Dan Margulis Applied Color Theory
Eurostandard Inks and Custom CMYK
Eurostandard inks – anyone?
Posted by: "Marco Olivotto - LoL Productions
snc"
Mon Jan 12, 2009 4:54 am (PST)
Hello all,
I would like to ask if any of the listmembers have any
significant experience to share with the usage of Custom CMYK with
Eurostandard inks.
The problem we're having is this: we're dealing with
two pressing companies which recommend we should use the ECI profile ISO
Coated v2 (300%). The problem with this profile is that it has a very high
black ink limit (95%, as far as I can see), and the black generation is
rather heavy. It's not a Light GCR profile at all. Several of our critical
images, especially when there are dark and desaturated colors, print too
dark. The colors, in general, are OK (hue, I mean), but there is an obvious
problem with black being too heavy. My idea was that I should try Custom
CMYK using Eurostandard inks as a base, but this doesn't really work. For
instance, a certain skin tone (75L 19a 24b) converts into ISO Coated v2 as
7C 38M 43Y 2K and into FOGRA27 as 9C 39M 43Y 0K – fair enough. But it
converts into an unbelievable 9C 51M 47Y 1K if I use Custom CMYK,
Eurostandard inks, Light GCR, 80% black ink limit and 300% total ink. The
fact that M > Y when a is significantly lower than b is absurd, to me;
and indeed, if I preview the output by assigning the ISO Coated v2 profile
to the image converted as above it looks unacceptable. The shifts in color
are neither obvious nor uniform: a light yellowish red as the one listed
above will turn more magenta, but at the same time a purplish brown will
lose points both in the a and the b; a light blue will get colder and
definitely move towards green.
The bottom line is that if I use this profile the
colors will be unacceptable. Even pasting this separation on top of the ISO
Coated v2, obviously retaining the numbers, and going to Luminosity mode
will produce heavy shifts.
The combination that works better, for me, is to use
Custom CMYK with SWOP inks (in Europe!!!) and do the Luminosity layer trick
on top of the ISO Coated v2. There is a color shift, but it is far less
damaging than the one I'm experiencing with Eurostandard inks. My strong
suspicion is that the values for these inks in the Photoshop engine are
wrong; or I have to admit that the canned profiles of ECI and FOGRA are
*not* based on these inks... which is obviously wrong (although, I must
say, I've found little or no information on the Web on the precise
character of Eurostandard inks). Any suggestion is welcome. I've found an
archived post from this list dating back to 2005 which points to the
"wrong" values of the M ink. I've tried the correction suggested
(+2 points in a and b values in the M and MY fields in the personal inks
window in Photoshop) but it doesn't really change much to me...
Finally, my current practice as a defensive move
against black being too heavy is to simply separate with ISO Coated v2 and
then apply a curve which will cut the black roughly above 50%, bringing
100% down to 80% or so. Black art, indeed... :-)
All the very best and thanks in advance for your
replies,
M.
---
Marco Olivotto
LoL Productions snc
1/A Via per Sasso
38060 - NOGAREDO (TN)
ITALY
Ph/Fax: +39-0464-490614
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: Terence Wyse
Mon Jan 12, 2009 8:27 am (PST)
Hello Marco,
The problem here is that Eurostandard is nothing like
the ink colors of ISO 12647-2 which is what ISO Coated is based on (the
actual "data set" is called FOGRA 39).
I took the LAB numbers from Eurostandard for the solid
primaries and secondaries (CMYK+RGB) and compared them to ISO Coated and
got an average delta E value of 8-12 (depended on whether I took the LAB
values from the custom ink values in Custom CMYK... or if I make a profile
and then checked them using the Photoshop color picker with the rendering
intent set to absolute). The Eurostandard inks were both "less
colorful" (reduced chroma) and had significant hue differences
compared to ISO Coated. Cyan and Blue (C+M) were the biggest offenders. The
Cyan ink had considerable blue/magenta hue shift causing the C+M overprint
to shift hue to purple/violet. Yellow had a significant hue shift towards
green.
If you've got a profiling application or know someone
who does, you could easily take the FOGRA 39 data on which ISO Coated is
based and build your own profile with whatever black generation you prefer.
Contact me off-list and I might be willing to build you a profile from
FOGRA 39 if you have a clear idea of what you're looking for.
Regards,
Terry Wyse
______________________________
Terence Wyse, WyseConsul
Color Management Consulting
G7 Certified Expert
___________________________________________________________________________
Re: Eurostandard inks – anyone?
Posted by: Stephen Marsh
Mon Jan 12, 2009 5:47 pm (PST)
Marco, I would forget which Custom CMYK inkset/stock
combo you are using, it does not matter if you are mixing SWOP with a
proper Fogra profile here - all that matters is that you like the resulting
numbers and that the proofing (soft and or hard) is acceptable.
I think the way quoted above is probably the best for a
production situation where consistency and repeatability may be required.
In addition to layering the Light GCR/UCR Custom CMYK
(SWOP) file in luminosity mode over the Fogra39 conversion - you could
deselect the CMY channels in advanced blending so that only the K data is
being blended in luminosity mode with only the underlying K data changing.
Best,
Stephen Marsh
___________________________________________________________________________
Re: Eurostandard inks � ’¶ anyone?
Posted by: "Marco Olivotto
Tue Jan 13, 2009 7:57 am (PST)
Stephen Marsh wrote:
In addition to layering the Light GCR/UCR Custom CMYK
(SWOP) file in
luminosity mode over the Fogra39 conversion - you
could deselect the
CMY channels in advanced blending so that only the K
data is being
blended in luminosity mode with only the underlying K
data changing.
This is extremely sensible, Stephen... good advice,
thanks! I would like to add that from the first tests I've made, and as
expected, there's no actual need to turn to Luminosity mode in this case,
because CMYK computes luminosity separately for the CMY and K channels. One
step less in the action :-).
Also, many thanks for your offer, Terry � ’¶
I will get in touch with you if needed and if I get stuck again.
And thumbs up to Henk for his communication off-list; I
am very pleasantly surprised at how many people are willing to help on this
group: this is not the first time this happens and it is a very good thing.
Best to all,
M.
---
Marco Olivotto
LoL Productions snc
1/A Via per Sasso
38060 - NOGAREDO (TN)
ITALY
Ph/Fax: +39-0464-490614
___________________________________________________________________________
Re: [colortheory] Re: Eurostandard inks � ’¶
anyone?
Posted by: Terence Wyse
Tue Jan 13, 2009 12:53 pm (PST)
Marco, I've got a profile made that should do what you
need.
I've made a profile from the FOGRA 39 data set where
the black starts right around c50m40y40 and peaks at about 85% using
moderate-to-light GCR and a 300% total ink limit.
If you want it, I can send it to you ASAP. With this
profile you won't have to jump through all those
layering/blending/luminosity hoops in an effort to "streamline"
your workflow! ;-)
Regards,
Terry
___________________________________________________________________________
Re: [colortheory] Re: Eurostandard inks �
?’¶ anyone?
Posted by: Terence Wyse
Tue Jan 13, 2009 8:17 am (PST)
I'm not sure I'm following you Stephen. You're saying
that jumping through these layering/blending hoops is more productive,
consistent and repeatable than having a proper ISO Coated profile with the
black generation you prefer to start with?
Regards,
Terry Wyse
___________________________________________________________________________
Re: Eurostandard inks – anyone?
Posted by: Stephen Marsh
Tue Jan 13, 2009 2:00 pm (PST)
Hi Terry, as you know, such a profile was not available
at that time. Marco was exploring manual curve edits vs. separation using a
different method (Custom CMYK) vs. layering and blending a CMYK file with a
more appropriate K generation. Since the OP and the comment that I made in
relation to the OP, you have kindly made what I presume is a profile that
is better than the one that is publicly available and blessed by FOGRA.
This latter, ground hog day situation of using a more appropriate profile
was not being commented on in my OP.
If this profile met the requirements and my standards,
then I would consider this the ideal option over hacking a less than ideal
separation made from a less than ideal standard industry profile.
Respectfully,
Stephen Marsh
___________________________________________________________________________
Re: [colortheory] Re: Eurostandard inks �
?’¶ anyone?
Posted by: Terence Wyse
Tue Jan 13, 2009 10:59 pm (PST)
While I'm not a...ahem...strong advocate of the legacy
"Custom CMYK" method, this COULD'VE been accomplished (or close
to it) using that method. All one would've needed is the CMYK+RGB
primary/secondary Lab values plus the C+M+Y and paper white Lab and the
tone curve for FOGRA 39. Most, if not all, of this data could've been
extracted from the standard ISO Coated profile right from within Photoshop.
But I simply chose the route that was the easiest and most familiar to me
and that was to import the FOGRA 39 directly into a profiling application
and let IT do the work for me.
:-)
Regards,
Terry Wyse
___________________________________________________________________________
Re: [colortheory] Eurostandard inks � ’¶
anyone?
Posted by: "Marco Olivotto - LoL Productions
snc"
Tue Jan 13, 2009 12:53 pm (PST)
Marco again...
The case I opened is closed thanks to Terry Wyse, who
sent me a profile based on FOGRA39 which he created and which works exactly
as I would expect it to work. It is, of course, a more straightforward way
to deal with my problem although the other possible solutions were also
very interesting.
Thanks Terry, and, I stress it again, this is a *very*
helpful list, both for the discussions and in practice.
Now, though, if only Adobe decided to include at least
a basic profile editor in Photoshop other than Custom CMYK the world would
be a (marginally) better place. ;-)
All the best,
M.
---
Marco Olivotto
LoL Productions snc
1/A Via per Sasso
38060 - NOGAREDO (TN)
ITALY
Ph/Fax: +39-0464-490614
___________________________________________________________________________
.
Lab values of inks (was: Eurostandard inks)
Posted by: "Marco Olivotto - LoL Productions
snc"
Wed Jan 14, 2009 9:09 am (PST)
I am sorry that I didn't read the posts yesterday when
I sent my latest one, or I would have added this...
Terry wrote:
While I'm not a...ahem...strong advocate of the legacy
"Custom CMYK"
method, this COULD'VE been accomplished (or close to
it) using that
method. All one would've needed is the CMYK+RGB
primary/secondary Lab
values plus the C+M+Y and paper white Lab and the tone
curve for FOGRA
39. Most, if not all, of this data could've been
extracted from the
standard ISO Coated profile right from within
Photoshop. But I simply
Yes, I thought about this and tried – but things
are not so simple. If you load a SWOP Web Coated profile as your default in
Photoshop and check the Lab values for the relevant colours you mention,
you will notice they're significantly different than the values in Custom
CMYK when you go and try to edit the inks in the Custom window. Same goes
for FOGRA: the official values are available, but when I tried dialling
them into their boxes the resulting profile bore little resemblance to any
"proper" FOGRA profile. To state it clearly – it was a
complete, utter mess.
Somewhere in PP5E Dan mentions that the numbers in the
engine are kludged, and I bet he's 100% right; I personally can't find any
obvious consistency between the values in there and other results even
inside Photoshop. Someday we may discover that the geometric mean of all
those values approaches Phi (golden section) when divided by the serial
number of the program, and we might then see some kind of light(ness)
– but I can't see this happening. ;-)
Also, one thing I noticed a while ago (and there was a
discussion on this list about this): the white point definition is
particularly meaningless. I was trying to put together a bogus profile to
proof how an image would print on yellowish paper, but there was no way I
could simulate yellow as paper colour when exporting to pdf via InDesign
and simulating the proper output condition, not even with *ridiculous*
amounts of yellow in the W. My ultimate feeling is that trying to seriously
mess with these numbers may lead to early insanity – something most
of us don't need, I guess. Now and again: "if only Adobe
would...". Known issue, huh?
Many thanks again for your attention, everyone.
M.
---
Marco Olivotto
LoL Productions snc
1/A Via per Sasso
38060 - NOGAREDO (TN)
ITALY
Ph/Fax: +39-0464-490614
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: Dan Margulis
Wed Jan 14, 2009 9:09 am (PST)
Already in this thread there seem to be three
misconceptions about how to use Custom CMYK to replace black generation in
an existing profile.
Marco writes,
Several of our critical images, especially when there
are dark and
desaturated colors, print too dark. The colors, in
general, are OK
(hue, I mean), but there is an obvious problem with
black being too
heavy. My idea was that I should try Custom CMYK using
Eurostandard
inks as a base, but this doesn't really work. (snip)
The bottom line is that if I use this profile the
colors will be
unacceptable. Even pasting this separation on top of
the ISO Coated
v2, obviously retaining the numbers, and going to
Luminosity mode
will produce heavy shifts.
The combination that works better, for me, is to use
Custom CMYK
with SWOP inks (in Europe!!!) and do the Luminosity
layer trick on
top of the ISO Coated v2. There is a color shift, but
it is far less
damaging than the one I'm experiencing with
Eurostandard inks.
Since you're using a luminosity layer in principle the
color formulations on top make no difference. The issue is whether the top
layer matches the bottom for darkness, not color. You're using FOGRA27 as a
base. Custom CMYK Eurostandard defaults give a darker separation than that.
SWOP v2 gives you something lighter. In either case there will be
significant shifts but I agree that
probably the Eurostandard defaults are worse.
To get it closer you have to edit the dot gain setting.
Since that isn't doable in SWOP v2, forget that option. Instead, find a dot
gain setting in Custom CMYK that closely matches the FOGRA27 darkness. The
higher the dot gain setting, the lighter the resulting sep. The
Eurostandard Custom CMYK default is 9%, about which default, the less said,
the better. However, I just did a test file where the bottom layer was
separated in FOGRA27 and the top layer (set to Luminosity) in Custom
CMYK/Eurostandard/16% and the results were just fine, a close match and
certainly far more printable.
Along the same lines, Stephen writes,
In addition to layering the Light GCR/UCR Custom CMYK
(SWOP) file in
luminosity mode over the Fogra39 conversion - you
could deselect the
CMY channels in advanced blending so that only the K
data is being
blended in luminosity mode with only the underlying K
data changing.
This is a bad idea. The black is being made much
lighter. To compensate, the CMY has to be much darker. Combining the
original CMY with a new, lighter black washes out the image. As long as the
top layer matches the bottom for darkness, the CMY component needs to be
active--it will be properly darker on the top layer.
And Terry writes,
While I'm not a...ahem...strong advocate of the legacy
"Custom CMYK"
method, this COULD'VE been accomplished (or close to
it) using that
method. All one would've needed is the CMYK+RGB
primary/secondary Lab
values plus the C+M+Y and paper white Lab and the tone
curve for FOGRA 39.
The value of said values is zero, in the warped world
of Custom CMYK. The interface looks like an invitation to enter measured
numbers, but anyone who does will get an unpleasant surprise. The only way
to get acceptable results is trial and error. If it looks like the cyan ink
needs to be bluer you make it bluer and if the magenta isn't saturated
enough you make it more saturated. Easy enough for those with a lot of
experience at calibrating stuff but beyond the capabilities of most others,
which makes it even more disgusting, if possible, that the Photoshop
engineers insist on giving people like Marco the choice of a fifth-rate
profile like FOGRA27 or upgrading to Custom CMYK, when it would be so much
easier to just let him edit what he wants to.
Dan Margulis
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: Terence Wyse
Wed Jan 14, 2009 7:26 pm (PST)
Not sure about the misconceptions since you can't use
Custom CMYK to replace the black generation in ANY existing profile.
My suggestion of plugging in LAB values to create a
custom ink set in Custom CMYK was under the assumption that it actually
worked. I understand what you're saying Dan but I'm not sure that
"proves" one way or another that entering in your own Lab values
in fact doesn't work.
In any case, it would appear creating a new profile
from the proper data set and setting your own black generation was indeed
the most efficient solution. Whether or not this capability is accessible
to the average Photoshop user is inmaterial to me.
You made reference to FOGRA 27 repeatedly. I'm not sure
where that comes from since ISO Coated v2 is based on the FOGRA 39 data
set. Perhaps you could clarify.
Regards,
Terry Wyse
___________________________________________________________________________
Re: Eurostandard inks – anyone?
Posted by: Stephen Marsh
Thu Jan 15, 2009 8:15 pm (PST)
Dan Margulis wrote:
This is a bad idea. The black is being made much
lighter. To
compensate, the CMY has to be much darker. Combining
the original CMY
with a new, lighter black washes out the image. As
long as the top
layer matches the bottom for darkness, the CMY
component needs to be
active--it will be properly darker on the top layer.
Dan, when I suggested this as it was indicated that the
colour was off when blending in Luminosity...so my mental "knee-jerk
reaction" was not to monkey with the CMY component.
I agree that even when the dot gain is corrected in the
Custom separation, that such a different K generation to the original will
indeed be lighter if one only replaces the original K with the new. There
are of course extra steps one can take to limit this.
This then creates the issue of having to boost the CMY
in an appropriate manner to match the new K component - which may lead to
more tail chasing than the original colour shift with the Luminosity blend.
As the thread ended with the generation of a new profile using more
appropriate settings for the images/printer in question, this issue was not
explored.
Stephen Marsh
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: Dan Margulis
Sat Jan 17, 2009 7:30 pm (PST)
Terry writes,
Not sure about the misconceptions since you can't use
Custom CMYK to
replace the black generation in ANY existing profile.
Maybe *you* can't. *I* can, and told Marco how.
My suggestion of plugging in LAB values to create a
custom ink set in
Custom CMYK was under the assumption that it actually
worked.
It's a reasonable assumption to make, as that's what
the interface seems to suggest. However, it doesn't happen to be correct.
I understand what you're saying Dan but I'm not sure
that "proves" one
way or another that entering in your own Lab values in
fact doesn't work.
