 17:26:33 -0800
   From: Peter Constable
Subject: sheetfed vs. web coated SWOP profiles

Stephen writes:

Dan uses the LEGACY CUSTOM CMYK ENGINE which IS EDITABLE.

The Adobe v2 profiles are proper third party ICC profile software
package generated ICC profiles and not the custom CMYK legacy of
Photoshop.

For clarification, Adobe's v2 profiles are created using Adobe's own in-house software, not third party software.

_peter

Peter Constable
Adobe Systems, Inc.
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 01:46:23 -0000
   From:Stephen Marsh
Subject: Re: sheetfed vs. web coated SWOP profiles

    And now one final question and I'll not bother you folks for a time.
What exactly is a LEGACY CUSTOM CMYK ENGINE?  This is a new one for me at
least.

Howard, versions prior to v5 used SEPARATION TABLES.

Thus this "legacy Custom CMYK" Photoshop CMYK option in say v6, 7, CS etc.

In your profile menus in Photoshop - you should be able to select a pre-built or pre-generated, pre-installed ICC profile (ones usually created with third party profile software).

Photoshop can also make CMYK conversions using the old legacy separation table methods - which are really treated as ICC profiles in v6 or higher (but these legacy profiles are not the same as one made by a third party profile package).

Stephen Marsh.
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 02:03:35 -0000
   From:Stephen Marsh
Subject: Re: sheetfed vs. web coated SWOP profiles

Peter writes:

For clarification, Adobe's v2 profiles are created using Adobe's own
in-house software, not third party software.

Thanks Peter, I would have thought that you would have used a branded product from GretagMacbeth or someone, like "ProfileMaker" or some other such standard ICC profile generation software package (after making colorimetric machine readings to plug into  this software).

I did not know that Adobe had their own software (not that I should know this :)

I may have to pass some profiles through a profile inspector...

So when can we expect Adobe to offer it's professional colour customers (ie Photoshop users) - the ability to edit or create ICC profiles?

Having the ability to alter GCR and other factors is one of the reasons that the legacy custom CMYK is still used a lot in day to day real world production.

Stephen Marsh.
___________________________________________________________________________

   Date: Thu, 9 Dec 2004 21:10:39 -0600
   From: "drhobbes"
Subject: Adobe's CMYK profiles

    Sometimes the obvious is obscured by the confusing.  Legacy Custom CMYK Engine is a reference to Adobe's CMYK profiles in Photoshop 5.5 and earlier.  Beginning with Photoshop 6.0, the CMYK color profiles were changed to the Adobe v2 version.  This version is based on TR001 data--evidently a requirement for a SWOP-acceptable CMYK profile.  The complaint that some have made of "editing deficiencies" in Adobe's profiles, appears to be based on the inability to edit ink values and on the lack of black ink simulation and paper white simulation in soft proofing.  Dot gain (also known as TVI, or tone value increase), black ink generation,and ink limits (also known as TAC, or total area coverage) are editable.  For those trying to create a custom profile that brings profile-rendered colors into line with colorimetric measurements, the difficulty in editing ink values may be perceived as a serious deficiency.  For the majority of us who are just trying to get a printed result that meets our expectations, that may not be quite so critical.  "Ink values" refers to profiles of the four printing inks, described by LAB or xyY values, something most of us don't need to worry about unless we're consultants or specialists in custom profile creation.

    When a file is converted to CMYK, the image appearance does not change.  The numbers that describe CMYK colors in the image will print properly if the press conditions are those for which the profile is intended.  Unless these conditions are known, the image will appear satisfactory on the monitor but may not look at all the same when printed.  The CMYK numbers tell the press how you expect it to represent the image colors, but they leave the monitor image alone.  You can duplicate an RGB image, convert one of them to CMYK with a 12% dot gain, tag it with that profile, then open the second into a CMYK color space that has a 40% dot gain setting.  They will look alike on the monitor, but the printed output of one will scarcely resemble the other.  Thus the potential difference in appearance of the same image on the monitor vs. its appearance on paper if the image's conversion profile was edited to contain settings that were not suited for the actual printing conditions.  As noted before, it doesn't have to be exact if a good proof is provided and the press crew is competent.  It's remarkable how much an image can be edited at the time of printing, though this cannot be considered a safety net if an improper profile is used for the conversion.  If you have a CMYK image and want to see what would happen under different press conditions than those for which its profile was intended, you can use Assign Profile to change the monitor image appearance to show how the numbers in that existing file will print under those different press conditions.  This time you will see a change in image appearance on the monitor, but the numbers will continue to refer to the original printing conditions for which the profile was intended at the time of conversion.

    The first of the duplicate images referred to above was tagged to keep it from changing its appearance when you converted the second file with the 40% dot gain profile.  I remind you of this because there have been endless discussions about tagged vs. untagged files.  A tag is not only unnecessary for files sent to a printer but may even cause problems with a RIP, and then there are printers will strip out tags as a routine procedure. Granted that the appearance of an untagged image file will change if someone down the line opens it into a different CMYK profile than you used, the color numbers in that file will remain unchanged.  So even if the display on the stranger's monitor looks vastly different from the display on your own monitor, the printing device is going to follow the specifications in the profile in effect at the time the image was converted from LAB or RGB to CMYK.   It will take conscious effort on the part of the stranger to alter those numbers.  Keep in mind that the printing device cannot think.  If you give it the wrong specifications for the actual printing conditions, the output will not be what you expected.

    At this point, unless someone wants to correct my thinking, it would appear that use of one of Adobe's CMYK profiles for image prepress work is both workable and satisfactory provided intelligent guidelines are followed in editing the chosen profile before conversion (once a file has been converted to CMYK, you can edit the color space to your heart's content without affecting anything but the image's appearance on the monitor--just like with Assign Profile).  Needless to say, the choice should be based on expected printing conditions, not just on dot gain value.  For Dan's suggested changes in things like black generation, ink limits, etc., you can refer to Professional Photoshop.  However, Dan earllier has recommended in this forum that the black dot gain be 2 - 4% higher, cyan and magenta equal, and yellow 2% lower.  Before deciding on black generation setting, GCR, TAC, etc., those with little prepress work experience really should give Professional Photoshop a good reading.  What you want to avoid is using the default settings in the Adobe profiles.  Some users look down on the Adobe supplied profiles because they are not "editable" and the default settings do not represent real world conditions.  In fact it would appear that these profiles are well designed for the job for which they are intended, and they are editable in the sense that we can make just about all the changes that we need.  Granted that the defaults do not represent real world conditions, but hey! default settings are not intended to represent the best way to do things.  

    Recently I made a comment about the CD's womanbluecast, which one reader reported to have a blue cast that made its appearance on his monitor different from that in Dan's book.  Apparently this was due to his use of an unedited (default) version of a CMYK profile with the default black ink limit left at 100%.  Changing it to 85% brought the color back into line with the illustration.  Why?  Again, from Dan's book, black content affects final color.  Adding black increases color tone, and reducing it reduces that tone--thereby changing the appearance of the colors in the image.  The color that is present in a higher percentage than others will be the one that shows the most difference.

Howard Smith
___________________________________________________________________________

   Date: Thu, 9 Dec 2004 21:19:19 -0600
   From: "drhobbes"
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

First of all I want to thank Stephen and Peter for the additional information that turned up here right after sending in a lengthy dissertation that may help clear things a bit for newcomers.  It took me three years to reach this point, and then only because of  generous assistance from members of this forum.

Howard Smith
___________________________________________________________________________

   Date: Thu, 09 Dec 2004 19:18:12 -0700
   From: Andrew Rodney
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

On 12/9/04 7:03 PM, "Stephen Marsh"  wrote:

Thanks Peter, I would have thought that you would have used a branded
product from GretagMacbeth or someone, like "ProfileMaker" or some other
such standard ICC profile generation software package (after making
colorimetric machine readings to plug into  this software).

Thomas Knoll wrote his own software to do this. The profile is outstanding.

So when can we expect Adobe to offer it's professional colour customers (ie
Photoshop users) - the ability to edit or create ICC profiles?

When I asked him about it, he told me it has no GUI (Thomas doesn1t need one). So this is (like Camera RAW) a piece of software Thomas had to write himself to satisfy his needs. Be cool to have such a product but alas, it1s not available to mere mortals.

Andrew Rodney
http://digitaldog.net/
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 02:01:41 EST
   From: Dan Margulis
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

Howard Smith writes,

Through this post I would like to suggest to Dan that he really should
clarify this point in his next revision.

On page 270 there is a three-page section discussing exactly these points. The relation between a Custom CMYK setting and the SWOP v.2 setting are fully explained. It would be difficult to imagine how my recommendation to learn to use the Custom CMYK dialog because the v.2 settings CANNOT BE EDITED OR CHANGED could possibly be made more clear.

Those of us who don't have his  extensive background in prepress and
printing press operation are justified in believing from his book that it's a simple
matter of just changing the settings in the Adobe v2 profiles.

The Adobe profiles CANNOT BE EDITED OR CHANGED. This concept has been reinforced many times on this list, particularly by Stephen Marsh.

Similarly, nothing in the book could possibly justify anyone in thinking that the process is simple. It has three full chapters on the adjustments possible in Custom CMYK.