As Marco pointed out, having tried it, they're not even
close. I've tested it extensively, because if it were actually true that
one could do this, it would have made all the difference in the world to
the adoption of the methods you practice.
In any case, it would appear creating a new profile
from the proper
data set and setting your own black generation was
indeed the most
efficient solution.
The two options as I understand them are:
DAN'S OPTION:
1) Evaluate the profile that the printer recommends and
observe that it's a disaster.
2) Find out what Custom CMYK dot gain value matches it.
3) Write an Action that separates a file twice and puts
the Custom CMYK version on top in Luminosity mode.
4) Drink beer.
TERRY'S OPTION:
1) Evaluate the profile that the printer recommends and
observe that it's a disaster.
2) Post message to colortheory list asking for help.
3) Wait for moderator to approve message.
4) Hope that a color management consultant reads
message and responds.
5) If it happens, exchange offline correspondence with
color management consultant.
6) Get color management consultant to send you a better
profile.
7) Hope that color management consultant agrees to do
this for free.
8) Evaluate the profile that the color management
consultant provides and decide whether it's worth using.
9) If not, revert to Dan's option.
10) If yes, convert files with this profile until your
first actual experience with the printer indicates you shouldn't, whereupon
you decide which option you want to choose all over again.
11) Pop Prozac.
It does not seem that we are in agreement as to which
of these methods is more efficient. That disagreement does not prevent me
from expressing appreciation for your generosity in providing the profile
to Marco, and thereby apparently solving his problem for the time being,
until he faces up to a new printer or finds out that this printer's
performance isn't quite what he was led to believe.
Whether or not this capability is accessible to
the average Photoshop user is inmaterial to me.
I should think that experience would have by now made
clear that it is *very* relevant to you. Remember, this debate has been
going on for a very long time. It was eleven years ago that the Photoshop
engineers announced that Custom CMYK was obsolete, and I said that if they
were serious about it they had to incorporate at least as much capability
in whatever the next step was, because nobody interested in quality was
going to downgrade to a system of uneditable profiles. I do not fault the
Photoshop engineers for thinking that they were right at the time, but
there comes a point when they--and you--have to admit that they have been
proven wrong. That point should not require eleven years to reach.
Remember, Marco and, earlier, Jeremy, are not
CMYK-proficient. They have little experience dealing with printers, but
they want to have their work print in a predictable way, maximizing their
chances for success even if the printing is not of good quality. And so, in
a world where few printers perform reliably and fewer know anything about
prepress, they're stuck learning Custom CMYK unless they want third-quality
results (or can persuade a color management consultant to help them every
time they switch printers or learn something new about the printer they're
using.)
Helping out Marco is most commendable, but what happens
when he learns that this printer doesn't really print to any known
standard, and your profile, though good in theory, is significantly off in
practice? Having to have an on-call color management consultant when the
solutions already exist for free in Photoshop doesn't really seem
practicable.
I note that, since the CS4 release, a Photoshop
engineer, on the ColorSync list, again advocated pulling Custom CMYK out of
Photoshop and made it clear that the motivation was revenge, a deliberate
attempt to sabotage CMYK users. Shortly afterward, on Adobeforums, a
question came up about yet another thing that can currently only be done in
Photoshop via Custom CMYK (generating a quality 4/c B/W), and the advice
came in from others, mostly nonexperts AFAIK, that the choices were either
Custom CMYK, buy a profiling package, or hire a color management
consultant. Nobody offered any criticism of Adobe. Nevertheless, the same
engineer, who had not previously participated in the thread, butted in with
the following (this is the full post): " 'Custom CMYK' should have
been renamed 'Obsolete PoorQuality CMYK' a long time ago."
And that was it. No solution offered (because there is
none to offer within Photoshop other than Custom CMYK), just a gratuitous
insult of an Adobe client.
When one method is being belittled and declared
obsolete for a period of eleven years, yet those interested in quality
won't abandon it, perhaps some positions should be reconsidered. And as for
being material to you, all you have to do is ask yourself why those who can
produce better images have prospered over these eleven years while color
management consultants have by and large starved. Selling color management
services to those who don't care about quality is a flawed business model.
Trying to sell methods that produce third-rate color to those who do is
equally flawed IMHO.
You made reference to FOGRA 27 repeatedly. I'm not
sure where that
comes from since ISO Coated v2 is based on the FOGRA
39 data set.
Perhaps you could clarify.
It came from the OP, which specifically references
FOGRA27 as giving values he considered reasonable. If he had said some
other profile I would have used that, it would presumably have changed the
dot gain setting, but otherwise isn't pertinent to what I posted.
Dan Margulis
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: Terence Wyse
Sun Jan 18, 2009 12:03 pm (PST)
On Jan 17, 2009, at 9:57 PM, Dan Margulis wrote:
Maybe *you* can't. *I* can, and told Marco how.
No, you CAN'T since you yourself stated you can't match
the color of the primary and secondary inks of the profile (ISO Coated)
using Custom CMYK. Finding and adjusting the dot gain is only half the
story. The other half is, uh, color. Finding the CMYK "numbers"
that you think are correct won't mean a hill of beans if the colorants of
the actual inks used in the correct profile or on press are substantially
different than what you assumed.
Let's look at this again:
The two options as I understand them are:
DAN'S OPTION:
1) Evaluate the profile that the printer recommends
and observe that
it's a disaster.
2) Find out what Custom CMYK dot gain value matches
it.
3) Write an Action that separates a file twice and
puts the Custom
CMYK version on top in Luminosity mode.
4) Drink beer.
5) Find out that the hues of the inks are substantially
different and either...
6) Spend several hours color-correcting the
<blank> out of the images or..
7) If it already went to press...
8) Cry in your beer.
TERRY'S OPTION:
1) Evaluate the profile that the printer recommends
and observe that
it's a disaster.
2) Download the relevant ISO standard data set.
3) Use it to make a new profile with the K generation
and GCR that suits your fancy.
4) Have a glass of champagne while congratulating
yourself on how easy that was and how many hours that saved you. EVEN IF
you had to purchase a profiling package (or a profile from a service), you,
at worst, broke even and will have the ability to create more such high
quality profiles in the future.
It does not seem that we are in agreement as to which
of these
methods is more efficient.
Forgetting my (and yours) tongue-in-cheek assessment of
the more efficient solution, I deemed my method more efficient before you
came to the discussion. I was evaluating the efficiency of my solution
(create a new profile) with what Stephen had suggested as an option
(layering/blending and what not). Turns out yours is at least no more
efficient than mine (really, how many trips back into Custom CMYK would it
take before you stumbled on the "correct" dot gain of the
profile, never mind that the ink colorants would be different?)
That disagreement does not prevent me from
expressing appreciation for your generosity in
providing the profile
to Marco, and thereby apparently solving his problem
for the time
being, until he faces up to a new printer or finds out
that this
printer's performance isn't quite what he was led to
believe.
The printer's actual performance in this case is
immaterial. As far as we know, the profile suggested by the printer was
correct but Marco felt they printed "too dark" prompting him to
suggest that perhaps a new profile but based on the same data set as the
original profile but with less aggressive black generation was in order to
*perhaps* correct for a printing condition where the black wasn't
controlled as well as it should be. Here's the questions I would have: They
printed "too dark" compared to what? * To your display? Is it
calibrated and profiled?
* To a proof that YOU printed? Is your system properly
profiled to simulate the proper printing condition?
* To the PRINTER'S proof? In other words, the proof
THEY provided looked fine but it printed to dark? Then you have every right
to ask for a re-print..end of story. If, on the other hand, they were too
dark on the proof and they MATCHED the proof, then, well, that's kind of
your mistake. If you didn't even GET a contract proof from them to sign off
on, then, again, that's the customer's mistake.
I should think that experience would have by now made
clear that it
is *very* relevant to you. Remember, this debate has
been going on
for a very long time. It was eleven years ago that the
Photoshop
engineers announced that Custom CMYK was obsolete, and
I said that if
they were serious about it they had to incorporate at
least as much
capability in whatever the next step was, because
nobody interested
in quality was going to downgrade to a system of
uneditable profiles.
I do not fault the Photoshop engineers for thinking
that they are
right at the time, but there comes a point when
they--and you--have
to admit that they have been proven wrong. That point
should not
require eleven years to make.
People interested in *quality* would've invested in a
decent stand-alone profiling package 11 years go. I would suggest if you
want to be professional in this business and work in the CMYK print world,
you should, at the very least, invest in a decent display
calibration/profiling package and at least CONSIDER investing in a print
profiling package. Doesn't even require that you get a spectro or anything
if all you want to do is re-engineer existing profiles with different black
generation an so on. You can either do that or employ any number of custom
profiling services. They're not that hard to find and the fee is very
reasonable and may make sense vs. purchasing the software yourself.
Helping out Marco is most commendable, but what
happens when he
learns that this printer doesn't really print to any
known standard,
and your profile, though good in theory, is
significantly off in
practice?
If the printer SAYS he prints to a standard but DOESN'T
in fact print to that standard, it's time to go shopping.
Having to have an on-call color management consultant
when
the solutions already exist for free in Photoshop
doesn't really seem
practicable.
But this "free" (it's not free really)
solution doesn't seem to work very well does it? And, based on Adobe's
general attitude towards that functionality, it doesn't appear that it's
going to get any better any time soon.
The way I see it, one can bitch and moan from the past
11 years to the foreseeable future and see if the situation changes OR, if
your life/ job depends on CMYK separations of the highest quality, bite the
bullet and either invest in what it takes to provide that for your business
or find someone or a service who will.
When one method is being belittled and declared
obsolete for a period
of eleven years, yet those interested in quality won't
abandon it,
perhaps some positions should be reconsidered.
You REALLY THINK that Custom CMYK provides HIGHER
quality profiles/separations compared to any number of front-line profiling
applications that are out there? Really? That's interesting.
And as for being
material to you, all you have to do is ask yourself
why those who can
produce better images have prospered over these eleven
years while
color management consultants have by and large
starved. Selling color
management services to those who don't care about
quality is a flawed
business model. Trying to sell methods that produce
third-rate color
to those who do is equally flawed IMHO.
I'm not sure you understand what the vast majority of
color management consultants actually DO for a living (at least the ones I
know about and excluding some industry pundits that last had a clue about
color management from some time in the late 90s). I'd certainly like to
think that I get called by folks that DO CARE about quality but currently
don't have the means to do that with their in-house talent so they need to
look outside their organization. At least 90% of *my* consulting business
is spent with printers interested in achieve higher quality results by
means of better process control, standardization, etc. Their aim is to NOT
be like those printers you constantly rail against as being clueless about
printing and prepress. So rather than constantly belittle color management
consultants (is that code for "calibrationist"?), maybe you
should be ENCOURAGING our efforts because there's a large contingent of us
that are working hard to solve the very issues that you speak of. If the
goal is making the offset (and other) printing processes more predictable,
then you might find us a useful ally to you and those on this list.
Regards,
Terry Wyse
______________________________
Terence Wyse, WyseConsul
Color Management Consulting
G7 Certified Expert
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: J Walton
Mon Jan 19, 2009 5:08 am (PST)
Terry,
It seems like you are missing something. Or perhaps I
am missing something obvious?
I agree that matching the *color* of an ICC profile
using Custom CMYK is a waste of time, a fact with which I'm sure Dan will
agree. There's a reason why nobody learned to adjust the LAB values in
Custom CMYK. But why are the primary and secondary inks of any issue if all
you are replacing is the black? AFAIK, Dan is basically adjusting Custom
CMYK to get the DOT GAIN the same, selects the black generation he wants,
and then he puts the Custom CMYK separation on top of the the profile
separation and sets it to luminosity mode.
It seems like you pretty much could replace the black
generation in ANY existing 4/c profile using that method. And it wouldn't
take very long, certainly it would happen quicker than you could download a
data set off the interweb and create a custom profile.
And when you go through those hoops enough times you
wonder - why can't Adobe just give me an automated method of doing this
using a familiar interface, say, Custom CMYK?
The whole shouting and pointing fingers and accusing
people of not being professional because they didn't buy something
superfluous 11 years ago seems kind of silly when you take a step back. The
fact is Adobe should be providing such a basic tool - it's like Nike
selling shoes without laces. You can get mad at people who refuse to buy a
profiling package that costs MORE than Photoshop to do something that
Photoshop has historically done, but wouldn't it be more constructive to
direct that angst where it belongs?
J Walton
___________________________________________________________________________
.
Re: [colortheory] Eurostandard inks – anyone?
Posted by: Terence Wyse
Mon Jan 19, 2009 9:40 am (PST)
On Jan 19, 2009, at 2:40 AM, J Walton wrote:
Terry,
It seems like you are missing something. Or perhaps I
am missing
something obvious?
I agree that matching the *color* of an ICC profile
using Custom CMYK
is a waste of time, a fact with which I'm sure Dan
will agree. There's
a reason why nobody learned to adjust the LAB values
in Custom CMYK.
But why are the primary and secondary inks of any
issue if all you are
replacing is the black? AFAIK, Dan is basically
adjusting Custom CMYK
to get the DOT GAIN the same, selects the black
generation he wants,
and then he puts the Custom CMYK separation on top of
the the profile
separation and sets it to luminosity mode.
My assumption was to re-separate from the original to
get the right separation from the get-go rather than try to repair the
incorrect separation. In my mind, that would be faster/easier and give
superior results compared to other methods. But we can agree to disagree on
that.
It seems like you pretty much could replace the black
generation in
ANY existing 4/c profile using that method. And it
wouldn't take very
long, certainly it would happen quicker than you could
download a data
set off the interweb and create a custom profile.
I think it would be incorrect to simply consider the
black channel as something independent of the CMY channels. If you start
with the wrongs assumptions about the CMY ink colorants, the black
generation/GCR will certainly be affected.
As far as quicker than downloading a data set and
creating a new profile, maybe on a single image but not if you several or
perhaps tens-hundreds of images to correct. You only have to download and
build a new profile ONCE.
And when you go through those hoops enough times you
wonder - why
can't Adobe just give me an automated method of doing
this using a
familiar interface, say, Custom CMYK?
I don't know, ask them. They haven't done much with it
in 11 years so I guess I would consider moving on to some other solution
that offers the same functionality plus much more.
The whole shouting and pointing fingers and accusing
people of not
being professional because they didn't buy something
superfluous 11
years ago seems kind of silly when you take a step
back.
If you're talking to me, I never accused anyone of
being unprofessional, I merely pointed out that I (and MANY others) don't
consider Custom CMYK a professional profile creating or editing tool.
Besides functionality, there's several ways this can be demonstrated
technically.
I'm NOT saying that Photoshop users all need to rush
out and purchase a profiling package. All I'm saying is that if your
separation needs go beyond the many standard profiles (GRACoL, SWOP, all
the flavors of ISO profiles that are freely available) and you either need
to create your own from custom measurement data or you need to customize
your separation parameters on a regular basis, then certainly one ought to
consider other options besides looking to Adobe for these functions.
The fact is
Adobe should be providing such a basic tool - it's
like Nike selling
shoes without laces.
Well, I think a better analogy would be...does Nike owe
you a new pair of shoes everytime you outgrow the pair you have?
You can get mad at people who refuse to buy a
profiling package that costs MORE than Photoshop to do
something that
Photoshop has historically done, but wouldn't it be
more constructive
to direct that angst where it belongs?
You must be misinterpreting what I wrote. I'm not MAD
at anybody for not buying a profiling package. I could care less. I simply
get tired of people complaining about Custom CMYK when there are plenty of
other BETTER solutions out there.
As to the cost, think of how many times you've upgraded
Photoshop in the last 11 years....if you'd purchased a good profiling
package 11 years ago, you would've had to upgrade MAYBE 2-3 times in that
same time span. While the initial entry fee for a profiling package can be
steep, the upgrades haven't been all that expensive, perhaps 1/4 of the
original cost of the package. I'm not saying it's INEXPENSIVE to get a
profiling application, but it's probably not as bad as some people think
and the upgrade cycle is positively glacial compared to Adobe's.
I'll admit, I get just a LITTLE rankled at the color
management consultant ("calibrationist"?) bashing whenever
there's an opportunity. I think much of this bashing came as a result of
certain vocal pundits but I assure you they don't speak for the rest of us
who do this for a living. The vast majority of us have no axe to grind with
anybody and come from a similar background as many of you on this list.
Myself, I came from a traditional prepress/printing background starting in
the late 70s and worked in production into the 90s before I went out on my
own. I did paste-up, stripping, camera separations, drum scanner operator
for about 10 years and on into the "desktop" production era.
Point is, I looked at color management as simply a more sophisticated way
to handle color issues that we've always had to deal with in the past. It's
opened my eyes to many possibilities compared to the way I was thinking
back in the 90s. So I guess I've embraced the technology for what
possibilities it may bring rather than trying to fight it and hang on to
outmoded ways of thinking.
Regards,
Terry Wyse
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: "Ron Kelly"
Mon Jan 19, 2009 12:44 pm (PST)
Terry:
There's some validity to your point - we shouldn't get
stuck in our ways. I don't think Dan Margulis is someone who is stuck,
though. I can't think of anyone who better typifies the attitude of
"constant improvement."
Supporting CMYK is built in to Photoshop; it's part of
the core. If Adobe pulled that functionality from the program, a lot of us
would be looking for a replacement. We prepare files for press and that's
at the core of our business.
Photoshop does not, for example, have sophisticated
word processing support. No paragraph styles, and that's fine. I'll use
word processing programs and page layout programs for that purpose.