It is misleading at best that this seems to work just as he describes when
we are dealing with monitor images. Evidently it does not work as well with
files sent off to the printer.

I have no idea what this can be referring to. If the CMYK setting is correct, whether it was generated by Custom CMYK or not, it works for both monitor and for printer.

Dan Margulis
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 09:09:29 -0000
   From:Stephen Marsh
Subject: Re: Adobe's CMYK profiles

Hi Howard, I will add comments where necessary - where I have trimmed text away it needed no comment, or I could not comment.

I understand that you are attempting to get a handle on a complex subject that has confused more than one member of this list in the past. Stick with it, I think you are close!

Beginning with Photoshop 6.0, the CMYK color profiles were changed
to the Adobe v2 version.

v4 or earlier only used the separation table method. It could load, and save out separation tables.

Things changed in v5.x - it had the ability to load old style Adobe separation table engine or to use ICC profiles (there was also a third option to load separation tables from other sources outside of the legacy Adobe CMYK engine). It could no longer save sep tables, these became an ICC profile when saved. Dot gain also changed in v5. It was still called 20% but it did not have the same LAB value reading in the midtone as in v4. This is why there are two prebuilt ICC profiles called Photoshop 4 Default CMYK or Photoshop 5 Default CMYK or whatever they are named (they should all be based off SWOP Coated 20% dot-gain, Med. GCR, 100%K etc). From v5 to now, 20% still means what it did in v5.

Then v6 really changed things, everything became ICC profile based and the 'old, legacy' CMYK separation method became 'Custom CMYK'.

This presumably uses the same old methods behind the scenes but the data appears as a ICC profile since that is all that later versions deal with.

The complaint that some have made of "editing deficiencies" in
Adobe's profiles, appears to be based on the inability to edit ink
values and on the lack of black ink simulation and paper white
simulation in soft proofing.  Dot gain (also known as TVI, or tone
value increase), black ink generation,and ink limits (also known as
TAC, or total area coverage) are editable.

The problem is ICC profile editing deficiencies in Photoshop, not the profile itself. Or another way to word this is...Adobe have only given its CMYK power users ONE fixed set of conditions to separate into - without offering any other alternate profiles. Adobe also do not offer an ICC profile for newsprint, where as their old Custom CMYK engine did.

For me it is this simple: Dear Adobe, please give the print industry easy, affordable access to profile editing abilites in Photoshop or if you do not wish to turn an image editor into a profile editor, then at least offer more profiles. But who cares about the print market, they stopped being a serious customer back in v4 or v5. Digital photography and other areas is where the future and money is.

Some profile makers do make a 'suite' of profiles for similar conditions. The aim-point may be TR001, but they may produce UCR, various GCR ratios and even ink weights like 280/300/330 etc. This is much better than only one profile. Chromix did this for TR001.

Even though some users get in trouble and may not need or wish to know about prepress variables - some users of Photoshop do require these things. I would like to think that in this day and age we have progressed from the mindset of "there is only one flavour of CMYK".

In the old days, many users just hit the mode/cmyk button and that was as much thought into prepress that was given. But some users knew a little more, and would alter variables to suit their conditions. These users who do need to have flexible separation options made and still make good use of the Custom CMYK controls offered by Adobe. Many scorn this old system (often those who have a professional interest in competing current ICC technology). Many embrace this system (often those who need results that are image dependent and who understand that a single conversion method is far from ideal).

When a file is converted to CMYK, the image appearance does not change.

In-gamut colours will not change. I knew what you meant to say though.

Out-of-gamut colours - get clipped if using Custom CMYK or Relative Colorimetric rendering with a proper ICC profile. O-o-G colours will be compressed if using Perceptual rendering.

Also, if the profile creation software has a proprietary 'secret sauce' method built into the Perceptual intent - one can get a significant contrast boost. The Adobe v2 profiles do not have this (this good in my view). If the image needs to be lighter, the dot gain should be addressed, not a contrst curve applied which may wipe out highlights.

 At this point, unless someone wants to correct my thinking, it
would appear that use of one of Adobe's CMYK profiles for image
prepress work is both workable and satisfactory <<

I will cut you off there!

Yes, the Adobe profiles are good/satisfactory - if they actually describe the final conditions. Same as for any profile.

The situation and question that you are really asking here is:

If I have 'an average, nice but unimpressive RGB image' - that looks good on the monitor and in test press simulation inkjet prints...what is the best conversion for an unknown condition - but we presume that it is safe to assume SWOP type conditions (as opposed to say European or Japanese)? Does one convert using the v2 SWOP profile, or the Custom CMYK engine?

That is a harder question to answer.

I will give you this hint though, in the real world when results matter over theory - theory often fails.
 
Some users look down on the Adobe supplied profiles because they
are not "editable" and the default settings do not represent real
world conditions.\

I think that bit is mixed up. The Adobe v2 ICC profiles do represent real world conditions. It is just a matter of your real world conditions matching those described by these profiles.

If the prebuilt profile does not match the condition, then one can simply use it and hope for the best (not good).

Or one can find a profile that does match the condition, or simply use Custom CMYK to make a separation that does suit the condition better than the proper ICC profile. In theory the proper ICC profile is better, but in practise often not as good as Custom CMYK.

In a nutshell, I do use the Adobe v2 profile (and others) - but 90% of the time I would CHOOSE to use the Custom CMYK setting instead.

Q: Would I use a proper ICC profile instead of Custom CMYK if it delivered the dot gain, black plate weight, black ink limit and total ink limit that is required?

A: Yes.

Q: Is it easy for the average person to possess such a profile or the tools, knowledge and skill (not to mention time/budget) to generate such a profile?

A: No.

Is it any wonder that Custom CMYK is such a valuable tool to many users, considering the nature of the print industry?

The last question is the one which Dan has obliquely commented on about machine generated profiles often misrepresenting pastel colours without some human editing.

This is easy to prove/disprove - but the current exercise/image of the blue woman is not the best way to do this. Even profile vendors make a distinction between a raw profile and one which has been 'tuned'.

I know how/why ICC works - but it often sounds like smoke and mirrors.

Stephen Marsh.
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 08:07:03 -0700
   From: Andrew Rodney
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

On 12/10/04 12:01 AM, "Dan Margulis"  wrote:

The Adobe profiles CANNOT BE EDITED OR CHANGED. This concept has been
reinforced many times on this list, particularly by Stephen Marsh.

To clarify, they CAN be modified with a 3rd party profile editor. They cannot be modified in Photoshop it self.

Andrew Rodney
http://digitaldog.net/
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 10:30:26 EST
   From: Dan Margulis
Subject: Re: Adobe's CMYK profiles

Howard Smith writes,

The complaint that some have made of "editing deficiencies" in Adobe's
profiles, appears to be based on the inability to edit ink values and on the lack
of black ink simulation and paper white simulation in soft proofing.  Dot gain
(also known as TVI, or tone value increase), black ink generation,and ink
limits (also known as TAC, or total area coverage) are editable.

No, they are not. The v.2 profiles are COMPLETELY UNEDITABLE within Photoshop. You cannot change dot gain. You cannot change black ink generation. You cannot change total ink limit, or black ink limit, or UCA. You cannot change ANYTHING AT ALL. Not one jot, not one tittle, not no way, not no how. You have to take them exactly as they are, or leave them and use Custom CMYK. The minute you change to Custom CMYK, you are trashing the v.2 profile in favor of an entirely different profile that is being built from scratch using defaults that date from Photoshop 5.

Therefore, your very long post is based on a single fundamental error--that when you have the v.2 profile loaded, and switch to Custom CMYK, you are somehow magically editing that profile. Again: when you switch to Custom CMYK, you are not editing the v.2 profile, because that v.2 profile CANNOT BE EDITED OR CHANGED IN ANY WAY. You are throwing it away and beginning from scratch.

Consequently, you have reached a completely wrong conclusion in something like the following.

Recently I made a comment about the CD's womanbluecast, which one reader
reported to have a blue cast that made its appearance on his monitor different
from that in Dan's book.  Apparently this was due to his use of an unedited
(default) version of a CMYK profile with the default black ink limit left at
100%.  Changing it to 85% brought the color back into line with the illustration.  
Why?

Not for the reasons you stated. The original v.2 profile does NOT have a black ink limit of 100%. Only the Custom CMYK engine has that default. The reason you saw 100%  was that you switched into Custom CMYK, and the minute you did that, the v.2 profile was toast, history, finished, forgotten about, of no account, and no longer a factor. The monitor rendition looked too blue at first because the v.2 profile does not represent light blues accurately. It looked better when you went to Custom CMYK because you were no longer using the v.2 profile, but rather an entirely new, made-from-scratch profile using an engine that doesn't have this particular problem, although it has others.

Dan Margulis
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 08:49:36 -0700
   From: Andrew Rodney
Subject: Re: Adobe's CMYK profiles

On 12/10/04 8:30 AM, "Dan Margulis"  wrote:

Therefore, your very long post is based on a single fundamental error--that
when you have the v.2 profile loaded, and switch to Custom CMYK, you are
somehow magically editing that profile.