The support to one of it's core functions, CMYK, black
generation and modification - is weak. It has beenweak for years, and many
on this list, especially Dan have said so.
Why doesn't it offer better support? Why must you go to
another program to try and cope in this aspect? I think that's the question
here.
I would agree that it should be central to Photoshop.
Why have we got umpteen different ways to address the color and range of a
file but only one kludgey method for controlling/modifying black
generation?
It would be interesting to know who insists it be so,
and why, wouldn't it?
Ron Kelly
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: "Henry"
Mon Jan 19, 2009 3:13 pm (PST)
My observation is that Photoshop has been continually
drifting further away from pre-press needs and has evolved to become a
product that is aimed more at photographers than any other group. I credit
this to nothing more than a business decision. But is does seem a shame
that Adobe's original partner remains at the dance while they have left
with another. It's bad form.
My experience in printing began with camera work
shooting mechanicals, making separations, stripping, platemaking - I bought
Photoshop and brought it to that party. I was in that group of first users
sending files to imagesetters. I knew very few photographers that even knew
the program existed. I had already gained a personal understanding of dot
gain way before Photoshop came along. So, I played with the Lab values in
Custom CMYK since the very beginning and from my first disappointing
results I asked myself, "what in the world am I doing wrong?". I
could get the Custom CMYK to do what I wanted more or less with regard to
black generation but color would not behave. Then I began to ask why myself
in the world would Adobe make such a frustrating tool. Why shouldn't the
Lab values make it work? I could never figure it out, but I *could* get the
black plate I wanted. To me, Custom CMYK seems sort of like a cruel joke.
Since then I've asked myself, "why can't it work
correctly in the simple way that the tool itself implies that it
should?". Is there a way that it could be made to work properly?
Is there an official Adobe reason as to why not?
Henry Davis
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: Michael Jahn
Tue Jan 20, 2009 4:44 am (PST)
Hi Ron,
And MANY of us in the printing vertical would REJOICE
and be THRILLED that Photoshop suddenly stopping letting users make CMYK
!!!!!
Do you select screen angles ? Do you know what trapping
or ink squence ? I mean, why on earth would i want CMYK make for one print
condition to print a job on an HP indigo, a Xerox iGen or a Kadak Nexpress
?
I LOVE the face that Lightroom does not offer CMYK !
there, I said it, and I mean it. CMYK does not belong
in Photoshop. Many many many people have ALWAYS felt that way. 10 years
forward, people will think the notion that you need to make CMYK before you
send it to print is silly.
Betcha 5 dollah !!!
-- I will even send that 5 dollars to dan so he can
hold the bet, and
will not be surprised if this is in 5 years.
--
Michael Jahn
Jahn & Associates
___________________________________________________________________________
.
Re: Eurostandard inks ? anyone?
Posted by: "Lee Clawson"
Tue Jan 20, 2009 9:44 am (PST)
on 1/19/09 9:38 PM, Michael Jahn wrote:
there, I said it, and I mean it. CMYK does not belong
in Photoshop.
Many many many people have ALWAYS felt that way. 10
years forward,
people will think the notion that you need to make
CMYK before you
send it to print is silly.
Well there was a long time when the designers were
asked to send Quark, PageMaker and InDesign files
"ready-to-print". If not, the printers charged extra for the
conversions. Then proofed your files and still had a hard time matching
their own proofs.
And when companies could ask a photographer for digital
images they asked for CMYK for the same reasons.
But I agree it would make life much easier for everyone
if the submission specs changed to deliver RGB. Except for one question;
which RGB ??
Lee Clawson
2/\V/\7 Studio
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: "RJay Hansen"
Tue Jan 20, 2009 11:29 am (PST)
On Mon, Jan 19, 2009 at 7:38 PM, Michael Jahn wrote:
10 years forward,
people will think the notion that you need to make
CMYK before you
send it to print is silly.
People have been saying this for the past 10 years.
Just because *everyone* doesn't need CMYK doesn't mean those of us who do
should have to do without.
There are very valid reasons for making CMYK available
to users. Dan has pointed out many where ultimately the file is converted
back to RGB.
RJay
___________________________________________________________________________
Quotes Without Comment (was: Eurostandard inks)
Posted by: Dan Margulis
Tue Jan 20, 2009 12:55 pm (PST)
I said it, and I mean it. CMYK does not belong in
Photoshop.
Many many many people have ALWAYS felt that way. 10
years forward,
people will think the notion that you need to make
CMYK before you
send it to print is silly.
--Michael Jahn, 2009
"CMYK [is] on the way out…. The winds of
digital change, blowing harder every day, will soon reach gale
force…. Signs, portents, and products abound: RGB will be the
communications format of the just- over-the-horizon future. Graphic arts
professionals, diversify."
--Lead article in Color Publishing magazine, 1993
"Six months into the future, we will all discover
that it’s smarter to scan and save all our images in RGB and only
convert them to CMYK at the RIP. Everyone will see the light one morning.
We’ll all wake up…"
--Brian Lawler, 1997
"It is obvious that storing and manipulating
images in CMYK is not the future. Color management systems will also
dictate this. CMYK will be an output choice done late in the
game."
--Bruce Fraser, 1997
"Standardized color management will remove the
mystery from color printing. It will cause printing costs (and prices) to
drop and the volume of jobs to increase dramatically. Color printing will
become a commodity, an output service comparable to that offered by
PostScript imaging services. Some firms will prosper, but many won’t
be able to make the change and will close their doors, or move into other
lines of work. Once we get through the period of flux, the industry will
stabilize once more…. Predictable color will be no more complicated
than typographic layout is now."
--George Alexander, 1998
Dan Margulis
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: "Ron Kelly"
Tue Jan 20, 2009 12:55 pm (PST)
Michael:
I guess your clients are primarily photographers who
want to avoid the complexities of CMYK and work only in RGB. They can go to
experts when necessary to get a printed job done well.
What about the experts? Why would you want them to stop
using Photoshop?
Ron Kelly
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: Michael Jahn
Tue Jan 20, 2009 12:55 pm (PST)
Hi Rjay,
I do not think people have been saying "We should
be exchangeing RGB" for the last 10 years -
I think Don Hutcheson was saying that sometime before
that !
And I never said people do not need CMYK.
I think - like we have moved from DCS eps files - most
will recognize that selecting the wrong CMYK, or inappropriate CMYK - too
far upstream - makes a bigger mess of things.
I recall a time in my career - as digital file exchange
began - when were were exchanging color separations as halftone screened
film, each with a specific dot shape, screen angle and rosette pattern - I
remember some insisting and sending digital versions of that. I also
remember people trying to scan that film to digital data files and imposing
them - it was a major issue when people could not agree on resolution, and
we had ads using different screen angles on the same flat.
Eventually, people all agreed that screening is late
binding - that is, we should do this in the rip.
I guess as we move toward shorter runs and more digital
printing, we need to recognize that these systems do not use SWOP inks or
really use ink densities in the same way as tradition offset printing.
It is that shift that will create pressure to NOT
create color separations (or CMYK files) too soon.
But hey, people thought I was crazy when i suggested
PDF might be a great prepress format, many who worked for Adobe.
Not to say I am NOT crazy (many will attest to that)
Anyway, this is a great debate and no one is ever
really right or wrong - people always do what they are comfortable with,
and continue until the see an advantage (or price break)
--
Michael Jahn
___________________________________________________________________________
.
Re: [colortheory] Eurostandard inks – anyone?
Posted by: "RJay Hansen"
Tue Jan 20, 2009 2:34 pm (PST)
Hi Michael,
Well I almost said the last 15 years instead of 10,
which as Dan's last message showed is closer to the mark.
I'm a prepress guy myself, so I can probably pretty
confidently state that I've seen most of the ways "average" users
can submit bad cmyk files. Our customers cover a wide range of ability and
Photoshop knowledge. From those who you wonder if they have to get someone
else to turn their computer on to those who know how to manipulate
individual channels so they can for instance, create an image in Photoshop
and confine black elements to the black channel only.
So I'm not for a "lowest common denominator"
approach with program like Photoshop that's aimed at professionals. Anyway,
I believe there's already a version of Photoshop that doesn't do CMYK. It's
called Elements.
I've got to comment another idea commonly trotted out
by RGB workflow proponents, that being that once you've separated an image,
you've hosed it for any other purpose. I would wager that the vast majority
of 4-color print images are used one time. Separate it for that one time
use and you've got no problem. And if it's going to be used multiple times
in different scenarios, it's really easy to do File-->Save As...
RJay
___________________________________________________________________________
Re: Quotes Without Comment (was: Eurostandard inks)
Posted by: "Jim Donovan"
Tue Jan 20, 2009 2:34 pm (PST)
Thanx for all the comments Dan!!!!! I saw no need to
repost any of them. Will the rest of the "many,many,many people"
please chime in with the argument against CMYK being part of Photoshop.
Would love to hear from all or any of the "many,many,many",surely
there are many,many,many of them on this list,being the leading color
theory group on the planet......or maybe not.Please state your case...or
not. Jim Donovan
___________________________________________________________________________
Re: Eurostandard inks - anyone?
Posted by: "Greg Welch"
Tue Jan 20, 2009 9:44 am (PST)
Dear Michael,
The only reason I use photoshop is because it can
manipulate cmyk files. There are many other programs much better than PS at
RGB work. QFX comes to mind.
Understood, that you as a specialist would like to
maintain the color control when the public ventures into cmyk. Before
photshop and icc life in color was more expensive perhaps better for some
$$ but not possible for others who now do contribute to images printed.
Let's go forward, RGB is a great workflow for most but not for all.
So I would vote for more CMYK tools in PS :-) After
having taken my Kodachrome away please don't take my CMYK away too, and
leave me with icc only solutions. Oh that's an Ink (horse) of a different
color, issue. Printing can be a craft as well as a commodity. Just a
different viewpoint, thank you!
Greg Welch
___________________________________________________________________________
A remark about the CMYK debate (long)...
Posted by: "Marco Olivotto - LoL Productions
snc"
Tue Jan 20, 2009 9:44 am (PST)
Hi all,
only today I've had time to go through the most recent
digests, and I was surprised to find a rather heated debate based on my
previous post about Eurostandard inks.
I think I should further clarify the reason why I
originally wrote it; and, of course, I am grateful to anyone who threw in
their opinions because this helps me understand the subject better. Dan
makes a good point when he writes that I am not CMYK proficient. No dobut
in this. But, alas, I have to make it with my own means for a number of
reason. Why?
Top line: this is Italy. Maybe not third world, but
over 15 years of activity (CD and DVD production, sometimes occasionally
books) I haven't been able to find 1) someone who can actually guarantee a
serious repeatability of results in print under similar conditions and, 2)
someone who can actually clarify which workflow is the best when you have
to go to press under such conditions. The problem I have to face, believe
it or not, is that whenever you speak to a printer, they'll go –
"give us RGB, we'll match your monitor". Period. While I am not
*that* proficient, I've seen enough to know that this is bulls**t.
Professional pre-press is dead, at these latitudes; nobody does that
anymore. My last encounter with a proper pre-press person dates back to 10
years ago – and this is the sad truth. I may sound bold, but I am
quite sure that if I spend ONE WEEK studying Custom CMYK in depth by
trial-and-error and with the kind help of competent people as those on this
list, I'll know more than the average operator I have to speak to every
other day.
If you don't believe me, one funny
"professional" suggestion I received exactly one week ago was
this: "never separate b&w in 4/c – forget Heavy GCR, it just
won't work. Make a duotone, instead, with K and a Pantone Grey, that's a
lot better." Granted, as long as you have a 5-colour printer and as
long as you can master the duotone curves properly. It seems to me we are
trying to throw away a problem by creating two more. Or not? And this,
notice, was suggested for a book which has TWO b&w images among FIFTY
others, regularly in colour. No mention of the major costs of printing the
fifth colour, of course.
So, about a year ago, at a point we decided it was just
about time to give it a try on our own. And we started studying CMYK which
is, admittedly, conceptually a bit more difficult than RGB.
With one particular printer we are *forced* to use, we
discovered that the results were consistently darker than we expected:
certainly darker than our monitors (yes, calibrated) and certainly darker
than *their* proofs (we don't proof in-house).
Rough line of reasoning: the proofs look like a decent
match of what we see on screen, to us. The printed result doesn't. So the
obvious problem is proof-to-print matching. My personal theory about the
nature of this problem, having examined about a hundred printed results and
the originals, is that something goes wrong in their press when black ink
goes above 70% or so (this is what I seem to see, but don't ask me why this
should be!). Lighter areas are ok. I asked, and got no useful reply –
or I couldn't understand it (it was along the line of "bysinchronous
bitflop", yes). Since the profile they recommend is ISO Coated v2
300%, which has a black ink limit of 95% *and* a serious black generation
(you can't call it a Light GCR at all, probably because it uses black to
compensate the relatively low total ink limit, compared to, say, FOGRA39),
this looked like a recipe for disaster in the shadows. Actually, it caused
a number of serious disasters in the shadows, to us! Moving to FOGRA27 or
FOGRA39 wasn't a real option because the total ink limit is higher than
300%, which is what they ask for.
Therefore, Custom CMYK seemed a viable option. The
problem, to us, was this: almost all the discussion about it in PP5E is
based on SWOP inks, which are different than ours. Between a European and a
Japanese alternative, we chose the first – only to discover that the
colours were absolutely different than we obtained with the ISO profiles
– and *now* I know why. The way out of this is obviously going to
Luminosity mode, as discussed; but then luminosity wouldn't match. This
prompted the original question I posted.
The custom profile so kindly provided by Terry is a
reasonable alternative; and the final explanation about dot gain provided
by Dan is as much reasonable – actually, I prefer it in principle
because it allows me more room to roam, if needed, and because (to me) it
sheds some light on a complex subject. I originally completely overlooked
the dot gain issue for a laughable reason: I accepted the standard proposed
by Photoshop, which is 9% for Eurostandard inks. This is obviously flawed
(and, again, *now* I know); I admit my first thought was: "how can
this be? 20% for SWOP and 9% for other inks?" Because this would mean
that the average printing conditions in Europe are much better than in the
States – (choke!). It was only when Dan pointed out that he wouldn't
comment about the 9% that I thought this might be the culprit, and in fact
it is. With, may I add, the fact that the dot gain of the Cyan ink is 4%
higher than any other by default – which doesn't make sense (and this
is discussed in the book, indeed). After a couple of hours of funny games
with the dot gain curves I was able to reach a perfectly acceptable
compromise. Bottom line: I have *two* solutions now. All this is something
I am learning by experiment – which may be good or not, but this is
the actual situation I live in. And exactly because I am not proficient at
CMYK, and have no direct experience in the pressroom, my learning curve is
steep. I can only thank anyone who suggested a solution, because each
solution actually shows a different side of the problem; and while there
are different sides, there is room for improvement, which is what I am
after.
My final thought is that, whether one is proficient or
not in CMYK, when it's time to go to press it is necessary to at least
understand the mechanism of the thing... otherwise sparks will fly, and
soon. Now – buying a profile editor is an option – but at least
having a go at what Photoshop provides is also some kind of option, in my
opinion. I will gladly run the risk to ruin my next job completely, if this
gives me a chance to grasp the underlying mechanism a bit better, because
ruining one job in order to have ten good in the future is a fair trade-off
in my mind. I am used to have dozens so- so, after all, so what? Finally,
as you certainly can tell from what I write, I am a strong believer in the
education based on trial-and-error, and this case is no exception to my
eyes.
Thanks again to everyone!
All the best,
M.
---
Marco Olivotto
LoL Productions snc
1/A Via per Sasso
38060 - NOGAREDO (TN)
ITALY
___________________________________________________________________________
Re: A remark about the CMYK debate (long)...
Posted by: "Ron Kelly"
Tue Jan 20, 2009 12:55 pm (PST)
Marco:
My sympathies for your struggle, and a salute to your
persistence.
I would make one comment on your venture into preparing
press ready files: start proofing in-house.
I've heard many, including Mr. Margulis himself, say
that they don't need a proof anymore, that the screen is sufficient. I
can't agree with them, and I've tried plenty hard to rely on my screen. If
it was all I needed it would be a much appreciated savings in time and
material.
I have calibrated screens and proofers, but I find out
things from the printed proof that the screen doesn't tell me. Things like
this:
- reflected-light media are fundamentally different
than transmitted light media. The subtleties of color just don't show up on
screen like they do in print, the difference in dynamic range is too great.
-color saturation is another example; on-screen
sometimes looks fine, or maybe it looks ridiculously dull but if I print it
I'm not guessing if it's good or not - I know.
-fine details like sharpening are represented by
facsimile on-screen, but the printed result shows you *what it is.*. Unless
you have a lifetime experience preparing files, and much more than my 15
years or so, you can't judge from as screen as accurately as you can from a
print.
-I'm sure I could go on, but the last point I'll make
is that if your final result is a printed piece, the most representative
thing you can do (which is not to say that it's perfect) is to view it as
such.
It is possible to do without, and some do but I can
unequivocally say that I couldn't do as good a job without my own proof. If
you proof your own files, you'll save yourself time and money when dealing
with a printer, even if they also proof the file (they should) as you will
lessen the number of revisions to your work before it's sent to plates.
There are many solutions available at reasonable cost,
and if you have even an intermittent on-going need to go to press you would
do well to develop this capability.
Ron Kelly
___________________________________________________________________________
A necessary clarification (was: Eurostandard Inks)
Posted by: "Marco Olivotto"
Tue Jan 20, 2009 2:34 pm (PST)
I need to clarify that when I was writing FOGRA27 or
FOGRA39 I was actually shortening "Coated FOGRAXX (ISO 12647-2:
2004)", that is, Adobe's profiles based on the relevant data set. My
fault, indeed – but that's what I meant.