Hold the presses, I actually totally agree with Dan on this point. This is totally DUMB that Adobe doesn1t inform you that the switch has nothing to do with the U.S. Web Coated (SWOP) v2 profile you were just working with. You1re supposed to notice that v2 is missing? They should make this much clearer.

Andrew Rodney
http://digitaldog.net/
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 13:16:29 -0600
   From: "drhobbes"
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

Dan Margulis gave the following response to my comments on CMYK profiles:

On page 270 there is a three-page section discussing exactly these points.
The relation between a Custom CMYK setting and the SWOP v.2 setting are fully
explained. It would be difficult to imagine how my recommendation to learn to
use the Custom CMYK dialog because the v.2 settings CANNOT BE EDITED OR
CHANGED could possibly be made more clear.
The Adobe profiles CANNOT BE EDITED OR CHANGED. This concept has been
reinforced many times on this list, particularly by Stephen Marsh.

Similarly, nothing in the book could possibly justify anyone in thinking that
the process is simple. It has three full chapters on the adjustments possible
in Custom CMYK.

 If the CMYK setting is correct, whether it was generated by Custom CMYK or not, it works for both monitor and for printer.

Dan,

    First of all, thank you for taking the time to respond to my post.

    On page 270 of Professional Photoshop you state very clearly that the CMYK settings can be changed if one uses the traditional Photoshop engine (Custom CMYK), then you just as clearly state that "(the) profile...supplied with Photoshop...can't be edited within Photoshop."  Isn't the act of changing dot gain, ink limits etc. considered editing?  Then you state that "few knowledgeable users employ (U.S., Web Coated SWOP v2)...".  Dan, what is a custom CMYK if it isn't one made by editing the settings in one of Photoshop's CMYK profiles?  I agree that both you and Stephen Marsh have clearly stated that the v.2 settings cannot be edited or changed.  O.K., but where does the Custom CMYK come from if not from a modification of one of Adobe's profiles?  Perhaps it would make this whole thing a little more clear if you or Stephen Marsh would be willing to post an example of the specific steps you take in creating a Custom CMYK profile that you will use for converting something you intend to send to press.

    I totally agree that nothing in your book could possibly justify anyone thinking that any of this is simple.  I didn't wear out three copies for the fun of it.

   As much as I respect your knowledge and your writing ability, I cannot agree that your explanation of the creation and use of custom CMYK profiles cannot be made more clear.  You have a lifetime of experience in this field, but some of us are a little short of that, which could explain why you feel it's crystal clear and I do not.  For whatever it's worth, several other well known Photoshop writers are convinced that they also have explained these things very clearly.  They haven't.  I am not a stupid person, but after three years of pretty serious study I still do not know how to create a custom CMYK profile without resorting to modification of an Adobe supplied profile or calibration instrumentation.  Because of your adamant stand against calibrationism, and my respect for your judgment, this latter is not a route I am interested in taking.

    We both have short tempers, Dan, so I can understand why you might not want to pursue this.  Just be assured that my intent here is to get an answer to my questions, not to challenge you or any of the others who have given their time so generously.   We all have similar goals, but some of us are slower than others.

Howard Smith
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 13:35:09 -0600
   From: "drhobbes"
Subject: Re: Adobe's CMYK profiles

Well, Dan, I've just got to start reading current e-mail before sending in my replies to earlier messages.  You've pretty well answered my last post before you received it.

I'll have to study your current message about Custom CMYK just like I've done with Professional Photoshop--over and over until I've wrung everything out of it.

As I noted in the post that you haven't seen yet, this looks a whole lot more clear and understandable to you than it does to some of us.  This is the first time anyone has ever explained that Custom CMYK has no relationship to the Adobe profile that had been active at the time the Custom choice was made.  My belief was that choosing "Custom" meant that we were being given an opportunity to make changes to the v.2 profile.  That's why it was totally puzzling to keep getting these comments about the v.2 profiles being uneditable.  I thought that switching to Custom was simply giving me the opportunity to change the v.2 settings, not taking me into new territory to build a profile from scratch.  Perhaps New And Different would be preferable to Custom.

Now I can just hope that you read your current e-mails before responding to my unfortunate last post--unless of course you want to add any more information to what you just posted.

Thank you for not giving up on me.

Howard Smith
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 14:11:29 -0600
   From: "drhobbes"
Subject: Re: Re: Adobe's CMYK profiles

Stephen,

    If you have written any books or articles on Photoshop or on printing, please post the information so I can order them.  Your answers to my questions about Adobe Custom and v.2 profiles were both complete and understandable.

    Thank you for going to so much trouble to answer my posts.  I hope the time will come when I'll have an opportunity to reciprocate for all the generous responses to my sometimes misguided questions.

Howard Smith
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 13:43:39 -0600
   From: "drhobbes"
Subject: Re: Adobe's CMYK profiles
 
On 12/10/04 8:30 AM, "Dan Margulis"  wrote:

Therefore, your very long post is based on a single fundamental error--that
when you have the v.2 profile loaded, and switch to Custom CMYK, you are
somehow magically editing that profile.

Then on 12/10/04, Andrew Rodney added:

Hold the presses, I actually totally agree with Dan on this point. This is
totally DUMB that Adobe doesn1t inform you that the switch has nothing to do
with the U.S. Web Coated (SWOP) v2 profile you were just working with.
You1re supposed to notice that v2 is missing? They should make this much
clearer.

Oh, I totally agree!  Neither you nor Dan can appreciate how much it means to me to get this cleared up at last.  Three long years wasted!  Well, on the other hand I did learn some other things...

I appreciate you folks not giving up on me, even when it must have been trying your patience to keep telling me over and over that the v.2 profiles couldn't be edited in Photoshop.  It's so nice to have this make sense.

Howard Smith
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 17:28:06 -0500
   From: Terry Wyse
Subject: Re: Re: sheetfed vs. web coated SWOP profiles
 
On Dec 10, 2004, at 2:16 PM, drhobbes wrote:

 Isn't the act of changing dot gain, ink limits etc. considered editing?

The POINT is you're NOT editing the actual profile you currently have selected as your CMYK working space. Plug in ANY profile you want as your working space profile and then pull down "Custom CMYK". It's identical in each case which should be your first clue that the default settings of Custom CMYK have nothing whatsoever to do with the profile you have selected as your working space.

Dan, what is a custom CMYK if it isn't one made by editing the settings in one of
Photoshop's CMYK profiles?

It's not. See above. In fact, density and dot gain values are something rather foreign to an ICC profile. You can certainly INFER from the colorimetric values that it might have "x" dot gain or "y" density but it's nothing that's really inherent to the profile. To put it another way, there are no tags or data in a profile that explicitly tell you what the press/print conditions were.

I agree that both you and Stephen Marsh have
clearly stated that the v.2 settings cannot be edited or changed.   O.K., but
where does the Custom CMYK come from if not from a modification of one
of Adobe's profiles?

With CustomCMYK, you're simply starting with a basic ink set (SWOP, Toyo, whatever), plugging in some dot gain values (along with the usual total ink limit, UCR/GCR, etc.) and it attempts to build a simplistic profile from the supplied data. I wouldn't say it's "bad" in any way, as long as it meets your needs, but I have to believe that it's starting with very simple assumptions about press behavior as compared to something like SWOPv2 that's based on REAL press behavior (TR001). Just my opinion though.

Regards,
Terry
_____________________________
WyseConsul
Color Management Consulting
704.843.0858
http://www.colormanagementgroup.com
___________________________________________________________________________

   Date: Sat, 11 Dec 2004 00:08:32 -0000
   From:Stephen Marsh
Subject: Re: sheetfed vs. web coated SWOP profiles

I wouldn't say it's "bad" in any way,
as long as it meets your needs, but I have to believe that it's
starting with very simple assumptions about press behavior as
compared to something like SWOPv2 that's based on REAL press behavior
(TR001).

Agreed Terry! Custom CMYK is not bad - if you get the press results you require who could care if the technology does not use a wide base of colours for it's calculations or offer softproofing etc.

Just as the Adobe v2 SWOP profile is not bad, if it gives you the press results that you were after. And if one had more flexible choices in ink weights and GCR methods with a TR001 aimpoint (or other standards), then custom CMYK would be required even less than it is today.

Stephen Marsh.
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 19:00:44 -0600
   From: "drhobbes"
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

Terry,

    Thank you for your detailed response to my questions about Photoshop profiles.  This looks very simple now, but it wasn't so easy to grasp when I was on the outside looking in and with no previous experience with the strange things being witnessed.  I understand Dan's frustration with my being unable to understand what he sincerely felt was an explanation that should have been immediately clear to almost anyone.  He just didn't realize that people like me were out there.  The fact that the Photoshop v.2 profiles go out the window when you check Custom is just a little thing, but then so is a flu virus.

     There's just one question left.  If you have the time to answer it, your help certainly will be appreciated this time as well.

    What kind of profile do the majority of prepress people use for conversion to CMYK?  Do they most commonly use the legacy Photoshop Custom CMYK, off-the shelf third-party profiles, their own machine-generated custom profiles, profiling data or profiles provided by the printing firm or by a service bureau, or do they more commonly send in files converted with and/or tagged with any manner of strange profiles that must be worked around by whoever makes the printing plates?  From what has been said earlier, Dan Margulis uses the legacy Photoshop Custom CMYK profile with settings he recommends in Professional Photoshop.  I wonder if I misunderstood this as well?