Sorry if someone was confused by these improper terms.
M.
---
Marco Olivotto
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: Dan Margulis
Wed Jan 21, 2009 8:34 am (PST)
Terry writes,
No, you CAN'T since you yourself stated you can't
match the color of
the primary and secondary inks of the profile (ISO
Coated) using
Custom CMYK. Finding and adjusting the dot gain is
only half the
story. The other half is, uh, color.
That's right, and if you understood the procedure you
were criticizing, you would know that the color is not determined by Custom
CMYK, but by the existing profile. The only thing that has changed is the
CMY to K relation.
Forgetting my (and yours) tongue-in-cheek assessment
of the more
efficient solution, I deemed my method more efficient
before you came
to the discussion. I was evaluating the efficiency of
my solution
(create a new profile) with what Stephen had suggested
as an option
(layering/blending and what not). Turns out yours is
at least no more
efficient than mine
Yes, on the assumption that your profile is good, the
actual workflow is fine--if we forget about the need to find a color
management consultant first, or spend $2,500 on a profiling package and
learn it.
(really, how many trips back into Custom CMYK
would it take before you stumbled on the
"correct" dot gain of the
profile, never mind that the ink colorants would be
different?)
As noted, the ink colorants would be the same. But in
answer to your question, it requires zero trips into Custom CMYK, one
merely need open any file with a properly embedded tag. The process takes
about a minute. Anyone claiming to be competent in color management should
be able to do it.
The printer's actual performance in this case is
immaterial. As far as
we know, the profile suggested by the printer was
correct but Marco
felt they printed "too dark" prompting him
to suggest that perhaps a
new profile but based on the same data set as the
original profile but
with less aggressive black generation was in order to
*perhaps*
correct for a printing condition where the black
wasn't controlled as
well as it should be. Here's the questions I would
have:
They printed "too dark" compared to what?
* To your display? Is it calibrated and profiled?
* To a proof that YOU printed? Is your system properly
profiled to
simulate the proper printing condition?
* To the PRINTER'S proof? In other words, the proof
THEY provided
looked fine but it printed to dark? Then you have
every right to ask
for a re-print..end of story. If, on the other hand,
they were too
dark on the proof and they MATCHED the proof, then,
well, that's kind
of your mistake. If you didn't even GET a contract
proof from them to
sign off on, then, again, that's the customer's
mistake.
These are interesting questions and if time permits
they ought to be investigated. However, it would be difficult to conceive
of any circumstances where Marco's strategy would not help. Given a printer
of unknown quality, supplying a light-GCR separation is definitely more
likely to get the desired result--as both Marco and Jeremy have recently
found out the hard way.
People interested in *quality* would've invested in a
decent stand-
alone profiling package 11 years go.
They did. Photoshop. It had to be at least decent,
because otherwise there could not have been quality work
done--substantially ALL work went through Custom CMYK before 1998 and a
heavy majority was done that way at least until Photoshop 6. So, unless you
are asserting that it was not possible to have done quality sep work in
Photoshop in the 20th century, Custom CMYK has to be fairly good.
Faced with the undeniable reality that much good work
was being done with Custom CMYK, the first half of the argument in favor of
change was and is as follows.
1) In order to be able to do in the future what you
already can do faster for free, you will need to pay $2,500 or hire a color
management consultant.
2) If you do not avail yourself of this generous offer,
you are not "interested in quality." you are
"brain-dead," "extremely dense," "clueless about
color", and "stuck in the stone age of printing."
If the printer SAYS he prints to a standard but
DOESN'T in fact print
to that standard, it's time to go shopping.
Right. This is the second half of the new received
wisdom.
1) Commercial printers, who historically have known
nothing about prepress, are henceforth responsible for being prepress
experts.
2) Commercial printers, who historically have been all
over the quality map, will henceforth all be precision operations.
3) If your job is being printed by any firm that does
not comply with both 1 and 2 above, it is your fault because you should
have found one of the handful of printers that does.
4) If your job is screwed up because the printer *says*
he complies but doesn’t, not to worry, because it is the printer's
fault, not yours, and when you explain it to your clients they will
understand.
But this "free" (it's not free really)
solution doesn't seem to work
very well does it? And, based on Adobe's general
attitude towards that
functionality, it doesn't appear that it's going to
get any better any
time soon.
If that's the case, then the choices are the same four
as in 1998: hire a color management consultant; spend $2,500; downgrade to
a system of uneditable profiles whose current quality level is considerably
to the south of mediocre; or use something that's decent, fast, and free.
It wasn't a hard choice then and it isn't a hard choice now.
You REALLY THINK that Custom CMYK provides HIGHER
quality profiles/
separations compared to any number of front-line
profiling
applications that are out there? Really? That's
interesting.
You REALLY THINK that George W. Bush was our greatest
president? More competent than Abraham Lincoln, Franklin Roosevelt, or any
number of other frontliners? Really? That's interesting.
Profiling packages are a lot like cameras. The quality
of the instrument is usually secondary to that of the person operating it.
A reasonably motivated Custom CMYK user, even someone of Marco's limited
experience, say, would be wildly unlikely to produce anything with dot gain
as far out of whack as US Sheetfed Coated v2. It would be virtually
impossible for him to produce anything as bad as Coated FOGRA27. And
surely, he would not think that sheetfed uncoated and web uncoated are the
same thing, and issue the same profile for each.
At least 90% of *my*
consulting business is spent with printers interested
in achieve
higher quality results by means of better process
control,
standardization, etc. Their aim is to NOT be like
those printers you
constantly rail against as being clueless about
printing and prepress.
I don't rail against them for being clueless about
prepress because nearly all printers have *always* been clueless about it
and there isn't any reason to suppose that things will change. I rail
against authors and color management consultants who create the impression
that one should slavishly defer to the printer's views on how files should
be prepared.
I do not recall railing against printers as being
clueless about printing. I do point out that the majority of commercial
printers cannot be relied on for excellent process control. I do not
criticize this, because process control is expensive, and "good
enough" color is, in fact, good enough for many clients, though
presumably not for members of this list. I do rail against color management
consultants and others who insist that all printers should be treated as
though they had excellent process control, by giving them files that will
fail in its absence.
Any success you may have in guiding printers to better
process control is, of course, much to be desired.
So rather than constantly belittle color management
consultants (is
that code for "calibrationist"?), maybe you
should be ENCOURAGING our
efforts because there's a large contingent of us that
are working hard
to solve the very issues that you speak of.
I encourage all efforts that result in higher print
quality, such as assisting printers to do their job more consistently and
to communicate better with their clients, or correctly explaining to users
what their options are. I discourage all efforts that diminish print
quality, such as advising users to rely on what printers say without
verifying that they know what they're talking about, or advising users to
employ profiles that will produce bad results when the printing is not of
the highest quality. I deprecate all efforts by color management
consultants to enrich themselves at the expense of final quality, such as
by enablng printers to secretly overrule their client's GCR decisions in
order to save a few pennies on ink, or by trying to restrict users to
third-rate profiles unless they hire a consultant.
If the goal is making the
offset (and other) printing processes more
predictable, then you might
find us a useful ally to you and those on this list.
To some extent you have been. A randomly selected
printer today probably won't be very good, but will be better than was the
case ten years ago, and some of the credit for that must go to color
management consultants. As far as providing profiles for us, we would
certainly like you to be an ally, but your value is heavily compromised by
our inability to adjust your work. Anyone is advising printers to overrule
our black generation decisions and take wild guesses as to what our intent
was, that person or group is an enemy and not an ally.
OTOH, high-end CMYK retouching has been a prosperous
field for a long time, and even in today's economy, those who have the
skill are doing better than almost anybody else in the field. This is in
sharp contrast to the color management field, which was a train wreck
before the economic collapse. After eleven years of tremendous growth in
all things color, and eleven years of propaganda about how this way is the
best way and everybody will be adopting it real soon now, the dominating
company in the field has downsized itself beyond recognition, and color
management consultants can't find work. So it might well be asked how
things would have played out differently if yawl had had *us* as an ally.
People who work with CMYK for a living are interested
in quality, not slogans. Discussion about what is or is not the wave of the
future, whether resistance is or is not futile, whether people who use
alternate methods are brain dead or merely mentally impaired, and the like
does not impress either us or our clients.
It's not as if it hasn't been made clear what would
have been needed to effect that alliance. I told the Photoshop engineers
prior to Photoshop 5 and in print after it was released. Chris Murphy and I
wrote about what it would take in 2001. The necessary progress has not been
forthcoming. But the basics are the same. Offer things that make us more
efficient and/or give us better results, and we'll march together. Try to
force downgrades on us, and you'll relive the last eleven years.
Dan Margulis
___________________________________________________________________________
.
Re: [colortheory] Eurostandard inks – anyone?
Posted by: "RJay Hansen"
Wed Jan 21, 2009 5:23 pm (PST)
On Wed, Jan 21, 2009 at 10:10 AM, Dan Margulis wrote:
might well be asked how things would have played out
differently if
yawl had had *us* as an ally.
Dan,
I believe the correct spelling is ya'll (as an
Oklahoman I can speak fairly authoritatively on that).
RJay
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: Terence Wyse
Wed Jan 21, 2009 5:23 pm (PST)
I think many if not most of your conclusions are dead
wrong, but, frankly, I'm done discussing it (as if it were ever a
discussion in the first place). No matter what valid points I might make,
you will always be right at least in your own mind so it's probably a waste
of my time to continue trying to find any common ground that we might agree
on.
You took what COULD HAVE BEEN a learning experience for
those on your list and simply turned it into another one of your color
management consultant-bashing sessions. Fine. I've got better things to do.
Regards,
Terry Wyse
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: J Walton
Wed Jan 21, 2009 7:00 pm (PST)
On Wed, Jan 21, 2009 at 4:41 PM, Terence Wyse wrote:
I think many if not most of your conclusions are dead
wrong, but,
frankly, I'm done discussing it (as if it were ever a
discussion in
the first place).
With which conclusions do you not agree? There's a lot
of people on this list whose name is NOT Dan Margulis (at least hundreds if
not thousands), so there's nothing preventing a discussion from taking
place. Just because you initially appeared to misunderstand what Dan was
saying doesn't mean the discussion can't continue. If we all agreed on
everything there would be no reason to have a discussion. It is only in our
DISAGREEMENTS that we find interest. And with your expertise a disagreement
could be very interesting.
No matter what valid points I might make, you will
always be right at least in your own mind so it's
probably a waste of
my time to continue trying to find any common ground
that we might
agree on.
I'm always interested in hearing valid points, and I'm
sure you brought some up. But they became hidden when you attacked the
suggestion that Adobe should give us the tools we need to produce the best
separations possible.
Here's places where we all pretty much agree.:
: : What Dan did was a hack that he shouldn't have to
do.
: : ICC profiles made with a professional application
will produce images of MUCH higher quality than is possible with Custom
CMYK.
Here's where we seemed to disagree:
: : One could adjust the black generation on any
profile by matching the dot gain in Custom CMYK and converting a copy on
top set to Luminosity.
Here's where I think we secretly agree:
: : Adobe should give us the ability to adjust dot gain
and black generation in profiles that contain measurement data.
You took what COULD HAVE BEEN a learning experience
for those on your
list and simply turned it into another one of your
color management
consultant-bashing sessions. Fine. I've got better
things to do.
That's unfortunate. Because if you have something to
offer it would be nice to hear it. But beyond that, it would be only fair
to admit that this could have been a learning experience for you as well.
That's the only time a disagreement becomes pointless and frustrating.
That's where we all need Howard's attitude (or at least a bit of it :-)
J Walton
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: "Ron Kelly"
Thu Jan 22, 2009 3:26 am (PST)
Mr. Walton:
Excellent post, with one exception: Custom CMYK *can*
produce excellent separations. It can also produce crap, but that much can
be said of any Photoshop tool.
If a separation came from a "professional"
ICC profile or through Custom CMYK and it's excellent, then . . . it's
excellent.
Also, those of us who use Custom CMYK and are
*professionals* might resent the implication that they are substandard.
Otherwise very well said.
Cheers,
Ron Kelly
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: Terence Wyse
Thu Jan 22, 2009 7:20 am (PST)
On Jan 22, 2009, at 12:43 AM, Ron Kelly wrote:
Also, those of us who use Custom CMYK and are
*professionals* might
resent the implication
that they are substandard. Otherwise very well said.
Ron, as ICC profiles go, they ARE sub-standard but
that's not to say that on any particular image it might not indeed do a
fine job, even a high-quality job but a *dedicated* ICC profiling
application will generally produce better results. Not ALL of them mind you
(there's plenty of crappy software out there for making ICC profiles also).
The fact is, you can't take ANY industry-standard set
of measurement data (ISO/FOGRA xx, GRACoL, SWOP, IFRA, SNAP, etc.) and use
Custom CMYK to generate a profile from it. I mean, that's simply BASIC and
what any profiling application needs to be able to do. Anything else it may
be able to do is superfluous in my book.
Here's a list of what I would consider any profiling
application worth it's salt should be capable of:
* Ability to import CGATS formatted measurement data or
extract this data from an existing profile by means of measurement data
"tags" in a profile or via an A2B (CMYK->LAB)
conversion/extraction.
* In the case of CMYK profiles, the ability to
customize black generation (GCR/UCR, etc.), black start point, black limit
and total ink limit.
* Create rendering intent tables for saturation,
perceptual and colorimetric.
* Ability to choose from different gamut mapping
algorithms or some other means of "tuning" the gray balance,
color and contrast rendering of a profile.
* Some means of editing the profile after the fact. By
editing, I do not mean altering the original separation parameters but a
means of correcting deficiencies in the tonal gradation, neutral balance
and color accuracy of the profile.
Of this list, I really only see ONE of these
capabilities in Custom CMYK.
I'll repeat myself, while Custom CMYK *may* be able to
produce high quality separations on occasion, it is not a high quality
profile creation tool.
Regards,
Terry Wyse
______________________________
Terence Wyse, WyseConsul
Color Management Consulting
G7 Certified Expert
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: "Ron Kelly"
Thu Jan 22, 2009 9:20 am (PST)
Terry:
I wasn't suggesting that Custom CMYK could create
profiles. What I said, and what I shall restate is that Custom CMYK can
allow skilled users to make excellent separations for a given press
condition.
Ron Kelly
___________________________________________________________________________
Re: Eurostandard inks - anyone?
Posted by: "Pylant, Brian"
Thu Jan 22, 2009 10:48 am (PST)
I wasn't suggesting that Custom CMYK could create
profiles. What I said, and
what I shall restate is that Custom CMYK can allow
skilled users to make
excellent separations for a given press condition.
Custom CMYK *can* create profiles. It's as easy as
selecting "Save CMYK" from the CMYK profile dropdown. However,
the original quote was:
it is not a high quality profile creation tool.
No one ever said it was, or claimed it to be. It's a
very limited feature at best. However, when you need control over black
generation, total ink, black ink, dot gain, etc. and only have Photoshop at
your disposal, then it's the only tool you can use.
Shame on Adobe for their neverending resistance to
provide something more useful, but in the meantime excellent results are
possible in the hands of a knowledgeable user. (Note that I am *not* saying
that in comparison to any particular profile, simply stating it as the
standalone fact that it is.)
As with many of the past arguments on this list, I get
the distinct impression that both sides are arguing over points that are
(at least) slightly different, but arguing as if they are the exactly the
same with rebuttals that do not directly address the points they are
attempting to rebut. (Be that intentional or via misunderstanding I can not
say.) Not apples vs oranges, but perhaps something more akin to granny
smith vs. golden delicious.
BRIAN PYLANT
Manager, Electronic Prepress
Disc Makers
7905 North Route 130
Pennsauken, NJ 08110
Toll free: 1-800-468-9353 ext. 5539
www.discmakers.com
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: J Walton
Thu Jan 22, 2009 10:48 am (PST)
On Thu, Jan 22, 2009 at 8:04 AM, Ron Kelly wrote::
I wasn't suggesting that Custom CMYK could create
profiles.
Actually, it can. You just don't have control over the
measurement data - only the back-end stuff.
What I said, and
what I shall restate is that Custom CMYK can allow
skilled users to make
excellent separations for a given press condition.
I think Custom CMYK can beat a custom profile the same
way a photoshop novice can make a better color correction than I can. It's
possible, but not likely, and if it does then it's an accident. Custom CMYK
in the hands of an expert will never consistently be as good as a profiling
package in the hands of the same expert. But if it is, it's an accident.
Of course, the best-case scenario is being able to have
a good Custom CMYK, one that can take in measurement data and make some
basic adjustments on the back-end of the profile. Then you have the
familiarity of Photoshop and the tools an expert will need to take the
measurement data and customize its use for a particular image.
J Walton
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: "Ron Kelly"
Thu Jan 22, 2009 2:28 pm (PST)
J:
I don't want to continue around this mulberry bush
indefinitely, but I will disagree with you one more time.
I can and do get excellent results, using Custom CMYK
in ways that I have learned from "Professional Photoshop" books
and classes from Dan. It's not an accident.
If a slide rule and a computer both come to the same
answer, it's quite possibly indicative that you can get the correct answer
more than one way. That doesn't mean that the slide rule user was lucky,
Ron Kelly
___________________________________________________________________________
Re: Eurostandard inks – anyone?
Posted by: "Ben Saltzman"
Thu Jan 22, 2009 2:28 pm (PST)
OK, I have read everyone of these threads from the
start. Was anything accomplished? I just see mad people.