Howard Smith
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 19:11:18 -0700
   From: Ron Kelly
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

Mr. D.R. Hobbes:

Hello, friend. One who is brave to ask questions of the high and mightly wizards; many more of us in the shadows are listening.

By the way: how is Frodo doing, and all the other Hobbes-it's? Please pass along my regards.

FWIW: I have never had much luck, despite all the advice to, using SWOP v.2 or any of the canned profiles. They just don't come close enough, and as I realize now I can't change 'em so if they're off the mark, you're cooked, eh?

I use my own fine-tuned legacy Photoshop Custom CMYK profiles, and it's working fine, except that I always wonder if I'm living in a little backwater. So I try to keep a weather eye out.

I always find myself wondering, "Is this the best way to do things? Could they be better?" but I never know the answer to that question, and I suspect I'm not alone.

Cheers,
Ron Kelly
___________________________________________________________________________

   Date: Sat, 11 Dec 2004 00:09:33 -0600
   From: "drhobbes"
Subject: Apology

Dan,

    Please accept my apologies.  You were right and I was very much in error.  Now that I can see the whole picture the description you provided on pages 270 and 271 is both clear and understandable.  Since you can expect to encounter some readers who may be even more dense than me, you might consider putting in a specific example (for example, how you prepared a profile for the photograph of the cowboy that was included in your book)..  Sometimes those of us who can't grasp simple explanations may understand better if we are given either redundant explanations or better yet a simple, real-life example.

    So you can understand why anyone would be confused by a perfectly clear explanation, let's put it this way, Dan.  Suppose you were attending a lecture on medical immunology and the speaker was describing symptoms of a disease that affects the skin, the nervous system, and the immune system.  Let's say that in the course of that lecture he states clearly and unequivocally that the interrelationship of the symptoms is due to their common origin in the embryo.  Now he and other knowledgeable biologists would know perfectly well what he was talking about but you and most of the physicians might feel left out. It's not a question of intelligence, but a question of familiarity with the subject. Now suppose he went on to explain that the embryo is formed from basically three different layers of tissue, and that the outer layer (the ectoderm) is the origin of the outer skin, the mouth, the nervous system, and the immune system.  You would understand the skin part, but what about the nervous system and the immune system?   Being quick enough to recognize that not everyone in the audience had the background to follow his lecture, he might go on to explain for their benefit that the skin along the midline of the back forms a deep groove that closes over to form the tube that becomes the rudiment of the spinal cord from which the nerves will branch and that some of the nervous tissue will differentiate into components of the immune system.  Once he offered these details that are familiar to biologists but not to prepress experts, you (and most of the biologists in the audience) would understand what he was talking about when he resumed his description of disease symptoms that involve all three systems.  Even without a degree in biology or medicine you would be able to understand how these symptoms were related to the common origin of the affected tissues.  Actually not all of this is elementary to a many biologists or to most physicians, but that's beside the point.  

    Anyway, I'm sorry to have caused you such frustration.  Thank you for gritting your teeth and going ahead and explaining it in detail that even I could understand.

Howard Smith
___________________________________________________________________________

   Date: Sat, 11 Dec 2004 10:40:04 +0000
   From: Shangara Singh
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

It was 10/12/04 10:28 pm, when Terry Wyse wrote:
 
The POINT is you're NOT editing the actual profile you currently have
selected as your CMYK working space. Plug in ANY profile you want as
your working space profile and then pull down "Custom CMYK". It's
identical in each case which should be your first clue that the default
settings of Custom CMYK have nothing whatsoever to do with the profile
you have selected as your working space.

Terry

Not if you select Photoshop 4 Default CMYK. <g> OK, it's a small change but a change nonetheless.

What's more, because Custom RGB uses the settings for the current RGB profile, it's easy to assume that Custom CMYK also behaves similarly. Of course, once you know, you know. However, before you know, there's ample elbow room to make wrong assumptions. Consistency is the key to reducing assumptions, IMHO, and sadly that's lacking in the interface.

In some areas the user interface principles have been so heavily violated that the user has no choice but to unlearn what they have learned and learn an illogical behaviour - and unlearn it again when working in other apps that are more logically laid out and which follow user interface principles.

Shangara.
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 20:34:16 -0900
   From: KmKm/Crystal Images
Subject: Custom CMYK
 
So, please Dan, what do you mean "from scratch"?  The Custom CMYK profile must be based on something?  If not v2 or whatever it looked like it was, then what? And what are those "other problems"?  I traveled down this path with my project and made similar wrong assumptions, apparently--though output on my own printer has been fairly agreeable with my monitor and my intent.  Would love to understand more.  Kathleen
--
Kathleen M.K. Menke
Crystal Images
http://kcd.com/ci
___________________________________________________________________________

   Date: Fri, 10 Dec 2004 21:07:16 -0600
   From: "drhobbes"
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

On December 10, 2004, Ron Kelly wrote:

FWIW: I have never had much luck, despite all the advice to, using SWOP
v.2 or any of the canned profiles. They just don't come close enough,
and as I realize now I can't change 'em so if they're off the mark,
you're cooked, eh?

I use my own fine-tuned legacy Photoshop Custom CMYK profiles, and it's
working fine, except that I always wonder if I'm living in a little
backwater. So I try to keep a weather eye out.

I always find myself wondering, "Is this the best way to do things?
Could they be better?" but I never know the answer to that question,
and I suspect I'm not alone.

    Thanks for asking, Ron.  Frodo is doing just fine, counting his potential Christmas gifts already!  (Jaken's feelings are hurt that you didn't ask about him also...).

    You asked "Is this the best way to do things".  If you can't find a better way, it's the best way.  The worst way is to wait for something better to come along.  Experimentation is the yeast of life.  The more you put into it, the bigger the rise you get out of it.

Howard Smith
___________________________________________________________________________

   Date: Sat, 11 Dec 2004 08:56:52 -0800
   From: Eric Bullock
Subject: Re: sheetfed vs. web coated SWOP profiles

What kind of profile do the majority of prepress people use for
conversion to CMYK?  

The key part of that question is "the majority". I would venture to assert that the bulk of CMYK separations are done with the Adobe SWOPv2 profile...which has become sort of a "when in doubt" default for many people out there. Short of more specific information it isn't an awful choice.

-eb-
___________________________________________________________________________

   Date: Sat, 11 Dec 2004 11:56:46 EST
   From: Dan Margulis
Subject: Re: Custom CMYK
 
Kathleen Menke writes,

So, please Dan, what do you mean "from scratch"?

I mean that once the CMYK setting is changed from "SWOP v.2" to "Custom CMYK", every last trace, particle, iota, and jot of the SWOP v.2 profile is absolutely, positively, unconditionally, and irrevocably erased, destroyed, eliminated, extirpated, taken out of, removed, and missing from what comes next; and that the process of profile editing begins afresh with something else.

The Custom CMYK profile must be based on something?  If not v2 or whatever
it looked like it was, then what?

The traditional Photoshop CMYK default profile, last updated in Photoshop 5.

And what are those "other problems"?

Underestimation of black dot gain, overestimation of cyan dot gain, a 100% black ink limit, a non-skeleton black generation, and previews that are misleading in the shadows because the engine assumes that 100k is absolutely black.

Many threads related to these topics are posted at
http: //www.ledet.com/margulis/ACT_postings/SeparationIssues/Separation.htm

Dan Margulis
___________________________________________________________________________

   Date: Sat, 11 Dec 2004 14:38:56 -0600
   From: "john opitz"
Subject: Re: Adobe's CMYK profiles

 Hello Stephen,

How are you ? It's been a while since we conversed. This post I'm making is not to be against color management. Color management is useful.  And since the current thread of the Custom cmyk is a hot topic, some people not knowing, changing to custom cmyk is creating a new profile and its (not) modifing an existing profile, and I know some...... that think this as well. And their in publishing. Well, I'm going to add to what some might consider color management as magic, to remove the smoke (and mirrors) about it. This is not my take on it, but I agree with it. It's from the book "Real World Color Management (1st edition)". "Where the Models fail". Page 45. I will not quote the book word for word, anyway (that God).

The CIE models is designed to predict the degree..... of two (solid) color swatches seen at a specific distance, background, lighting that would appear to match. We (humans) see color in context.  Color Managements' goal is not to get a colorimetric match, just a pleasing image.
 A "well trained" (key-words here) "eye" beats a "colorimeter" every time ( Ooo. other key-words here) for evaluating final color.

  Now this sounds like I'm against color management, Stephen. I'm just agreeing with the book...... Now I understand it so clearly now, when the Man says: "If you want quality.....customize", using the eyeball method. Btw, this might not be the exact quote (Professional Photoshop, which edition...unknown). But something close. Pretty scary that I can quote authors. You think I read too much? More so, retain too much? In all, with the above kept in mind one who uses color management (machine "colorimeter" generated profiles) will not be disappointed. No more chasing rainbows and setting traps for unicorns, for me.