For my use, it would be nice to know what default CMYK
you folks use and for what purpose. Yes, I know that there is no magic
bullet. But where do you start?
Please state the CMYK you use as a default and a two
sentence explanation. This would be helpful information. No fighting
please.
Love, Ben
___________________________________________________________________________
.
Re: Eurostandard inks - anyone?
Posted by: Michael Jahn
Thu Jan 22, 2009 2:49 pm (PST)
On Thu, Jan 22, 2009 at 10:44 AM, Pylant, Brian wrote:
"However, when you need control over black
generation, total ink, black
ink, dot gain, etc. and only have Photoshop at your
disposal, then it's the
only tool you can use."
and when all you have is a hammer, everything - even
screws - can worklike a nail.
which is what several of us try to point out.
Me, I use a screw gun.
then you added;
"Shame on Adobe for their never ending resistance
to provide
something more useful...."
Adobe will continue to enhance (add features), and
streamline (remove features) that will ensure a wider audience might be
convinced to invest in their applications. So - as very few people concern
themselves with things like color fidelity (or something exotic like swatch
matching) it follows that if they suddenly added a world class color
management toolset that this would NOT make sales leap.
Adding something like that Content-Aware Scaling DOES
make sales leap (in new sectors) http://www.youtube.com/watch?v=zJy9PNU12hc
So, adding pro level CMS no business case - it is that
simple really - so, no "shame on you" Adobe from the
stockholders, and all the Adobe partners like that Adobe did not squash
them like a bug.
--
Michael Jahn
___________________________________________________________________________
Re: [colortheory] Eurostandard inks – anyone?
Posted by: Terence Wyse
Thu Jan 22, 2009 2:49 pm (PST)
On Jan 22, 2009, at 11:04 AM, Ron Kelly wrote:
I wasn't suggesting that Custom CMYK could create
profiles.
Yes, but that's what Custom CMYK does, it makes
profiles.
What I said, and
what I shall restate is that Custom CMYK can allow
skilled users to
make excellent separations for a given press
condition.
I don't see how it can make profiles for a *given* or
*unique* press condition because there's no *working* mechanism for
reliably doing so (you can neither feed it actual press data nor input the
colorimetry in a reliable way). The only mechanism that appears to allow
you to enter parameters *unique* to a given press condition would be the
dot gain tables...but dot gain is really not the most reliable way to
describe the tone response of a press or any other printing device. Note
that I don't consider UCR/GCR, K generation, total ink, etc. parameters
that would describe a unique press condition. There's any NUMBER of GCR and
total ink settings that could be applied to a unique press condition so
there's nothing unique about it.
Regards,
Terry
______________________________
Terence Wyse, WyseConsul
Color Management Consulting
G7 Certified Expert
___________________________________________________________________________
Re: [colortheory] Re: Eurostandard inks – anyone?
Posted by: Terence Wyse
Thu Jan 22, 2009 3:25 pm (PST)
Of course, there is no ONE profile that's ideal, even
for a starting point or default. Depends where you're at geographically and
what type of printing is typical of your projects.
In the USA, you'd have to say...
GRACoL2006 Coated1 for sheetfed printing on coated
stock.
SWOP2006 Coated3 for web offset (publication) printing
on a quality coated stock.
For uncoated stocks (I'm thinking "offset"
stock), I think FOGRA 29 has a lot going for it. It is very close to the
"G7" tone response curve and the colorimetry seems about right.
Until we get a "G7" version of an uncoated stock, FOGRA 29 should
do fine.
Regards,
Terry
___________________________________________________________________________
Re: [colortheory] Re: Eurostandard inks – anyone?
Posted by: Michael Jahn
Thu Jan 22, 2009 3:25 pm (PST)
On Thu, Jan 22, 2009 at 11:54 AM, Ben Saltzman wrote:
Please state the CMYK you use as a default and a two
sentence explanation.
This would be helpful information. No fighting please.
Answering as IQCOlour;
We ignore any profiles embedded in an RGB or CMYK image
file (TIFF, JPEG, etc...) and we ignore any "output intent" that
is placed into an PDF (we process any and all images or objects no matter
what color space (or better said "color mode" (1, 8, 24, 32 bit
or DeviceN) and we pass this through a proprietary Color Space (ADT - AKA -
IJK) and create "IQ" Color separations using our LUT.
Answering as IoFlex
We can either honor / preserve or replace/ignore
profiles as required by YOUR workflow - but, in a perfect world, all RGB is
tagged Adobe 1998 walking in and "as you like it" on the way out
(normally, as WE like it is RGB PDF with the specific Output Intent profile
embeeded - PDF/X-4 "workflow"
in step with the "STOP SPDFS" and "Stop
the Transparencide" mantra -- http:
//michaelejahn.blogspot.com/2008/06/stop-spdfs-stop-transparencide.html
--
Michael Jahn
___________________________________________________________________________
.
Re: [colortheory] Re: Eurostandard inks – anyone?
Posted by: Henry Davis
Thu Jan 22, 2009 3:25 pm (PST)
On Jan 22, 2009, at 2:54 PM, Ben Saltzman wrote:
OK,
I have read everyone of these threads from the start.
Was anything accomplished? I just see mad people.
I observe that people are reasonably frustrated over
the kludge that is Custom CMYK. The question I asked earlier generated no
response. Perhaps there isn't a "nice" truthful response
available, but I am interested in speculative and even angry answers if
that will get the ball rolling at Adobe to make the tool be what the name
of the tool implies. I'll repeat again here:
"Since then I've asked myself, "why can't it
work correctly in the simple way that the tool itself implies that it
should?". Is there a way that it could be made to work properly? Is
there an official Adobe reason as to why not?"
Custom CMYK could be the tool that it is implied to be.
For me, the tool has been a lie from the beginning, and it has continued to
be a lie in spite of being called out. If that doesn't frustrate one to
some degree of anger, then one must have a very lenient attitude toward the
developers and their management.
Henry Davis
___________________________________________________________________________
Re: Eurostandard inks - anyone?
Posted by: Henry Davis
Thu Jan 22, 2009 5:04 pm (PST)
On Jan 22, 2009, at 5:39 PM, Michael Jahn wrote:
Adobe will continue to enhance (add features), and
streamline (remove
features) that will ensure a wider audience might be
convinced to
invest in their applications. So - as very few people
concern
themselves with things like color fidelity (or
something exotic like
swatch matching) it follows that if they suddenly
added a world class
color management toolset that this would NOT make
sales leap.
You don't want a tool box that has a complete set of
tools? Perhaps your degree of satisfaction would be the same if you had
only two selection tools? I notice that you didn't include the notion of
churn in your business model. What's wrong with the idea of discovering how
the market would react to the offering of a more complete toolbox?
Understand in advance that I will not respond to any
posts demanding a definition of "complete toolbox".
Henry Davis
___________________________________________________________________________
.
Re: Eurostandard inks - anyone?
Posted by: "Pylant, Brian"
Thu Jan 22, 2009 9:20 am (PST)
I believe the correct spelling is ya'll (as an
Oklahoman I can speak
fairly authoritatively on that).
Actually no, it would be y'all (since the apostrophe
replaces the "ou" in "you all").
:o)
BRIAN PYLANT
Manager, Electronic Prepress
Disc Makers
7905 North Route 130
Pennsauken, NJ 08110
___________________________________________________________________________
.
Re: Eurostandard inks - anyone?
Posted by: Michael Jahn
Fri Jan 23, 2009 4:10 am (PST)
Hi Henry,
Henry Davis teased;
You don't want a tool box that has a complete set of
tools?
LOL -- well, sure, and I can drive my helicopter
submarine automobile boat up to the jiffy mart to pick up a quart of
Lactose free milk.
What's wrong with the idea of
discovering how the market would react to the offering
of a more
complete toolbox?
today - we have all w need (apparently) --Adobe CS4,
Print Engine 2 and Acrobat 9.
TRUST me when I say this - do not hold your breath for
any new versions in the near future - lets hope they survive and do not get
bought by Microsoft or someting...
Understand in advance that I will not respond to any
posts demanding
a definition of "complete toolbox".
now THAT is funny - I guess that would be the set of
tools i would need to work on my helicopter submarine automobile boat.
--
Michael Jahn
___________________________________________________________________________
Re: Eurostandard inks - anyone?
Posted by: J Walton
Fri Jan 23, 2009 12:46 pm (PST)
Hi Henry,
Evidently Michael not only doesn't want a complete
toolbox, he also doesn't want to have a serious discussion on the subject.
That being the case, why not let the rest of us discuss things without
interruption?
First of all, you can get Lactose-free milk at a jiffy
mart. So I don't understand why that's funny.
But really, we are NOT talking about something so
radical as a complete shift in purpose for Photoshop. So the helicopter
submarine car is more than ridiculous - it's juvenile.
We're talking about a shift in technology that ADOBE
ENCOURAGED, and yet did not support in their software. You could always
make separations in Custom CMYK, and if you knew what you were doing they
would be of sufficient quality. The alternative was to let the scanner
operator do the separation for every image, which certainly did not
guarantee a better conversion.
This is not like a flying car that can go underwater.
This is more like the whole world switched to hydrogen fuel, and Adobe
continues to make cars that use gasoline. This is what happens what a
company has a monopoly. The good part is we don't have a problem with
conflicting file formats. The bad part is they can serve up a complete turd
like Photoshop CS4 and we buy it because there's nothing else to get.
Right now you have the lack of innovation you saw at
Quark with the forced upgrade cycle you get from Autodesk. The funny thing
is that other products, especially After FX, have some really cool tools
that seem well thought-out. So it's possible for them to do cool
stuff--it's just that the Photoshop team is focusing on other areas right
now.
So we can add an incredibly ugly 3d coke can to our
image without leaving Photoshop, but we have to use 3rd party software to
adjust black generation in a CMYK conversion? I would guess that 90% of all
CMYK conversions happen in Photoshop. YOU CANNOT SAY THAT THIS IS NOT A
FUNDAMENTAL PART OF WHAT PHOTOSHOP DOES.
Now is the time we need something like Live Picture, or
X Res, to give Adobe a reason to focus on its main constituents.
J Walton
___________________________________________________________________________
Re: Eurostandard inks - anyone?
Posted by: Stephen Marsh
Fri Jan 23, 2009 7:23 pm (PST)
Ron Kelly:
Supporting CMYK is built in to Photoshop; it's part of
the core. If
Adobe pulled that functionality from the program, a
lot of us would be
looking for a replacement. We prepare files for press
and that's at the
core of our business.
The support to one of it's core functions, CMYK, black
generation and
modification - is weak. It has been weak for years,
and many on this
list, especially Dan have said so.
J Walton:
We're talking about a shift in technology that ADOBE
ENCOURAGED, and
yet did not support in their software. You could
always make separations
in Custom CMYK, and if you knew what you were doing
they would be of
sufficient quality. The alternative was to let the
scanner operator do
the separation for every image, which certainly did
not guarantee a
better conversion.
CMYK support could mean two things: (#1) separation or
creation of a 4 channel file that is recognised as being in CMYK mode by
various software; (#2) the ability to open a CMYK mode file (with or
without conversion of the initial values), display an accurate composite
preview and to subsequently apply edits to the composite or separate
channels - with the ability to save that file so that other software will
see it as a CMYK mode file. Ideally one can do both in the one tool (say
Photoshop). What concerns this thread is #1.
I think it is important to look at things from a
historical perspective, in order to understand where we are today. Back in
the day most high end or other 4C work was separated at the drum scanner,
RGB originals were not really an issue as the file obtained was CMYK mode,
even if the scanner was capturing the initial data as RGB. The high
resolution 4C separation was often placed into the layout and the CMYK
values were simply sent to the RIP. If the image did require editing, then
Photoshop was often used. Many used a legacy or Custom CMYK setting that
was inappropriate for colorimetric use with the scanner separated image.
Although this created problems for monitor based edits, it was not
impossible to work with images direct in CMYK mode, either visually and/or
’"via the numbers; If one *converted* the CMYK file to RGB, then
the colorimetric issues became more of an issue when the image was
eventually converted back to CMYK. Many did not convert scanner CMYK files,
while for others it was no concern.
Workflows changed over time, more work was being
scanned on flatbed scanners as RGB and the CMYK separations were being
performed in Photoshop, although many still converted to CMYK at the
scanner or in other software if it offered better results than Photoshop.
Back then, Photoshop could use the built-in separation options or use a
separate Look Up Table option. Similar to ICC profiles, the LUT option was
‘preferred’ if one had the ability to make them work. For most,
using the built-in separation option was the only method available to them.
With Photoshop 5, ICC profiles were added as a third option. In Photoshop
6, the LUT option was removed and the ICC method was touted as being the
preferred option. The legacy built-in separation options were consolidated
into what we now know as Custom CMYK.
Whether for right or wrong, many people embraced the
Photoshop Custom CMYK approach, due to the accessibility and flexibility
that it provided them in creating pleasing separations. When they used ICC
profiles instead of Custom CMYK, they found that they did not have the
control that they did using the outdated Custom CMYK method.
So where are we now? Some users achieve acceptable
results using ICC profiles, whether a standard industry condition suitable
for national distribution to multiple printers printing to that
specification, or a custom ICC profile for a single local press/proofing
condition. Others do not get the same results and wish for more flexibility
in creating separations, these users rely on Custom CMYK to get the job
done. Many of these users appreciate that Custom CMYK is not an ideal tool
and would welcome an option from Adobe to be able to adjust separation
parameters so that they are not at the mercy of the locked in TVI, GCR,
black start, black limit, GCR curve shape, total ink limit etc. Some wish
for even more flexibility, such as being able to use a Light GCR style
ratio for saturated areas and a Heavy GCR style for neutral areas in the
same image.
Adobe built an expectation with Custom CMYK (as flawed
as it is, I also appreciate what it does do right). As ICC profiles have
superseded Custom CMYK as Photoshop's preferred separation method, it is
only reasonable that some Photoshop users have an expectation of being able
to separate with the same flexibility as with Custom CMYK.
Stephen Marsh
___________________________________________________________________________
.
Re: Eurostandard inks anyone?
Posted by: "John Romano"
Fri Jan 23, 2009 12:31 pm (PST)
On 1/22/09 3:35 PM, "Ron Kelly" wrote:
If a slide rule and a computer both come to the same
answer, it's
quite possibly indicative that you can get the correct
answer more than one way. That
doesn't mean that the slide rule user was lucky,
No It means the slide rule user is wasting time and is
less productive. Causing his company to loose money and effectively be run
out of Business By the computer user.
John Romano
___________________________________________________________________________
Re: [colortheory] Eurostandard inks � ?? anyone?
Posted by: "Ron Kelly"
Fri Jan 23, 2009 6:25 pm (PST)
John:
I'll just plug away here on my good old slide rule.
Looks like you need a new computer; you're "loosing" too much
money with this one. Get one that does grammar while you're at it.
Good Luck,
Ron Kelly
___________________________________________________________________________
Re: Eurostandard inks – anyone?
Posted by: Dan Margulis
Fri Jan 23, 2009 6:25 pm (PST)
Terry writes,
Of course, there is no ONE profile that's ideal, even
for a starting
point or default. Depends where you're at
geographically and what type
of printing is typical of your projects.
As the tone of this thread has started to deterioriate,
maybe we can get it back on track with a discussion of the profiles being
recommended. I have suggested in past years that certain color management
consultants (not Terry) do not appear to grasp how important small
variations are, and to assume that certain profiles are "close
enough" when in fact they are not close at all. Also, there has been a
tendency to look at profiles that are OBVIOUSLY screwed up and to defend
them on specious grounds.
So I have some questions and comments for Terry about
the quality of these profiles.
In the USA, you'd have to say...
GRACoL2006 Coated1 for sheetfed printing on coated
stock.
This is a good recommendation. I'd be happy with that
as a starting point for sheetfed printing, if I could accept an uneditable
profile.
My question, though is, could you please comment on how
big of a difference there is between this profile and its predecessor, U.S.
Sheetfed Coated v2? The reason I ask is, this list for several years saw a
vigorous and bellicose endorsement of Sheetfed v2 for exactly the purpose
for which you and I are now endorsing GRACOL2006. We were told that it had
won awards for its quality, was based on the best measurements, that it was
put together by a genius, it did not need to be edited, and the usual
rhetoric about how Custom CMYK is only for those disinterested in quality.
Responding, I called Sheetfed v2 unusable because its assumptions about dot
gain were off the wall, resulting in separations that were grossly too
light. I said that anybody familiar with sheetfed printing could see in a
second that this profile was garbage, and that anybody halfway decent in
Custom CMYK would be able to make a significantly more accurate profile
without the benefit of any measurements at all.
But enough of my opinions. Let me ask yours, so that we
can know what kind of tolerances we have. In the test image described below
I designated as a midtone test a point reading 52L0a0b. On separation with
the GRACOL profile that we both apparently agree is pretty accurate, we get
50c41m41y17k. Separating the same original using Sheetfed v2 produces
49c38m38y7k.
These two profiles were recommended for the same
purpose: general preparation of a file for sheetfed printing. So my
questions are,
1) Do you consider that 50c41m41y17k and 49c38m38y7k
are grossly different values?
2) If a sheetfed printer is given an area of
50c41m41y17k, about how much tolerance would you give him to vary from
those values on press and still consider that he has adequate process
control?
3) Same question, this time with respect to performance
of profiles. Another profile prepared by another method would not match
this one exactly. But suppose you are reasonably satisfied with what we
have. Assuming that it has approximately the same GCR method (relative
strength of CMY vs. K), can you say about how much tolerance you would have
for a competing profile to vary from 50c41m41y17k, before concluding that
the other profile was not an adequate substitute?