  John
___________________________________________________________________________

   Date: Sat, 11 Dec 2004 14:19:23 -0700
   From: Andrew Rodney
Subject: Re: Re: Adobe's CMYK profiles

On 12/11/04 1:38 PM, "john opitz"  wrote:

 A "well trained" (key-words here) "eye" beats a "colorimeter" every time
( Ooo. other key-words here) for evaluating final color.

True but let1s not misunderstand Bruce1s point here. He1s saying that if the goal is to evaluate the quality of a complex color image, CIE colorimetry and instruments are not appropriate. However, these instruments and the work of the CIE is key to actually producing ICC device profiles. So we have a system by which we need to view complex color images and make evaluations and we have systems in which we need to fingerprint a device by viewing solid colors. One system isn1t right for the other goal and vise versa. This is why eye ball calibration is not very effective while instrument analysis of image quality isn1t very effective. Boils down to the right tool for the right job.

"If you want quality.....customize", using the eyeball method.

I1d say use the eyeball for estimating quality. For producing specific repeatable color analysis based upon the measurement of a solid color, an instrument is vastly superior.

Andrew Rodney
http://digitaldog.net/
___________________________________________________________________________

   Date: Sat, 11 Dec 2004 22:29:41 -0000
   From:Stephen Marsh
Subject: CMYK Colour Settings Test

With the recent posts on 'what happens when one changes the dialog to Custom CMYK from say the Adobe SWOP v2 pre-built ICC profile' perhaps a quick little test will help nail this issue in folks minds.

* Create a new file in greyscale mode at some tint value (50% K). Perhaps place a colour sampler so you can note values in the next steps

* Convert to profile to the SWOP v2 profile

* Save a history snapshot (named SWOP v2 if you like)

* Undo the conversion

* Convert to profile to the Custom CMYK without changing any other settings or values (this is where it has been mistakenly presumed that Custom CMYK references the v2 profile).

* Save a snapshot (named Custom CMYK if you like)

* Compare the colour values between the snapshots.

Also repeat with real images, perhaps saving a snapshot of each conversion methods black plate and then compare etc. Also note in the staus bar section of the Photoshop main window that one can have this to display document ICC profile (or file size etc).

Another little test:

Make a new CMYK file with solid patches of C, M, Y, K and other tones and colours etc. Assign the SWOP v2 profile to this file and read/note LAB values of the CMYK numbers. Then assign a Custom CMYK generated profile (such as Photoshop 5 Default CMYK.icc) and note the LAB values. It should now be obvious that the default Custom CMYK does not have the same LAB response as does TR001 (SWOP v2).

If you require more accuracy in your LAB readings, change your rendering intent in Color Settings to Absolute Colorimetric as you make these measurements - BUT DONT FORGET TO CHANGE IT BACK TO RELCOL or PERCEPTUAL once you are done with the evaluation (this is not really critical for this simple test mentioned above, but is required if you do need more accurate LAB measurements when doing colorimetric tasks).

Stephen Marsh.
___________________________________________________________________________

   Date: Sun, 12 Dec 2004 09:04:40 -0500
   From: Terry Wyse
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

I debated even attempting to come up with an answer to this question but I'll give it a shot:

I would say it's mostly between "USWebCoated(SWOP)v2" or "USSheetfedCoated". Most would probably default to SWOPv2 even if they are a commercial sheetfed shop (which is too bad). But I suppose there are just as many that upgraded from the Photoshop 4/5 days that simply set their color preferences to either Photoshop 4/5 defaults and left it there. Again, too bad.

Personally, I hope there starts a movement towards standard ICC profiles that model real press behavior and that these would be available in Photoshop. I can't see how it would be a bad thing to settle on a half dozen or so profiles that would cover SWOP/web publication work (TR001), commercial sheetfed (GRACoL TR004) and possibly Newspaper (IFRA). You could also throw flexo/packaging (FIRST) in their as well. I'm speaking mostly of the US printing industry. In Europe, you've got the ECI working on this and their collection of ISO characterization data.

I personally use a slightly tweaked version of USSheetfedCoated, tweaked more for the realities of linear CTP than for anything else (USheetfedCoated assumes too much gain, more like a film-to-plate workflow). I prefer the gray balance behavior and solid ink hues of this profile over SWOPv2. I work mostly with either commercial sheetfed or web printers so my bias is definitely towards commercial printing regardless of how the paper is transported through the press. :-) Given that, in my press optimization work I'm always pushing the press towards GRACoL guidelines as far as density and dot gain.

I could go on but I have to be heading to Nashville today to get some work done!

Terry
_____________________________
WyseConsul
Color Management Consulting
704.843.0858
http://www.colormanagementgroup.com
http://www.wyseconsul.com (coming soon)
___________________________________________________________________________

   Date: Sun, 12 Dec 2004 14:20:19 -0600
   From: "drhobbes"
Subject: Re: Re: sheetfed vs. web coated SWOP profiles
 
On December 12, 2004, Terry Wyse posted a complete and most informative response to my questions about which profile(s) are most popular for commercial printing use.  As the entire post immediately precedes this acknowledgement of his help, the message has been deleted for brevity.

Terry,

    Thank you for generously taking the time and effort to clarify this. It may not be all that big a deal to you, but it's more than valuable to me--and I'm sure to many of those who stand on the sidelines, watching and hoping that someone will ask all these wild questions. Your post has been saved for further study.

    This information may be common knowledge to you and to your peers who have years of education and experience in this field.  It's very fortunate for the rest of us that you understand  our desperation and were willing to take the time to further our education.  This is the kind of applied knowledge that is sorely lacking in almost all of the books on the subject, at least in my own extensive collection.

    For what it's worth, if anyone else who browses this forum could answer yet another of my unending questions, it will undoubtedly be helpful to many of ouir number.  Has it been your experience that sometimes you get poor printed output with profiles that are known to be right for the job?  And, on the other hand, have you found that some printers can produce good results even when the profiles used are not the ideal choice for the job? Along these same lines if anyone would be willing to provide either horror stories or examples of printers or others who went to extraordinary lengths to give you a good result, that could be as instructive as any kind of technical advice.  Sometimes it's easier to learn from examples than from detailed descriptions.  As an example, my first commercial printing job was color separated by a true expert, one of the best.  It was unfortunate for me that the printer, recommended to me by a co-worker's brother, had never done a color printing job in his life.  It was my fault for not asking questions before contracting for the job.  What I'm looking for here is some idea of just how much leeway we have when assigning profiles, preparing proofs and choosing printing firms (if there is any choice).  Is this an exact science or can we make little errors along the way and still survive?

Howard Smith
___________________________________________________________________________

   Date: Sun, 12 Dec 2004 16:50:34 -0500
   From: "jc castronovo"
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

This leads me to ask the question of how much control a pressman has on the job. If our files aren't exactly what they should be, when is the job hopelessly lost? I know that's a lot to ask for in such a forum as this, but I know that a lot can be done on press. How do we know when a pressman is just being lazy and blaming the files? What should we expect of him?
___________________________________________________________________________

   Date: Sun, 12 Dec 2004 16:17:45 EST
   From: Dan Margulis
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

Terry writes,

Personally, I hope there starts a movement towards standard ICC
profiles that model real press behavior and that these would be
available in Photoshop. I can't see how it would be a bad thing to
settle on a half dozen or so profiles that would cover SWOP/web
publication work (TR001), commercial sheetfed (GRACoL TR004) and
possibly Newspaper (IFRA). You could also throw flexo/packaging (FIRST)
in their as well.

The philosophy is an excellent one, but unfortunately, until Photoshop allows us to edit the profiles, it's a waste of breath.

It's ridiculous that we should be forced to use something as dated as Custom CMYK, but not as ridiculous as being repeatedly asked to change to something that doesn't meet our minimum requirements. If the experience of the nearly seven years since Photoshop 5 should have taught Adobe anything, it's that the ability to tweak profiles is an absolute prerequisite for the high-end CMYK community.  Until it's there, your otherwise commendable idea is a nonstarter.

Dan Margulis
___________________________________________________________________________

   Date: Sun, 12 Dec 2004 17:28:44 EST
   From: Dan Margulis
Subject: Re: Apology

Howard Smith writes,

Please accept my apologies.  You were right and I was very much in error.  
Now that I can see the whole picture the description you provided on pages 270
and 271 is both clear and understandable.  Since you can expect to encounter
some readers who may be even more dense than me, you might consider putting in
a specific example

First of all, not to worry about being wrong. Color is unique in that nearly everybody has certain blind spots, where they talk themselves out of seeing something that's obvious to everybody else. It happens to me, too; at least twice this year I've discovered I completely misunderstood some concept, and although I was getting to the right result, I was getting there for all the wrong reasons, and I can't imagine how I could possibly have been so stupid.

With respect to the book, because of this effect, that certain people can't get a handle on certain concepts, beginning with the last edition there are boxes in most chapters called "Stumbling Blocks": concepts that I think are clearly explained elsewhere but, for some reason, certain readers aren't grasping. For the next edition, I'll put this one in.

Dan Margulis
___________________________________________________________________________

   Date: Sun, 12 Dec 2004 18:06:03 -0900
   From: KmKm/Crystal Images
Subject: Custom CMYK

Thank you, Dan, for these answers and yes, I did go through the archives.  Still have a few questions, if you will indulge me.