SWOP2006 Coated3 for web offset (publication) printing
on a quality
coated stock.
Fair enough This is also a good profile. Those who
currently use SWOPv2 as a CMYK setting would be better off with this one.
For uncoated stocks (I'm thinking "offset"
stock), I think FOGRA 29
has a lot going for it. It is very close to the
"G7" tone response
curve and the colorimetry seems about right. Until we
get a "G7"
version of an uncoated stock, FOGRA 29 should do fine.
I would like to find out about this "about
right". My test image is a picture of a brownish tiger cat. I have
selected a certain part and removed all color, verifying that throughout it
measures 0a0b. I have tested this not just with FOGRA29, but with a variety
of profiles, some intended for this uncoated condition and others not. To
wit, SWOP v2, SWOP 2006-3, Custom CMYK coated defaults with appropriate dot
gain change, GRACOL, U.S. Sheetfed Coated v2, Custom CMYK uncoated defaults
with appropriate dot gain change, U.S. Sheetfed uncoated v2, and Euroscale
Uncoated v2. Let us consider two spots.
In the dark areas of the three-quartertones, I have a
value of 14L0a0b. FOGRA29 separates this pure neutral into 82c74m58y62k.
Are you aware of any printing conditions anywhere in the world where this
formulation does not yield a decided purple-blue? I'm not, and neither,
apparently, are the folks who prepared any of the profiles mentioned above.
The key number here is the relation between magenta and
yellow. Let me give you, for each of the other eight profiles, the magenta
value as compared to yellow. They are: +1, 0, 0, +3, 0, 0, +2, -1. FOGRA29,
at +14, is on some other planet. Do you think this is close enough to be
acceptable?
Not only would one get blue shadows by separating with
this profile, but even subdued colors are cooled down considerably. This
cat is a dark yellowish brown, 24L5a11b, somewhat closer to yellow than it
is to red. FOGRA29's idea of the CMYK equivalent of this is 61c70m73y49k.
That looks mighty purple to me, and to the other profiles. FOGRA29 has the
yellow only three points higher than the magenta. The other eight have it:
11, 14, 12, 12, 14, 11, 11, 17.
In light of the above, a) do you consider the variation
between FOGRA29 and all these other profiles to be significant enough to
concern ourselves over? b) do you stick by the statement that "the
colorimetry seems about right," and c) do you think that people would
be justified in wanting to change these bluish shadows to something more
standard?
Dan Margulis
___________________________________________________________________________
Re: Eurostandard inks � ’¶ anyone?
Posted by: Stephen Marsh
Sat Jan 24, 2009 12:20 am (PST)
Despite the regularity and similarity of these ground
hog day threads on overcoming the limitations of "standard" one
size fits all profiles, there are some good points that have been raised
(yet again).
Ron Kelly:
Why doesn't it offer better support? Why must you go
to another program
to try and cope in this aspect? I think that's the
question here.
I would agree that it should be central to Photoshop.
Why have we got
umpteen different ways to address the color and range
of a file but only
one kludgey method for controlling/modifying black
generation?
It would be interesting to know who insists it be so,
and why, wouldn't it?
The Photoshop development team does not appear to agree
that having the ability to alter the black generation or other components
of the separation of ICC profiles is something that Photoshop should do.
Developers and users often see things differently, some developers are more
open than others to adding features at the expense of simplicity or ease of
user support.
The expectation was set with Custom CMYK. It is
understood that ICC is different technology and that the separation
variables are hard wired into the profile and that one requires separate
profiles for each separation parameter change. This is obviously a less
flexible approach than Custom CMYK, although it can deliver better results
when the profile contains acceptable separation parameters and accurate
measurement data. It is also understood that profiles are usually created
using specialised third party software. One may note that when Adobe
created ICC profiles for distribution with Photoshop, they created their
own ICC profile generation software which is not available to the public
(they did they not use a third party application that was available
‘off the shelf’). Despite these issues, it was previously
demonstrated with the defunct Imation CFM (now part of Kodak?) that one
could dynamically alter the separation results of ICC profiles without
making new measurements or regenerating a new profile based on the original
measurements with new separation parameters. Perhaps Device Link Profiles
can do a similar job, or some of the other back end ink saving software
that has been mentioned in previous topics. Finally in Photoshop CS4, one
can access Device Link Profiles, however one still needs third party
software to create the DLP. Where there is a will, there is a way. In this
case there is a way, but is the will there? I don't think so.
Ron Kelly:
Supporting CMYK is built in to Photoshop; it's part of
the core. If
Adobe pulled that functionality from the program, a
lot of us would be
looking for a replacement. We prepare files for press
and that's at the
core of our business.
Where would we look for CMYK editing support Ron? If
Photoshop stopped working today, what software would you use to to work
with CMYK files like you would have in Photoshop? Mac OS X or Win OS?
During the 1990's there were numerous CMYK applications, some affordable
and some less so (in relation to Photoshop at the time). Since then they
have died by the way, absorbed and dropped by other companies buying them
out or simply being unable to compete with the Swiss Army Knife of image
editing, otherwise known as Photoshop. The code is not in the public
domain, even if these old tools will never be commercially published again.
There is a lot of public domain info for RGB and other modes, but not CMYK.
There does not appear to be many serious contenders for CMYK work available
commercially or shareware etc. Perhaps I need to look harder, I admit that
I have not spent too much time looking for another CMYK image editor suited
for prepress. The last one that I looked at was Binuscan Photo Retouch Pro,
which they have dropped before ever releasing the promised Win OS version
(I presume that Mac sales were too poor to justify development or continued
sale of this image editor, Binuscan appear to have changed their
direction).
Henry Davis:
My observation is that Photoshop has been continually
drifting
further away from pre-press needs and has evolved to
become a product
that is aimed more at photographers than any other
group. I credit this
to nothing more than a business decision. But is does
seem a shame that
Adobe's original partner remains at the dance while
they have left with
another. It's bad form.
Agreed, Henry, in many ways the general image editing
prepress market is dead or next to non existent in sales, the money is with
photographers, today with digital anyone is a photographer � ’¶
just as anyone is a prepress operator. So, if photographers are the core
market for Adobe Photoshop, some of them will require good prepress/CMYK
tools and workflow support in Photoshop. I think Photoshop's original
partner was RGB only scanning, if one looks back far enough, however I
agree that Photoshop has a long history of support for CMYK tasks and this
area should not be short changed.
Prepress specific updates to Photoshop are not too
common. I may be missing some things, but the big changes were in Photoshop
5, where spot colour channel support was added, not to mention the
introduction of ICC conversions and ‘Colour Management’.
Photoshop 6 introduced colour setting design changes and softproofing.
Photoshop CS4 added the ability to use device link and other profiles.
RJay Hansen:
I'm not for a "lowest common denominator"
approach with program like
Photoshop that's aimed at professionals. Anyway, I
believe there's
already a version of Photoshop that doesn't do CMYK.
It's called
Elements.
And as Richard Lynch demonstrated with his
‘Hidden Power’ add on kit to Photoshop Elements, one can do
CMYK (after a fashion) and use other tools and features that Adobe would
have a casual user upgrade to the full version to access.
I think it will take a small, independent developer or
perhaps a wide group via open source, to create the sort of CMYK tools that
many on this list have on their wish list. Where is Separationshop? The
developer could be a Mac based one, there is a wide user base for their
software and the Mac OS API provides ColorSync CMYK and other mode
conversions, somebody should be able to build a simple applicaton hacked
togehter using Core Image and ColorSync etc. Apple Preview offers some
limited ability in these respects.
Sincerely,
Stephen Marsh
___________________________________________________________________________
Re: Eurostandard inks – anyone?
Posted by: George Machen
Sat Jan 24, 2009 8:43 am (PST)
Get a load of this! Ha-haa-haaaa! Whom does Chris Cox
think he's trying to kid, anyway?!
- George Machen
--
http://www.gthelp.com/showthread.php?t=160345
Re: Photoshop CS4 (64bit) - custom CMYK #7
January 15th, 2009, 02:33 AM
Guest
Chris,
Can you elaborate a little more about how Custom
CMYK is obsolete? What do you mean that it's not
ICC compatible and what features have Custom CMYK
been replaced with?
Thanks!
--
Re: Photoshop CS4 (64bit) - custom CMYK #9
January 15th, 2009, 02:35 AM
Guest
rydog - it is obsolete, period. The Custom CMYK
engine produces low quality results, doesn't take
ICC profiles as input or a starting point. It's
Photoshop 2.5 technology THAT WE BARELY KEEP
WORKING. There hasn't been a replacement for
CustomCMYK yet. I KEEP TRYING TO GET A
REPLACEMENT IN THE APP, BUT AS LONG AS CUSTOMERS
AREN'T DEMANDING IT, PRODUCT MANAGEMENT POINTS TO
OTHER PROFILING APPS AND SAYING THAT'S GOOD ENOUGH.
[My emphasis; this presumably is Chris Cox.]
--
Re: Photoshop CS4 (64bit) - custom CMYK #11
January 16th, 2009, 02:28 AM
Guest
but as long as customers aren't demanding it,
OK, I DEMAND it! Where the hell is it?
Is it done yet?
Seriously, it would be better if this
"barely...working" anachronism weren't
included
at all. It leads users to believe that they can
take an icc compliant profile and tweak it in
Photoshop. The results can be acceptable but they
are rarely stellar.
Maybe product management needs to reconsider
Adobe's relationship with its original user base.
--
Re: Photoshop CS4 (64bit) - custom CMYK #12
January 16th, 2009, 02:29 AM
Guest
One of the problems is that there is at least one
alpha tester who does want to remove Custom CMYK
from Ps, but only as a vindictive jab at people
who are still using it - sort of a way to punish
them for not agreeing with his philosophy and
force them into the twenty-first century.
[I just wonder whom this "alpha tester"
really
is? Is Chris being a little disingenuous in this
thread?]
--
Re: Photoshop CS4 (64bit) - custom CMYK #14
January 16th, 2009, 02:29 AM
Guest
Ho - file a bug report/feature request, please.
Pfigen - well, we have had a lot of confused
customers who think they are tweaking an ICC
profile by using the ObsoleteCMYK engine (THEY
WOULDN'T LET ME RENAME IT IN THE UI :-).
[My emphasis.]
--
Re: Photoshop CS4 (64bit) - custom CMYK #15
January 16th, 2009, 02:30 AM
Guest
Folks at my job use it to generate black out of a
job and place it into other colors to reduce
registration problems on press. What would they
use if not CustomCMYK to do this?
--
Re: Photoshop CS4 (64bit) - custom CMYK #16
January 16th, 2009, 02:31 AM
Guest
It's dangerous to leave in a feature like that
without some kind of warning telling the user
that a custom CMYK setting has no relation to any
of the other included profiles.
Even from the very beginning of ICC support in Ps
5, it was made clear when you read about it that
Custom CMYK and the ICC dialog boxes had no
relationship with each other. Unfortunately too
many people thought they were connected and
suggested that you could actually edit ICC
profiles there. That caused a lot of confusion
and some of that early misinformation persists
today.
It's not you can't get pretty good results from
Custom CMYK for some printing scenarios, because
you can. It's more that it has become less and
less relevant as the ink definitions are from an
era predating direct to plate, and there
apparently there were a few half fixes to get it
where it has been since it's last revision.
Adding editing capabilities to Ps would probably
solve the problems that force some to still use
Custom CMYK for some purposes. As Rydog23 points
out, there are still legitimate uses for this
tool no matter how obsolete it may be. There are
other uses too if you don't happen to have access
to Profilemaker or PrintOpen or something similar.
[This presumably is Chris Cox.]
___________________________________________________________________________
.
Re: Eurostandard inks – anyone?
Posted by: "Peter Figen"
Sat Jan 24, 2009 11:55 am (PST)
Even from the very beginning of ICC support in Ps
5, it was made clear when you read about it that
Custom CMYK and the ICC dialog boxes had no
relationship with each other. Unfortunately too
many people thought they were connected and
suggested that you could actually edit ICC
profiles there. That caused a lot of confusion
and some of that early misinformation persists
today.
It's not you can't get pretty good results from
Custom CMYK for some printing scenarios, because
you can. It's more that it has become less and
less relevant as the ink definitions are from an
era predating direct to plate, and there
apparently there were a few half fixes to get it
where it has been since it's last revision.
Adding editing capabilities to Ps would probably
solve the problems that force some to still use
Custom CMYK for some purposes. As Rydog23 points
out, there are still legitimate uses for this
tool no matter how obsolete it may be. There are
other uses too if you don't happen to have access
to Profilemaker or PrintOpen or something similar.
George,
The above quote was from me, not Chris Cox.
I've been following this discussion for a while now. As
some of you know, I'm a pretty big supporter and sometimes defender of Dan,
but I'm also someone who fully embraces and used the most modern and best
tools I can find.
While there is some validity in arguing that Adobe
*should* upgrade Custom CMYK, which is in dire need of something, it's also
fairly unrealistic to expect them to do anything, taking into their history
of working on that feature.
I come at all this prepress stuff from being a
commercial photographer who had to learn these skills to diversify. I love
Photoshop for what it is, but I don't spend a lot of time worrying about
features I have very little control over. For instance, Photoshop has
notoriously sucked at certain types of masking, particularly, fine hair,
smoke, etc. When Ultimatte introduced Knockout, several years ago, those of
us jumped on that product as quickly as possible and no one complained
about the (then) $500 price tag for a very one trick pony.
I feel exactly the same when it comes to being able to
edit and generate custom profiles. Sure, it seems like a lot of extra
money, but what it really does is allow you to do the real professional job
that is required today. It's part of the prices of admission, as far as I'm
concerned, and complaining while waiting for Adobe to do something they
might never do anyway is wasted energy.
The thing that owning and using these tools -
ProfileMaker and a Spectrolino - in my case, does, is let me know that I'm
sending files that are the right ones for my clients printing, not one that
I think might be sort of in the ballpark. Ballpark isn't good enough for
them, and Custom CMYK just doesn't have the ability to compensate for the
quirks of everyone's non-conforming direct to plate press and proofer
calibrations.
I've never argued that Custom CMYK should be taken out.
I dont', but I don't think it's the best tool anymore. I've mentioned to
Chris Cox several times that perhaps something like the Imation CFM would
work. I had an early version of that product, and while it was buggy, it
did allow you to change black generation and total ink of existing
profiles. There's never been any response from him. I'm not going to sit
around hoping there is. I've purchased the tools that do the job today and
moved on. Not everyone needs the level of control and accuracy I do, but if
you do, then investing in ProfileMaker or Monaco Profiler can actually pay
for itself in increased efficiency and better results, which mean happier
clients who hire you more often.
Interestingly enough, I'm currently working on a CD
cover that is being produced at DiscMakers. When I contacted them about
printing specs the first response was that they are not color managed and
ignore all profiles. Just send CMYK, they told me. No mention of what
flavor. Finally they sent me a profile that was a Custom CMYK setting of
SWOPcoated13%DG.
When I contacted Bryan Pylant, who posts here, he said
they use either of a couple Custom CMYK settings, which he provided, OR to
use the SWOP Coatedv2 profile supplied in Ps. Here's the problem with this
scenario, even though he claims to be SWOP/GraCol compliant, it can't be
both because there are major differences these two recommended settings.
Assigning one profile to the other conversion causes a major tonal and
color shift on screen. All of this does not inspire confidence in the
process.
My best guess in this case, has to be with the SWOPv2
profile, because that's what a majority of people use as their Ps defaults,
and if Discmakers is ganging runs of all sorts, this would be the most
common common denominator.
Peter Figen
___________________________________________________________________________
Re: [colortheory] Re: Eurostandard inks anyone?
Posted by: Dan Margulis
Mon Jan 26, 2009 8:52 am (PST)
Henry writes,
I observe that people are reasonably frustrated over
the kludge that
is Custom CMYK. The question I asked earlier generated
no response.
Perhaps there isn't a "nice" truthful
response available, but I am
interested in speculative and even angry answers if
that will get the
ball rolling at Adobe to make the tool be what the
name of the tool
implies. I'll repeat again here:
"Since then I've asked myself, "why can't it
work correctly in the
simple way that the tool itself implies that it
should?". Is there a
way that it could be made to work properly? Is there
an official
Adobe reason as to why not?"
There is no difficulty in coding the capability. A
competent person could do it in one to two days.
Why the Photoshop engineers refuse to do it, and what
makes them consistently threaten to delete it without a replacement, has
been discussed several times on the list, e.g.
http:
//www.ledet.com/margulis/2006HTM/ACT-Asking_Adobe.htm
http:
//www.ledet.com/margulis/2007HTM/ACT07-What_Makes_Adobe.htm
Custom CMYK could be the tool that it is implied to
be. For me, the
tool has been a lie from the beginning, and it has
continued to be a
lie in spite of being called out.
I trust you are referring to post-1998, with the
introduction of uneditable profiles. Pre-1998 it was at least one of the
better options available, and in the early 1990s it would have to be rated
as quite impressive in its capabilities.
If that doesn't frustrate one to
some degree of anger, then one must have a very
lenient attitude
toward the developers and their management.
First, with the world in its current state, there are
more than enough things to be angry about than any shortcomings in
Photoshop.
Second, I suspect that most of the things that make us
angry about the Photoshop team are because we have a certain background
that they lack. In the same time that they have produced a dozen different
versions of Photoshop, many of us have produced close to a million
individual, distinct jobs. I assume that everyone who has been in that
position has been guilty of a large number of screwups. Certainly I have
from time to time handled jobs stupidly, done things I should have known
better than to do, and/or made misjudgments due to arrogance. Nobody likes
it when this happens to them. It takes us too long to admit that we have
done something badly. It's understandable that we get angry and look to
blame others for our own stupidity. If somebody vehemently criticizes our
work, we tend to take it personally.
When you consider how badly *we* take failure, even
though we have failed so many times that we're kind of used to it, it's
easy to understand how hard a *programmer* may take it. When you consider
that when somebody criticizes a job we did, we sometimes take it as a
personal affront even though we know they're just saying we made some bad
choices about color, you have to feel sympathy for programmers when people
criticize something that took them eighteen months to produce.
Have you ever wished you could do something to get even
with a client who criticized you? Have you ever wanted to insult him to his
face—and occasionally because he objected to some mistake you had
made? I have. And so, I can't fault anybody at Adobe for *thinking* similar
thoughts, even when they're wrong. Provided, of course, that they keep
their fantasies of sabotaging the workflow of their enemies, and the nasty
words they'd like tell them, under their hat, the same way we would have
to. What one *can* get irritated about is that management lets them get
away, year after year, with saying these things in public.
Dan Margulis
___________________________________________________________________________
Re: Eurostandard inks – anyone?
Posted by: Henry Davis
Mon Jan 26, 2009 2:19 pm (PST)
Boy, this topic has been kicked around.
Sorry I wasn't clear about the Custom CMYK timeline.
What I was getting at was the disappointing color that results from keying
in solid and overprint values both then and now. The control of GCR and
black generation was impressive, but it needed improvement with regard to
color. In the archive was this comment:
"I think I'd need both hands and a foot to count
the people who've used measured data and the custom CMYK dialog."
That just seems out of touch to me. The fields are
editable. Anyone with curiosity and a measuring device that gave Lab would
have tried it. I did it a lot early on until I found that the best thing I
could do with it was to alter the GCR/black.
I'm not angry with the Photoshop team. It must be a
real tough work environment. I can even sense their frustration. Adobe is a
big ship and it doesn't turn without a lot of effort. I appreciate that.
I've made mistakes and have eaten crow, but I've always fixed them - even
at my own expense. Expecting a big company to admit a mistake is
unrealistic and it's not necessary. A better level of dialogue on this
topic would be nice, though. OK, I suppose there is a little anger leftover
from v5 debacle, but it's because this topic has its roots in the v5
transition to profiles.
Reviewing the archive of this topic, it strikes me how
much of the dialogue is defensive in a snitty and legalistic sort of way.
The most telling thing is the absolutism that is asserted for the need and
use of profiles, along with an obfuscating attitude when it comes to direct
user support for them within the program. I don't see this attitude with
the Camera Raw module. Adobe has continued improving their support for that
format. Imagine the uproar if Camera Raw were given the same treatment:
Suppose Photoshop only supported one hard-wired translation of a Camera Raw
file and then took the same attitude to the debate.
Henry Davis
___________________________________________________________________________
.
Re: [colortheory] Re: Eurostandard inks � ’¶
anyone?
Posted by: Michael Jahn
Mon Jan 26, 2009 5:07 pm (PST)
I would like to just say one simple thing.
Adobe Photoshops ACE Engine and Adobe Photoshop using
LAB as a PCS is the issue as far as I am concerned.
This is a broken approach. It yeilds crappy results,
and we have simply lived with it because that is what the ICC decided was
the best approach so very many years ago when were were scanning
transparencies and viewing them on CRT monitors - and using film to expose
plates. We no longer do these things today.
Now, we capture images digitally and view them on LCD
screens, and we often print digitally often without using CMYK ink on an
offset press.
on TOP of all this - basic math is flawed.
So, before I would be asking anyone at Adobe to give me
some new tool, I would really have to say that we need to rething color
image processing in a very fundamental way.
It is simply too hard, and while clever people like Dan
and Henry can figure it out, well, this simply should not be so hard (and
it is)
I mean, I can ask anyone to make a design that has a
120 point letter M with serifs. I can even select a font like Arial, and if
I asked them to print it using 100 Black, and everyone shipped it to me -
everyone's output would indeed identical sized letters Ms... and probably
look more or less like the same look black.
but I can tell you all to make 30 cyan, everyone would
be ALL OVER THE MAP with few actually matching 30 cyan.
--
Michael Jahn
___________________________________________________________________________
.
Re: [colortheory] Re: Eurostandard inks � ’¶
anyone?
Posted by: J Walton
Mon Jan 26, 2009 5:40 pm (PST)
On Mon, Jan 26, 2009 at 2:58 PM, Michael Jahnwrote:
but I can tell you all to make 30 cyan, everyone would
be ALL OVER THE
MAP with few actually matching 30 cyan.
Why would everyone be all over the map if you are
specifying the ink percentage? That isn't making sense.
J Walton
___________________________________________________________________________
Re: [colortheory] Re: Eurostandard inks � ’¶
anyone?
Posted by: Michael Jahn
Mon Jan 26, 2009 6:49 pm (PST)
Hi J Waltron
I was suggesting that you would print out locally on a
desktop printer. You know, like on the printer in your office or studio.
Of course - i am pretty much convinced my "all
over the place" is probably most peoples "gee, they look the same
to me..."
--
Michael Jahn
Jahn & Associates
___________________________________________________________________________
Re: Eurostandard inks – anyone?
Posted by: Stephen Marsh
Mon Jan 26, 2009 7:58 pm (PST)
Henry Davis wrote:
Reviewing the archive of this topic, it strikes me how
much of the
dialogue is defensive in a snitty and legalistic sort
of way. The
most telling thing is the absolutism that is asserted
for the need
and use of profiles, along with an obfuscating
attitude when it comes
to direct user support for them within the program. I
don't see this
attitude with the Camera Raw module. Adobe has
continued improving
their support for that format. Imagine the uproar if
Camera Raw were
given the same treatment: Suppose Photoshop only
supported one hard-
wired translation of a Camera Raw file and then took
the same
attitude to the debate.
Yes, one can (only) choose among 4 rendering/output
profiles and not just one! If your preferred RGB working space is not one
of the four blessed by ACR, then one has to make another, extra mode
conversion once the file is in Photoshop. This is by design. Pixmantec Raw
Shooter could render into any installed profile. Adobe liked the Pixmantec
Vibrance command and improved upon it, however they must not have agreed
with their workflow.
Why ACR has been designed NOT to render the raw camera
data into any installed profile, just like Photoshop, is beyond me* (simply
make the other installed profiles available via an advanced checkbox
option, if this is going to confuse users). I am not talking about just any
profile, only RGB working space profiles, however any profile should be OK.
*This is not to spark a whole debate on ACR, I have
read the developer stated reasons at the Adobe ACR forum - I am just noting
that what some users wish is not always how the developer sees things.
Stephen Marsh
___________________________________________________________________________
Re: Eurostandard inks - anyone?
Posted by: "Pylant, Brian"
Mon Jan 26, 2009 8:52 am (PST)
When I contacted Bryan Pylant, who posts here, he said
they use either of
a couple Custom CMYK settings, which he provided, OR
to use the SWOP
Coatedv2 profile supplied in Ps. Here's the problem
with this scenario,
even though he claims to be SWOP/GraCol compliant, it
can't be both
because there are major differences these two
recommended settings.
Assigning one profile to the other conversion causes a
major tonal and
color shift on screen. All of this does not inspire
confidence in the process.
My best guess in this case, has to be with the SWOPv2
profile, because
that's what a majority of people use as their Ps
defaults, and if
Discmakers is ganging runs of all sorts, this would be
the most common
common denominator.
As I replied in our offline emails, I use the custom
CMYK settings for re-separating to get a lighter black than the SWOP
provides, not as an output profile to preview the press. We don't currently
have any other profiling software to achieve that, so I make do with what
Photoshop provides. And in the cases where I decide to go that route, I
don't assign the SWOP profile afterwards, and what I get off of the proof
and press are pretty much as expected. I apologize if I added confusion by
including those.
BRIAN PYLANT
Manager, Electronic Prepress
Disc Makers
7905 North Route 130
Pennsauken, NJ 08110
___________________________________________________________________________
Re: Eurostandard inks – anyone?
Posted by: Dan Margulis
Tue Jan 27, 2009 3:36 am (PST)
George writes,
Get a load of this! Ha-haa-haaaa!
Whom does Chris Cox think he's
trying to kid, anyway?!
The reasonable remarks in message #16 that you
identified as being from Chris Cox were actually made by Peter Figen. The
two others *are* from Chris Cox and I would comment as follows.
rydog - it is obsolete, period.
Chris Cox has been declaring this, in ever more
strident terms, for eleven years, and still AFAIK substantially every CMYK
expert makes *some* use of Custom CMYK.
The Custom CMYK engine produces low quality results,
It probably would, if operated by Photoshop engineers.
Of the four lead profiles that the Photoshop engineers have created for
release, only one (SWOP v2) reaches the level of mediocrity. The others,
anybody using Custom CMYK should be able to beat without difficulty.
The other three Adobe-generated profiles assume 1) that
a typical sheetfed press has far higher dot gain on coated paper than a
typical web press does and 2) that sheetfed and web uncoated printing are
one and the same, and the same profile, under different names, suffices for
both. Given that, for any Adobe representative to lecture anybody, even
rank beginners, on how to make good profiles is akin to Rod Blagojevich
telling us that we need to improve our ethics.
doesn't take ICC profiles as input or a starting
point.
That's the whole problem. And since Custom CMYK can't
readily be updated to take them, if the Photoshop engineers were actually
serious about wanting people to adopt an ICC workflow, their only
responsible action was and is to write something new that *will* take them
but not lose any of the Custom CMYK functionality. Instead, for eleven
years, they have insisted that we have to downgrade to a system of
uneditable profiles. And we see, after eleven years, how far their
stubbornness has gotten them, and the irreversible damage it's done to ICC
color management.
It's Photoshop 2.5 technology that we barely keep
working. There hasn't been a replacement for
CustomCMYK yet. I keep trying to get a
replacement in the app, but as long as customers
aren't demanding it, product management points to
other profiling apps and saying that's good enough.
The statement that customers aren't demanding it is too
ludicrous for comment. The statement that he keeps "trying to get a
replacement" for Custom CMYK is intentionally deceptive. For many
years Chris Cox has been threatening to delete Custom CMYK altogether, no
replacement. He has made it abundantly clear that his purpose is to revenge
himself on those he considers his enemies. Only a few months ago, on the
ColorSync list, we were treated to this:
Date: Mon, 08 Sep 2008 18:53:12 -0700
From: Chris Cox
Yeah, every time I try to get rid of it, somebody
screams bloody
murder.
Back to the present post. Chris Cox writes,
Pfigen - well, we have had a lot of confused
customers who think they are tweaking an ICC
profile by using the ObsoleteCMYK engine (They
wouldn't let me rename it in the UI :-).
It is correct that people often don't realize that when
they switch from, say, SWOP v2, to Custom CMYK, they're throwing out
everything and starting from scratch. Understandable, too, because the
interface clearly suggests otherwise. I reported this as a bug during the
Photoshop 5 beta period and pointed out that confusion was inevitable and
that a warning dialog should be inserted. The Photoshop engineers elected
to ignore the advice, and the inevitable confusion resulted. For example,
in
http:
//www.ledet.com/margulis/ACT_postings/SeparationIssues/ACT-Sheetfed_and_Web.htm
(2004) Howard Smith, who is usually pretty
sophisticated about such things, got suckered in by the deceptive
interface. I explained, and repeated my criticism. In
http:
//tech.groups.yahoo.com/group/colortheory/message/9553
Andrew Rodney, of all people, said the following
(capitalization is in original post):
Hold the presses, I actually totally agree with Dan on
this point. This is
totally DUMB that Adobe doesn't inform you that the
switch has nothing to do
with the U.S. Web Coated (SWOP) v2 profile you were
just working with.
You're supposed to notice that v2 is missing? They
should make this much
clearer.
In short, Chris Cox is attacking not Custom CMYK's
engine, but his own inability to design an interface properly--a failure
that one of the Photoshop engineers' most vocal supporters describes as
"totally DUMB".
The language quoted by George is not as strong as what
Chris Cox ordinarily uses, but the pattern is the same. It's all invective,
no constructive content. Somebody wants to criticize Custom CMYK, go ahead,
it's not a sacred cow, but you have to offer alternatives. For example,
*If you need to print a screen grab of a dialog box, or
a comic strip scanned or created in RGB, how do you ensure that the fine
lines print in black ink only, so that they will not be fuzzy?
*If you are printing a grayscale image but wish to
break it into CMYK, how do you ensure a stronger-than-average black, so
that colors won't shift?
*If you have what you think is a good profile, but the
printer won't accept its results because they do not meet his standards for
total ink, what do you do?
*If you are dealing for the first time with a new
printer who purportedly prints to standards but has a poor reputation for
quality, how do you give yourself the best chance of avoiding disaster, by
providing him a file that's lighter than usual?
*If you like Adobe's SWOP v2 profile but your clients
cannot accept that it produces purpler blues than any past profile, how do
you fix it?
By now the absolute majority of professionals have to
work in CMYK from time to time. These questions are important to them.
There is no easy solution to any of them if one downgrades from Custom CMYK
to a system of uneditable profiles. No alternatives are proposed in any of
the above Cox rhetoric, any more than there was in his earlier intervention
in Adobeforums. The question was asked as how to handle a 4/c grayscale and
several users replied correctly that the only way to do so within Photoshop
was in Custom CMYK. Chris Cox emerged out of the blue with a gratuitous
insult for these Adobe clients, and nothing in the way of a suggested
alternative.
The question of professional competence does not really
arise here, although the mere fact that Cox still doesn't get, after eleven
years, why people continue to use the method he condemns as obsolete,
speaks for itself. Even a star employee who had indulged in a tiny fraction
of the amount of public insulting that Chris Cox has been dishing out
would, at any other software company besides Adobe, have long ago been
summoned into his boss's office, in the presence of a representative of the
human resources department, and given to understand the following corporate
policies:
*We do not publicly apply invidious terms such as
"liar", "lying", "lies",
"brain-dead", etc., to any of our clients or business partners,
current or potential.
*When we enter public discussions about our products,
the purpose is to offer constructive commentary, not to divest ourselves of
insults.
*When we offer constructive criticism of someone else's
workflow, we suggest alternatives.
*When we consistently and contentiously advise our
expert clients that a certain procedure is wrong, and they decline to
change their ways for eleven years and counting, we consider the
possibility that we ourselves are mistaken.
*We do not weaponize software development by
threatening to delete features in order to punish perceived enemies.
*When we provide what we consider a better way of doing
things in our program, we strive not to disrupt existing workflows. We
prefer to allow users who prefer the older method to continue to employ it,
as we have recently done with improvements to the Brightness/Contrast
command, the Dodge/Burn tools, and the Convert to Profile dialog.
*Employees who continue to violate these policies after
formal warnings will be terminated for cause.
Dan Margulis
___________________________________________________________________________
Re: [colortheory] Re: Eurostandard inks anyone?
Posted by: Terence Wyse
Tue Jan 27, 2009 9:16 am (PST)
On Jan 23, 2009, at 9:14 PM, Dan Margulis wrote:
So I have some questions and comments for Terry about
the quality of
these profiles.
In the USA, you'd have to say...
GRACoL2006 Coated1 for sheetfed printing on coated
stock.
This is a good recommendation. I'd be happy with that
as a
starting point for sheetfed printing, if I could
accept an uneditable
profile.
I take that to mean "editable in Photoshop's
Custom CMYK..." because this profile is certainly editable in a
profile editor such as ProfileMaker's Profile Editor module (can I have one
more "profile editor' in that sentence? :-) ). But probably what you
REALLY mean is "editing" in the sense of altering the separation
characteristics. Once again, easily done, but to do this you would simply
start with the raw characterization data and build a new profile using,
yes, a 3rd party profiling application.
My question, though is, could you please comment on
how big of a
difference there is between this profile and its
predecessor, U.S.
Sheetfed Coated v2? The reason I ask is, this list for
several years
saw a vigorous and bellicose endorsement of Sheetfed
v2 for exactly
the purpose for which you and I are now endorsing
GRACOL2006. We were
told that it had won awards for its quality, was based
on the best
measurements, that it was put together by a genius, it
did not need
to be edited, and the usual rhetoric about how Custom
CMYK is only
for those disinterested in quality. Responding, I
called Sheetfed v2
unusable because its assumptions about dot gain were
off the wall,
resulting in separations that were grossly too light.
I said that
anybody familiar with sheetfed printing could see in a
second that
this profile was garbage, and that anybody halfway
decent in Custom
CMYK would be able to make a significantly more
accurate profile
without the benefit of any measurements at all.
Actually, "USSheetfedCoatedv2" is a fine
profile for what it is. From what I understand, USSheetfedCoatedv2 is
simply a profile made from a 3M Matchprint analog/film-based proofing
system. In the era of film/imagesetters, Matchprint was the defacto
standard off-press proofing system and had gain characteristics not unlike
film-to-plate systems of the day, albeit just a bit on the high side. I'm
guessing Dan that you probably made more than a few Matchprints in your
day, as have I, and probably noted that Matchprints were always
"fuller" in the 1/4- midtones that the press. I saw the same
thing when I used to make Dupont Cromalin proofs. Fuji ColorArt and Dupont
Waterproof laminate proofing systems were more accurate in my opinion of
simulating sheetfed printing dot gain.