The traditional Photoshop CMYK default profile, last updated in Photoshop 5.

Is it possible to state which of the current profiles, the Custom CMYK profile most closely relates to?

Underestimation of black dot gain, overestimation of cyan dot gain, a 100%
black ink limit, a non-skeleton black generation, and previews that are
misleading in the shadows because the engine assumes that 100k is
absolutely black.

"Underestimation of black dot gain"--meaning printed result will be more dark in output than anticipated by what is viewed on the monitor?

And if so, can this be compensated for (somewhat) by "lighten/contrast" and evaluating one's own proof off one's own inkjet?  (I assume this somewhat depends on which inkjet and which RIP).

The 100% black ink limit can be edited, correct?  As well as the UCR/GCR settings?  This to me seems the value of the custom CMYK profile?

But then there was that niggling comment about editing this profile only affects screen view and not output?  If so, then what's the point of editing?  I guess that's why I chose to embed profiles--to hold the settings in the final output.

I can only hope that I've made enough mistakes and false assumptions to cancel each other out so many times that maybe something will come out okay.  And if not, then there will be multiple reasons why or why or not, it seems. Can't wait to see what happens when they get ahold of my files in Hong Kong for tests proofs off of who knows what system?  Especially since they sent me this comment

"No matter which printer you use, the quality of your scans will lead to the same result, which will not vary with the printer."

So far, I have let this comment pass in stride. I figured at least I have someone in Hong Kong communicating directly with me, which has been hard to achieve with other print reps. Also they have been willing to discuss issues such as my Pagemaker files, PDF's, embedded profiles, dot gain, UCR/GCR, etc.

Similar to other posts I read in archives, my Hong Kong print rep advised 12% dot gain (which Dan questioned and probably rightly so), which I dutifully set, and UCR with total ink 330% and black ink 90%, but I chose light GCR with 5% UCA with same ink settings.  After converting to my new profile (and embedding/sorry Stephen), I then fiddled with images and came up with the best output I could on my own printer and sent off files and my composite proofs.  Anyone want to predict the results?  (They may be using film rather than CTP and printing straight from my Pagemaker files.  They have offered to test at both 175 and 200 lpi.

Thanks again.  Even if I have screwed up big time, the learning process here is fascinating.  Regards, Kathleen

--
Kathleen M.K. Menke
Crystal Images
http://kcd.com/ci
___________________________________________________________________________

   Date: Sun, 12 Dec 2004 22:04:29 -0800
   From: rudy harvey
Subject: Custom CMYK

It should be noted that the Custom CMYK settings existed long before the Swop V2 profiles.

Custom CMYK was in Photoshop 2.5 I believe. The Swop V2 profiles were introduced in Photoshop 6. That's gotta be at least 5 years. There's no way the Swop V2 profiles could be based on the Custom CMYK settings. They are both useful, but are completely different.

I frequently use the Custom CMYK to make a proof setup to simulate proofs from printers who won't or can't send me a profile for their proofer or press.

The The Swop V2 profiles we use for about a third of our rgb to cmyk conversions the other two-thirds use a custom profile we made and edited  just for conversions.

My .02

Rudy Harvey
Color Management Consultant
Adobe Certified Expert -Photoshop
Certified Photographic Consultant
___________________________________________________________________________

   Date: Mon, 13 Dec 2004 07:39:25 -0000
   From:Stephen Marsh
Subject: Re: Custom CMYK

Is it possible to state which of the current profiles, the Custom
CMYK profile most closely relates to?

Photoshop 5 Default CMYK.icc (which was generated by Custom CMYK by Adobe, it should match SWOP Coated 20%, Med GCR, 100K, 300TIL, 0UCA).

The legacy Custom CMYK engine in no way, shape or form ralates to any machine measured and profile generating package produced ICC profile.

About the only common thing is that they both mention SWOP in their name - but I can show you at least half a dozen "SWOP" profiles which are all different anyway.

As your are making your own profile via Custom CMYK by entering the settings your HK printer recommends, it should all be editable. But if you have made the conversion first before you know what separation specs to separate to...

If your files are in CMYK and you then change your Custom CMYK profile settings to some new ideal value, then you will need to reconvert as the profile is only being used for display once the file in CMYK.

If the file was RGB and you change your Custom CMYK then it would not matter and the RGB would be both separated and displayed with this profile.

Stephen Marsh.
___________________________________________________________________________

   Date: Mon, 13 Dec 2004 09:02:53 EST
   From: Dan Margulis
Subject: Re: Custom CMYK

Kathleen Menke writes,

Is it possible to state which of the current profiles, the Custom
CMYK profile most closely relates to?

No. They are all different in different ways.

"Underestimation of black dot gain"--meaning printed result will be
more dark in output than anticipated by what is viewed on the monitor?

A better phrase would be "muddier-looking", as all but the purest colors would be contaminated by excessive black ink.

And if so, can this be compensated for (somewhat) by
"lighten/contrast" and evaluating one's own proof off one's own
inkjet?.

No.

The 100% black ink limit can be edited, correct?  As well as the
UCR/GCR settings?  This to me seems the value of the custom CMYK
profile?

Correct.

But then there was that niggling comment about editing this profile
only affects screen view and not output?  If so, then what's the
point of editing?

The point is to get not just a better preview but a better separation. If you edit the profile AFTER the file is already in CMYK, then, yes, it affects only the screen view. However, if you edit it BEFORE the file is in CMYK, then it will be taken into account during separation, and the resulting CMYK file will be different (and thus will output differently) than if the profile had not been edited.

Dan Margulis
___________________________________________________________________________

   Date: Mon, 13 Dec 2004 07:41:55 -0700
   From: Andrew Rodney
Subject: Re: Re: Custom CMYK


On 12/13/04 12:39 AM, "Stephen Marsh"  wrote:

About the only common thing is that they both mention SWOP in their
name - but I can show you at least half a dozen "SWOP" profiles which
are all different anyway.

Pretty frightening! And I1d agree totally that what you say above syncs up with my experiences with 3SWOP2.

As your are making your own profile via Custom CMYK by entering the
settings your HK printer recommends, it should all be editable. But
if you have made the conversion first before you know what separation
specs to separate to...

Exactly. And while this allows some editing, it1s not the full Monty. That is, you1re still basing the conversions on a huge part of the engine you can1t configure. You can measure some 61 or so color patches and make a custom ink model and tweak but at this point, you probably have the equipment to roll a true CMYK ICC profile. So the classic CMYK engine does provide a heck of a lot more tweaking then you1d have if you only had the U.S. Web Coated (SWOP) v2 profile at your disposal, it1s still not giving you a huge degree of control over the separation process.

Andrew Rodney
___________________________________________________________________________

   Date: Mon, 13 Dec 2004 15:20:08 +0000
   From: Richard Kenward
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

I jc castronovo  writes

This leads me to ask the question of how much control a pressman has on the
job. If our files aren't exactly what they should be, when is the job
hopelessly lost? I know that's a lot to ask for in such a forum as this, but
I know that a lot can be done on press. How do we know when a pressman is
just being lazy and blaming the files? What should we expect of him?

Dear John

I believe that going to press checks is an important part of getting a "feel" of what is possible and what is not, especially if it is a job of any size and by that I mean number of pages involved.  It also gives you a pretty good idea of if the outfit are up to the job or floundering!

Cheers

Richard
--
Richard Kenward  www.precision-drum-scanning.co.uk
___________________________________________________________________________

   Date: Mon, 13 Dec 2004 09:35:51 -0600
   From: "drhobbes"
Subject: Re: Apology

From: Dan Margulis

With respect to the book, because of this effect, that certain people can't
get a handle on certain concepts, beginning with the last edition there are
boxes in most chapters called "Stumbling Blocks": concepts that I think are
clearly explained elsewhere but, for some reason, certain readers aren't grasping.
For the next edition, I'll put this one in.

Thanks, Dan! Undoubtedly there are many among us who will appreciate this addition.

Is there any possibility of your telling us when to expect publication of your new edition?

Howard Smith
___________________________________________________________________________

   Date: Mon, 13 Dec 2004 18:18:41 -0800
   From: Jono Moore
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

On Fri, 10 Dec 2004 19:00:44 -0600, drhobbes wrote:

    What kind of profile do the majority of prepress people use for
conversion to CMYK?

I use custom CMYK, because that is how I learned to do it in Photoshop years ago.

I've recently started playing with the Sheetfed v2 profile, after recommendations from the folks on this list. But it feels a bit too much like a black box to me. :-)

Unfortunately I'm not doing a lot of 4 colour work at the moment so I don't get a lot of chance to experiment.

As an aside, does anyone know what sort of settings the Adobe profiles use for black generation, etc?
___________________________________________________________________________

   Date: Mon, 13 Dec 2004 18:32:18 -0800
   From: Jono Moore
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

On Sun, 12 Dec 2004 16:50:34 -0500, jc castronovo  wrote:

This leads me to ask the question of how much control a pressman has on the
job. If our files aren't exactly what they should be, when is the job
hopelessly lost? I know that's a lot to ask for in such a forum as this, but
I know that a lot can be done on press. How do we know when a pressman is
just being lazy and blaming the files? What should we expect of him?

The pressman has final control of colour. They run "to the numbers" (Standard Ink Density), but have a _lot_ of control over the final output. A good pressman can save bad art (to some extent but probably at the expense of something else on the sheet).

But there is a lot more to it than just setting ink densities. A lot of issues arise from what else is on the sheet around the images - like large areas of solid colour with a picture in the middle (do you want to match that 4 colour build of a Pantone colour or do you want the picture to look good?) - and where the images are on the sheet in relation to sheet travel through the press - what is before and after the picture and how much ink the press has to carry.

And, of course, every press is different with different capabilities and deficiencies. It all boils down to mechanics and physics.
___________________________________________________________________________

   Date: Tue, 14 Dec 2004 11:36:28 EST
   From: Dan Margulis
Subject: Re: Apology

Howard Smith writes,

Is there any possibility of your telling us when to expect publication of
your new edition?

My current plans are to have a book on the uses of LAB in the second half of 2005, and a new edition of Professional Photoshop sometime in 2006.
Dan Margulis
___________________________________________________________________________

   Date: Tue, 14 Dec 2004 11:34:41 EST
   From: Dan Margulis
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

Jono writes,

As an aside, does anyone know what sort of settings the Adobe profiles
use for black generation, etc?

The black is slightly darker than the Light GCR setting in Custom CMYK, and has a maximum value of 90%.

Dan Margulis
___________________________________________________________________________

   Date: Tue, 14 Dec 2004 12:21:02 -0500
   From: "Gene Palmiter"
Subject: Re: Apology

Good to know. I bought a copy off Amazon a couple of months ago and could not tell which was the newest....and didn't get the newest. But, as it's confounding me enough for now, I can wait a year for the next version.
___________________________________________________________________________

   Date: Mon, 13 Dec 2004 11:25:55 -0900
   From: KmKm/Crystal Images
Subject: Custom CMYK


Thank you Stephen.  This is helpful.  I believe these were the assumptions I was going on.  Because I inadvertently converted first with the Photoshop defaults (I wasn't even aware of what they were or that they mattered), I believe I both converted (from default CMYK to Kathleen's Custom CMYK with my changed settings) and saved as embedded to update and hold settings through Pagemaker links and transfer of project (seven CD's worth) to Hong Kong.  In this way, I hope to learn from results of wet proofs.  Next time, I would approach the project differently, getting my settings where I want them before converting from RGB.  What I want to avoid is muddy prints and washed out highlights.  We'll see.  Kathleen
___________________________________________________________________________

   Date: Tue, 14 Dec 2004 15:28:10 -0600
   From: "drhobbes"
Subject: Re: Apology

On 12/14/04 Gene Palmiter wrote:

Good to know. I bought a copy off Amazon a couple of months ago and could
not tell which was the newest....and didn't get the newest. But, as it's
confounding me enough for now, I can wait a year for the next version.

Gene,

    Several days ago a lady on an elevator noticed me carrying one of my many Photoshop books and commented on how hard it is to learn.  When I responded that she will not learn it from a single book, she seemed disappointed.  "So I guess Photoshop for Dummies won't do it?" she asked.

    No, Gene, there is no magic book on the market.  Dan's Professional Photoshop comes closer than anything I've found.  In the beginning it was like trying to learn Russian by reading a Russian/English dictionary (I made his book my primary text when I scarcely knew enough to be able to save an image file in Photoshop).  Now, in spite of the low opinion Dan must have of my current knowledge of Photoshop from all my puzzled queries, reading it is almost like reading a Reader's Digest article.  You get from there to here by reading.  Over and over and over until the copy falls apart in your hands (about six readings per volume).  When you have no idea what he's talking about, look up things like Apply Image and blending modes in another Photoshop book (Professional Photoshop was never intended as a first text). Then come back to Dan's book and read it some more.  Ask questions.  Lots of questions, but only when you can't find the answer in your books.  If you just accept the fact that this is a difficult program to learn when you're starting from scratch, and you refuse to give up on it, you'll find yourself becoming pleasantly expert at it.  When you despair of gaining the knowledge exhibited by many of the respondents on this forum, keep in mind that they are speaking from years of experience coupled with attendance at and participation in national seminars, both Photoshop seminars and those in the broad field of commercial printing.  You have to start before you can become that expert.

    As for the book you've just bought--read it.  Every edition has things in it that you may not find in the others.  After studying Professional Photoshop for almost three years, I learned still more by finally obtaining and reading a copy of his original book (Makeready, I believe it was called).  Had I not bought and read all of his other editions, I would be searching from them now at used book dealers.

Sincerely,

Howard Smith
___________________________________________________________________________

   Date: Wed, 15 Dec 2004 16:23:14 -0000
   From: "ken cavanagh"
Subject: Re: sheetfed vs. web coated SWOP profiles
 
If US Web Coated (SWOP) v2 is built with the settings as Dan describes and 300% TAC, is this profile going to show a significant improvement over similar custom cmyk settings?

I gather it is a more refined profile, but is it difference that one would see in reproduction for a "typical" image?  

Thanks
Ken
___________________________________________________________________________

   Date: Wed, 15 Dec 2004 11:53:03 -0500
   From: Henry Davis
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

That the press operator has "final control of color" strikes me as one of those statements like "the pilot has final control over the quality of the ride".  In a way it is true, because a press operator can mess up an otherwise well prepared job, the same as a "lazy" pilot might   not offer a pleasing ride in smooth air.  But the weather occasionally plays a role.  Too few designers however are concerned about basic print design, and the points you present concerning the limits of on-press control should be Designer 101.  Never mind ghosting and other fun topics.

If the job looks different than expected, then all factors are suspect.   Looking at the beginning as well as at the end of the process when troubleshooting is a realistic and expedient approach.  Sometimes print brokers and designers accept responsibility for a botched job, but this usually happens when they are presented with all of the facts.  It may not occur to some of them how many times their rears have been saved by pre press and press operators who quietly stepped in and fixed their problems - without mention and for no fee.  When presented with facts that implicate poor design or image prep, it is amazing how many refuse to accept responsibility.  But then, it's less amazing when one considers that designers may be under the misconception that it has been their files that are actually used.  This is one of the "Catch 22's" of the print trade.

It will probably continue this way, because often times it's so much easier to fix a customer's files than get into a long drawn out dialogue that seldom produces anything but contempt for both parties.  Rarely, though, one does encounter customers that truly want to learn, even when it is by their mistakes.  How many of us really spend much effort learning from a project that came off just right?  Files that are fixed on the qt by pre press or press operators are jobs that come off just right, and the designer receives an award.

Henry Davis
___________________________________________________________________________

   Date: Wed, 15 Dec 2004 15:37:48 EST
   From: Dan Margulis
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

Ken writes,

If US Web Coated (SWOP) v2 is built with the settings as Dan describes below and 300% TAC, is this profile going to show a significant improvement over similar custom cmyk settings?

If you're talking a straight Custom CMYK default, yes; if a properly adjusted one, no.

I gather it is a more refined profile, but is it difference that one would see in reproduction for a "typical" image?

All conversion methods get slightly different results, and will work well on  certain images and less well on others. Assuming that you're not planning to  do much to the CMYK file after conversions, Images with deep blues tend to get  poorer results with the SWOP v.2 profile than with a reasonable Custom CMYK, but  fleshtones are better, etc., etc.

If you are planning any significant work after the conversion, SWOP v.2 is a poor choice because of somewhat  excessive black generation and because its shadows are too red. Plus, of course, you can't override its black generation settings.

Dan Margulis
___________________________________________________________________________

   Date: Thu, 16 Dec 2004 17:12:04 -0000
   From: "ken cavanagh"
Subject: Re: sheetfed vs. web coated SWOP profiles
 
Dan, thanks for your reply, which leads me to ask...by "properly adjusted" you mean not the default custom cmyk, but something similar to US web coated swop, like swop coated, light gcr, 90% BIL and 300% TAC?

The fleshtones being better is something I will experiment with.  Are there other subjects, in your experience, that have yielded a pro/con result from custom  cmyk vs. web coated swop v2?  You hint that there are with "etc., etc."

Since my work involves a variety of photographic subjects, (most of it is printed with web presses on textbook quality stock), this is certainly of interest.

Thanks
Ken

Ken Cavanagh
Photographer
Macmillan/McGraw-Hill
___________________________________________________________________________

   Date: Thu, 16 Dec 2004 12:29:25 -0500
   From: Terry Wyse
Subject: "Custom CMYK" ideas (was: sheetfed vs. web coated SWOP profiles)

Dan, since you seem to have the ear of Adobe, here's an idea I have to get away from the "Custom CMYK" stuff (based on sort of "mystery meat" press behavior) and towards customizable profiles based on real press behavior:

Within the "Custom CMYK" engine, embed the actual TR001 (Web/SWOP) and TR004 (commercial sheetfed) characterization data and then build a custom profile based on this (yea, I know, CGATS or whomever would probably want some kind of royalty for this but embed it in such a way that it cannot be extracted). Why not go the whole nine yards and include flexo and newsprint data? So you would simply choose the base characterization data, much like you do now when choosing an ink set, set your separation parameters and build yourself a custom TR00-whatever profile. It would also be nice to select a custom paper white value to use but that's more for proofing than actually producing a separation.

To prevent the casual user from doing something silly, I'd also like to see a "flag" or warning if an end-user requests, for example, a total ink limit that exceeds what's recommended by the standards group that supplied the data, like choosing 360% for a "SWOP" profile. UCR/UCA/GCR could be left totally at the user's discretion but things like total ink limit/dot gain/K limit should throw up a flag if mucked with beyond a certain threshold.

While were at it.....

When I'm doing custom press profiling for a customer, I typically make them a small "set" of profiles with differing amounts of UCR/GCR etc. that they can select from to tailor the separation for certain types of images. Same colorimetric/measurement data but just different separation parameters.

What drives me insane is that Photoshop will flash up all kinds of warnings if they've used one of these profiles that doesn't HAPPEN to be the standard one they've chosen as their default CMYK working space even though all these profiles were made from the SAME colorimetric data and will effectively return a soft-proof that's identical. Photoshop thinks the profile is "different" when, from a certain point of view, it's really not.

I can think of a couple of ways around this. One, have as part of the standard ICC format the inclusion of the original colorimetric data within the profile (GMB ProfileMaker already does this). Photoshop could then quickly check the "colorimetric" tag and realize that this profile is effectively the same profile with only the separation parameters being different.

The other option would be for Photoshop to do quick "profile round-trip" where they extract a small number of "patches" by doing a CMYK->Lab transform and compare this data between the two profiles (sort of like doing a checksum of the profile). It's a quick thing to do for maybe 100-200 patches and possibly this data could be cached in some way so it doesn't have to do this every time an image with an embedded profile is opened.

I know this isn't exactly the forum for this sort of thing but I thought I'd just throw it out there.
 
Calibrationist regards,
Terry
:-)
___________________________________________________________________________

   Date: Thu, 16 Dec 2004 15:03:12 EST
   From: Dan Margulis
Subject: Re: Re: sheetfed vs. web coated SWOP profiles

Ken Cavanagh writes,

Dan, thanks for your reply, which leads me to ask...by "properly adjusted" you mean not the default custom cmyk, but something similar to US web coated swop, like swop coated, light gcr, 90% BIL and 300% TAC?

Those are steps in the right direction. You'd also need to adjust cyan dot gain down and black dot gain up, for starters, plus you do not assume that the default 20% dot gain is correct.

Dan Margulis
___________________________________________________________________________

   Date: Thu, 16 Dec 2004 15:26:57 EST
   From: Dan Margulis
Subject: Re: "Custom CMYK" ideas (was: sheetfed vs. web coated SWOP profiles)

Terry writes,

Dan, since you seem to have the ear of Adobe, here's an idea I have to
get away from the "Custom CMYK" stuff (based on sort of "mystery meat"
press behavior) and towards customizable profiles based on real press
behavior:

I am getting ready to leave for the airport for ten glorious days in the Puerto Rican sun, so I can't reply at the length that the post deserves. Will do so either upon return or if I can tear myself away from the beach and find a reliable dialup connection.

Briefly, though: the idea is commendable, but it wouldn't work in real life. The Custom CMYK engine is so heavily kloodged that if you enter correct colorimetric information for ink values you get a separation that's too gray, among other problems. You have to change the numbers to specify poorer-quality inks.

Also, there aren't enough data points in the Custom CMYK setup to get much of a match with any other profile, although one might come close.

Plus, you'd have to ask why you'd even want to. Custom CMYK profiles don't support any of the nice soft-proofing features that are a major attraction of third-party profiles.

More in a few days.

Dan Margulis
___________________________________________________________________________

   Date: Sat, 18 Dec 2004 09:58:49 -0500
   From: Ric Cohn
Subject: Re: "Custom CMYK" ideas (was: sheetfed vs. web coated SWOP profiles)

Terry,

Perhaps you could post your request on the "Feature Request" section of Adobe's Photoshop User Forum. I believe if enough truly knowledgeable CMYK users posted their views that Adobe might decide it was worth the time and expense of adding and updating CMYK features. I can't think of any other area of Photoshop where there is such a gap between the usability and understanding of the "beginner" and the "advanced" use of a Photoshop feature.

Ric Cohn
___________________________________________________________________________

   Date: Sun, 19 Dec 2004 13:12:24 -0800
   From: Peter Constable
Subject: RE: "Custom CMYK" ideas (was: sheetfed vs. web coated SWOP profiles)

Terry-

Don't users want to know if profiles are different with respect to separation parameters like K generation?  I see difficulties here.

Example 1: Someone creates profiles using the same colorimetric data but different software tools (and different gamut mapping strategies).  They may want to know which type of gamut mapping/profile is associated with their image so they have the option of reseparating using a different strategy/profile.

Example 2: Someone creates profiles using the same colorimetric data but different K generation settings.  They may want to know which type of K generation is associated with their image so they have the option of reseparating using a different K gen parameter.

Keying warning messages off colorimetric data used by the profile may limit their usefulness?

_peter

Peter Constable
Adobe Systems, Inc.
___________________________________________________________________________

   Date: Sun, 2 Jan 2005 15:16:29 EST
   From: Dan Margulis
Subject: Re: "Custom CMYK" ideas (was: sheetfed vs. web coated SWOP profiles)

Two weeks ago, Terry Wyse posted a lengthy commentary addressed to me on editing of CMYK profiles. Because I was on my way to the airport at the time, to a place where web access is difficult, I replied only briefly and said that I would add more when I returned.  Now that I am properly suntanned and the island of Puerto Rico has been substantially depleted of rum, I offer the following.

Dan, since you seem to have the ear of Adobe...

Yeah, the deaf one.

...here's an idea I have to get away from the "Custom CMYK" stuff (based on
sort of "mystery meat" press behavior) and towards customizable profiles
based on real press behavior: Within the "Custom CMYK" engine, embed the actual
TR001 (Web/SWOP) and TR004 (commercial sheetfed) characterization data and then
build a custom profile based on this (yea, I know, CGATS or whomever would
probably want some kind of royalty for this but embed it in such a way that it
cannot be extracted).

As I indicated in my first post, this wouldn't work, because the Custom CMYK engine is so heavily kloodged that entering the "correct" colorimetric values won't work. You can get good profiles out of it but they have to be based on a "make the magenta ink less vivid" approach rather than on measurements. That's fine with me, because I've done so many profiles that way on so many different platforms that it seems a lot more natural (not to mention quicker, and higher quality) than doing it with a machine. However, it excludes anyone who hasn't had that kind of experience, and anyone else who would prefer to use machine measurements as a starting point.

Why not go the whole nine yards and include flexo and newsprint data? So
you would simply choose the base characterization data, much like you do now
when choosing an ink set, set your separation parameters and build yourself a
custom TR00-whatever profile. It would also be nice to select a custom paper
white value to use but that's more for proofing than actually producing a
separation.

Sure. Presumably you would also want to add more sampling points than Custom CMYK now has. That's what *should* have been done in 1998 and at every point since. What you're describing is basically a completely revamped version of Custom CMYK, incorporating all of the advantages that are available with the ICC format, while not throwing away the indispensable ability to edit.

UCR/UCA/GCR could be left totally at the user's discretion but things like
total ink limit/dot gain/K limit should throw up a flag if mucked with beyond
a certain threshold.

Fine with me. It would stop a lot of confusion about the seemingly contradictory settings in the separation algorithm. Like, if you are choosing a heavier GCR you are ipso facto specifying a lower total ink, and if you nevertheless insert a higher total ink limit, it would probably help to have the program warn you against it. Similarly, since SWOP dictates an ink limit of 300%, if you put 360% in, it's not a SWOP-compliant profile. So, the program could advise us of that.

While were at it.....When I'm doing custom press profiling for a customer,
I typically make them a small "set" of profiles with differing amounts of
UCR/GCR etc. that they can select from to tailor the separation for certain types
of images. Same colorimetric/measurement data but just different separation
parameters.

This is a great thing to do. The problems are, first, that devices shift over time, and second, that the customer may agree that your profiles are good but disagree that they are perfect.

What drives me insane is that Photoshop will flash up all kinds of warnings
if they've used one of these profiles that doesn't HAPPEN to be the standard
one they've chosen as their default CMYK working space even though all these
profiles were made from the SAME colorimetric data and will effectively return
a soft-proof that's identical. Photoshop thinks the profile is "different"
when, from a certain point of view, it's really not.

This is simply a symptom of a much larger disease. The idea of an alert when Photoshop encounters a nonstandard profile is a great one; unfortunately the Photoshop interface is so lame that anybody who takes in files from many different sources has to turn it off.

In a perfect world, I want to know whether a CMYK file has been separated in an unusual way just as much as I want to know whether an RGB file has an embedded profile. I don't guarantee to make use of either piece of information but I would like to know about their existence. Unfortunately, as you point out, if we leave the warning on we are rapidly driven insane by interminable bogus alerts.

The obvious solution, as I've pointed out many times, is to have an option in that alert box saying "Stop warning me about this particular profile until the next time I quit Photoshop."

Thanks for a thoughtful post.

Dan Margulis
___________________________________________________________________________

Adobe Photoshop training classes are taught in the US by Sterling Ledet & Associates, Inc.