Here's what's "good" about USSheetfedCoatedv2
(in my opinion):
* Excellent gray balance characteristics. Gray balance
is within about .5-1.0% of what most of us "G7" advocates
consider optimum for midtone: 50/40/40, 1/4 tone: 25/19/19, 3/4 tone:
75/65/65.
* Proper total ink limit, if a bit on the high side, of
350%. Back in my scanner days, that was considered reasonable for sheetfed
printing although these days you see this closer to 320%.
* GCR is on the light-to-medium side (K start of about
35%, midtone K of about 8%, Kmax of 85%).
Here's what's not good about USSheetfedCoatedv2:
* Nominal dot gain of around 25% which is way too high
for typical linear CtP printing today. An even bigger problem I think is
the fact that the dot gain actually peaks around 37% which would result in
1/4 tones that print way too "full". The resulting separation
using this profile would result in very weak/flat 1/4-to-midtone
detail/shape if printed linear CtP. As a comparison, the
"GRACoL2006_Coated1" profile has a nominal dot gain of roughly
15% which peaks further up the scale at around 45-47%. Call the difference
between USSheetfedCoatedv2 and "GRACoL" to be around 10%....which
is HUGE.
* Ink trap/overprint colors (C+M, C+Y, M+Y) have the
typical characteristic of most any off-press laminate proofing system in
that the ink trap is "perfect", which is to say not realistic for
offset printing. Using the typical ink sequence of K-C-M-Y, a proof made
using this profile would have blues that would be too magenta, greens that
would be too yellow and reds that would be too yellow. If you separate
using this profile, the printed results on the press would be in the
opposite direction (too cyan blues, too cyan greens and too magenta reds).
* On the technical side, this profile has an A2B table
(CMYK->LAB) that has only 9 grid points or "resolution".
Typical high-quality profiles would have around 17 grid points or about
twice the resolution. This lack of resolution on the A2B side is not
relevant for separations however, only if used for soft-proofing and print
proofing when used as a source profile. The B2A or "separation"
side of the profile has the typical 33 grid point resolution.
So the bottom line is that profile is really meant for
a different era in printing, the days prior to CtP printing.
But enough of my opinions. Let me ask yours, so that
we can know what
kind of tolerances we have. In the test image
described below I
designated as a midtone test a point reading 52L0a0b.
On separation
with the GRACOL profile that we both apparently agree
is pretty
accurate, we get 50c41m41y17k. Separating the same
original using
Sheetfed v2 produces 49c38m38y7k.
These two profiles were recommended for the same
purpose: general
preparation of a file for sheetfed printing. So my
questions are,
1) Do you consider that 50c41m41y17k and 49c38m38y7k
are grossly
different values?
Yes, definitely. Looking at in another way, the
"total ink" at L*=52 is around 20% different (135% vs. 154% for
USSheetfedCoatedv2 vs GRACoL Coated1).
2) If a sheetfed printer is given an area of
50c41m41y17k, about how
much tolerance would you give him to vary from those
values on press
and still consider that he has adequate process
control?
In "G7" terms, we'd give him about +/- .02
density at the standard midtone (50/40/40) visual density of .54. This
would translate to roughly +/- 1-1.5% or a 3% maximum spread between
colors. Beyond that, you'll have gray balance issues. Personally, I'd give
the pressman about 2-3% midtone shift as long as he kept things in gray
balance. A shift in gray balance is much more noticeable than a shift in
tonality.
3) Same question, this time with respect to
performance of profiles.
Another profile prepared by another method would not
match
this one exactly. But suppose you are reasonably
satisfied with what
we have. Assuming that it has approximately the same
GCR method
(relative strength of CMY vs. K), can you say about
how much
tolerance you would have for a competing profile to
vary from
50c41m41y17k, before concluding that the other profile
was not an
adequate substitute?
I couldn't answer that unless I actually saw the
profile. Having the "correct" midtone balance is only one aspect
of a profile. If the midtone balance was "correct" but it made
totally incorrect assumptions about the colorimetry of the pure inks and
overprints, then all bets are off.
Of the things that CAN be changed on press, ink density
and, indirectly, dot gain has a certain amount of latitude. But if the ink
colorimetry is off because of pure ink colorants or ink trap/ overprint,
this effectively cannot be change by the press operator without either
opening up different cans of ink or changing the ink sequence, either of
which is not reasonable to do.
SWOP2006 Coated3 for web offset (publication) printing
on a quality
coated stock.
Fair enough This is also a good profile. Those who
currently use
SWOPv2 as a CMYK setting would be better off with this
one.
For uncoated stocks (I'm thinking "offset"
stock), I think FOGRA 29
has a lot going for it. It is very close to the
"G7" tone response
curve and the colorimetry seems about right. Until we
get a "G7"
version of an uncoated stock, FOGRA 29 should do fine.
I would like to find out about this "about
right". My test image is a
picture of a brownish tiger cat. I have selected a
certain part and
removed all color, verifying that throughout it
measures 0a0b. I have
tested this not just with FOGRA29, but with a variety
of profiles,
some intended for this uncoated condition and others
not. To wit,
SWOP v2, SWOP 2006-3, Custom CMYK coated defaults with
appropriate
dot gain change, GRACOL, U.S. Sheetfed Coated v2,
Custom CMYK
uncoated defaults with appropriate dot gain change,
U.S. Sheetfed
uncoated v2, and Euroscale Uncoated v2. Let us
consider two spots.
In the dark areas of the three-quartertones, I have a
value of
14L0a0b. FOGRA29 separates this pure neutral into
82c74m58y62k. Are
you aware of any printing conditions anywhere in the
world where this
formulation does not yield a decided purple-blue? I'm
not, and
neither, apparently, are the folks who prepared any of
the profiles
mentioned above.
Couple of things:
* You must be using a perceptual rendering or
relcol+BPC because FOGRA 29 can't get anywhere close to L*14. Maximum
"density" is around L*28. But moving on...
* What you're seeing as a "bluish" neutral is
a result of a couple of things:
* Looking at the solid ink colorimetry, the yellow ink
is very strong relative to cyan and magenta. The result in the separation
would try and compensate for this strong yellow ink by reducing the yellow
in the neutrals.
* The "black" ink for FOGRA 29 is not neutral
but is actually a bit warm/red. This could also be contributing to the
reduction in yellow in the gray balance since you've got a fair amount of
black (62%) printing at that point.
Remove the black ink and the gray balance values become
a more reasonable 82c80m74y.
It might appear whacked to you based on
"typical" gray balance but it's right for that profile if the
press is run to the same L*a*b* values for the solid ink
"density".
FOGRA 29 is a "standard" printing condition
based on ISO 12647-2 on Paper Type 4 (not to be confused with "grade
#4" paper by the way) so the separation characteristics of that
profile is correct for a press run to those printing conditions/standards.
The key number here is the relation between magenta
and yellow. Let
me give you, for each of the other eight profiles, the
magenta value
as compared to yellow. They are: +1, 0, 0, +3, 0, 0,
+2, -1.
FOGRA29, at +14, is on some other planet. Do you think
this is close
enough to be acceptable?
If this press is printing to the specs of FOGRA 29,
certainly, it just doesn't happen to be what you're expecting for typical
gray balance. Like I've shown above, remove the (warm) black ink and the
gray balance values balance out to more typical values.
Not only would one get blue shadows by separating with
this profile,
but even subdued colors are cooled down considerably.
This cat is a
dark yellowish brown, 24L5a11b, somewhat closer to
yellow than it is
to red. FOGRA29's idea of the CMYK equivalent of this
is
61c70m73y49k. That looks mighty purple to me, and to
the other
profiles. FOGRA29 has the yellow only three points
higher than the
magenta. The other eight have it: 11, 14, 12, 12, 14,
11, 11, 17.
In light of the above, a) do you consider the
variation between
FOGRA29 and all these other profiles to be significant
enough to
concern ourselves over?
FOGRA 29 is completely different "cat"
because of it being a printing condition for uncoated stock. My only beef
with with FOGRA 29 is that it assumes a much more neutral paper white (a*+1
b*-2) than I typically see for uncoated "offset" stocks in most
pressrooms. The ones I typically see have a lot of optical brighteners
resulting in b* values in the -5 to -7 range. A paper this "blue"
would require more "yellow" in the separation to maintain gray
balance so you would probably see more normal values for the yellow ink in
neutrals.
b) do you stick by the statement that "the
colorimetry seems about right,"
It's "right" because the international
standard ISO 12647-2, paper type 4, FOGRA 29 *says* it's right. Whether
it's "typical" uncoated sheetfed printing here in the USA is
another matter.
and c) do you think that people would
be justified in wanting to change these bluish shadows
to something
more standard?
I think if a person were printing here in the USA on
"typical" uncoated offset papers, they may want to hedge their
bets and possibly split the difference between "typical" gray
balance and what this profile calls for. This is not because the profile is
in any way "inappropriate" for the printing condition it
describes, it's because the paper stock "color" that this profile
assumes is perhaps incorrect for typical uncoated papers that I've seen
used for uncoated sheetfed printing.
What people need to understand is that the ink strength
or "chroma" reduction when moving from a gloss coated sheet to
something like an uncoated stock is not uniform. While ink density for cyan
and magenta can typically drop between .20-.40, yellow ink density will
only drop perhaps .10-.15 which leaves the yellow ink relatively stronger
than cyan/magenta resulting in a too-yellow gray balance if using typical
gray balance numbers. So what you're seeing with FOGRA 29 is not at all
atypical or unusual. My only beef with FOGRA 29 is the assumption they make
as far as paper color. It's unfortunate at this point that we don't have
what I would consider a profile that's appropriate for "G7"
printing conditions on uncoated offset stock. When and if that happens, I
will change my recommendation from FOGRA 29 to something else but for now
it's the best we have in terms of a true uncoated printing characterization
for sheetfed offset printing.
Regards,
Terry Wyse
___________________________________________________________________________
.
Re: Quotes Without Comment (was: Eurostandard inks)
Posted by: "Jim Donovan"
Tue Jan 27, 2009 9:16 am (PST)
Reality hits...The many,many,many were in truth one on
an island as suspected.Not even the one defended that absurd statement.This
issue has been put to bed,along with conversion at the
R.I.P...........Thanx again Dan,Jim Donovan
___________________________________________________________________________
Re: Quotes Without Comment (was: Eurostandard inks)
Posted by: "Laurentiu Todie"
Tue Jan 27, 2009 12:34 pm (PST)
How would you build process plus Pantone files, without
CMYK in Photoshop?
(I kindah know the answer, but what if I were the
printer and I knew what I was doing? : )
How would you use the Black plate for any kind of
enhancement, including but not limited to sharpening?
Some masking is easier with access to all possible
channels.
Laurentiu
___________________________________________________________________________
Re: [colortheory] Re: Eurostandard inks – anyone?
Posted by: Dan Margulis
Wed Jan 28, 2009 6:12 pm (PST)
Peter writes,
I've been following this discussion for a while now.
As some of you know,
I'm a pretty big supporter and sometimes defender of
Dan, but I'm also
someone who fully embraces and used the most modern
and best tools
I can find.
I haven't got much to add to any of this post, but it
does serve to underscore many of the points of the thread.
While there is some validity in arguing that Adobe
*should* upgrade Custom
CMYK, which is in dire need of something, it's also
fairly unrealistic to
expect them to do anything, taking into their history
of working on
that feature.
No doubt.
I feel exactly the same when it comes to being able to
edit and generate
custom profiles. Sure, it seems like a lot of extra
money, but what it
really does is allow you to do the real professional
job that is required
today. It's part of the prices of admission, as far as
I'm concerned, and
complaining while waiting for Adobe to do something
they might
never do anyway is wasted energy.
The complaints are not precipitated by waiting for
Adobe to do something that it should have done in 1998. The complaints are
because Photoshop engineers, knowing that in the lamentable way in which
matters stand, Custom CMYK is critical to many, threaten to delete it
without a replacement, and make crystal clear that their reason is to take
vengeance on their perceived enemies.
The thing that owning and using these tools -
ProfileMaker and a
Spectrolino - in my case, does, is let me know that
I'm sending files that
are the right ones for my clients printing, not one
that I think might be
sort of in the ballpark. Ballpark isn't good enough
for them, and Custom
CMYK just doesn't have the ability to compensate for
the quirks of
everyone's non-conforming direct to plate press and
proofer
calibrations.
I wouldn't sell it short, but it is certainly difficult
to learn, so going the route you did is reasonable.
I've never argued that Custom CMYK should be taken
out. I dont', but I
don't think it's the best tool anymore.
It would be hard to think that *any* tool that was last
updated in 1998 is the best tool today. It is the best tool currently
available that does not cost several times what Photoshop itself does.
I've mentioned to Chris Cox
several times that perhaps something like the Imation
CFM would work. I
had an early version of that product, and while it was
buggy, it did allow
you to change black generation and total ink of
existing profiles.
There's never been any response from him. I'm not
going to sit around
hoping there is.
Talking with Chris Cox is a waste of time. If you want
to make the proposal, it would have to go to Adobe upper management,
because the Photoshop engineers will not agree to do it unless they are
ordered to.
Not everyone needs the level of control and accuracy I
do, but if you
do, then investing in ProfileMaker or Monaco Profiler
can actually pay for
itself in increased efficiency and better results,
which mean happier
clients who hire you more often.
That's both a happy story and a sad one. You are a
professional photographer. According to what I have read elsewhere, and
correct me if I'm wrong, you realized correctly some time ago that
long-term survival in the field was in question unless you diversified,
which you proceeded to do. Now, a large amount of your income if not an
absolute majority comes from graphic services other than photography. Of
that, a significant amount of work involves shepherding client jobs through
various printers. These clients employ you because they trust you to bring
them good results and they are not interested in hearing excuses about what
the printer does or does not know about color management.
Now, the happy part. People who can do this well are in
demand. A lot of people on this list can do it, and the past decade has
been a good one for them, as I assume it has been for you. Happy clients
are valuable things and unhappy ones very expensive, so price tags on
helpful hardware and software must be taken in perspective, understanding
that not everybody would make the same decisions you did.
Now, the sad part. Why do you suppose that you have
done better than you might have a straight photographer? I think the answer
is pretty simple. A lot of people are capable of competent photography;
many of them are hungry enough to price their services very low, and in
fact some of them are hobbyists who don't need to make a living at it, but
still are strong enough to compete with professionals. NOT many people are
able to color-correct images or to insure that they get printed to the
client's liking.
One can understand why, fifteen years ago, there
weren't many such people--they would have had to buy a million dollars'
worth of equipment just to get into the field. But what's the excuse today?
The promise of ICC color management was to democratize the process, to
allow anybody to get consistent results from a variety of different output
conditions. Nothing of the kind has happened. Instead, the field is still
dominated by people who have to do a lot of studying technique and
obtaining experience in the school of hard knocks.
So, while your success is deserved, it would be nice if
it wasn't so difficult. Nice for others, that is.
Interestingly enough, I'm currently working on a CD
cover that is being
produced at DiscMakers. When I contacted them about
printing specs the
first response was that they are not color managed and
ignore all
profiles. Just send CMYK, they told me. No mention of
what flavor.
Finally they sent me a profile that was a Custom CMYK
setting of
SWOPcoated13%DG.
And one of the reasons it is so difficult for others is
that they are continually getting told to trust what the printer says, when
(as this is yet another reminder) printers ordinarily haven't the foggiest
idea,
My best guess in this case, has to be with the SWOPv2
profile.
Best or worst, it's certainly a guess, which is another
reason why what we do is so hard. In a perfect world we're not supposed to
have to guess. In this world, we *do* have to guess, and our success in the
professional often depends on how accurate our guesses are and what steps
we take to prevent disastrous results if we're wrong.
Dan Margulis
___________________________________________________________________________
What exactly does Custom CMYK need?
Posted by: "Jeremy Schultz"
Fri Jan 30, 2009 5:33 am (PST)
I?ve been following a little of the discussion on
Custom CMYK, and I want to present a list of needed Custom CMYK
improvements to those in charge at Adobe. Since I haven?t followed the
discussion closely, can those who have posted about it please recap what
they would want?
Jeremy Schultz
Design and illustration for print and the web
1502 Nine Iron Drive
West Des Moines, Iowa 50266 USA
(515) 306-4348
___________________________________________________________________________
Re: What exactly does Custom CMYK need?
Posted by: Michael Jahn
Fri Jan 30, 2009 11:53 am (PST)
Jeremy,
Besides the concept that this "ICC" approach
is basically "unfixable' - as there several things 'wrong' with the
basic idea of using LAB as a PCS from a color science standpoint...
http://iqcolour.com/pdf/presentation/IQC_workflow1.pdf
the REAL issue is not color science based that has this
"who cares if it is broken" position by Adobe...
Adobe is concerned with gaining new people as
customers. Fixing this will NOT make a large group of people who do not own
a copy of Photoshop suddenly buy one, so there is absolutely no business
case for Adobe to spend any engineering resources on this.
As someone who has been developing software solutions
and marketing them, I wish this was not the case, but from my perspective,
that has been the issue since Photoshop 5 - no real business case.
Michael Jahn
Jahn & Associates
___________________________________________________________________________
Re: What exactly does Custom CMYK need?
Posted by: "RJay Hansen"
Fri Jan 30, 2009 1:35 pm (PST)
There's no business case to be made for serving
existing customers' needs?
rjay
___________________________________________________________________________
Re: What exactly does Custom CMYK need?
Posted by: J Walton
Fri Jan 30, 2009 1:35 pm (PST)
On Fri, Jan 30, 2009 at 8:48 AM, Michael Jahn wrote:
Adobe is concerned with gaining new people as
customers.
As is every business that has ever been in existence.
Fixing this
will NOT make a large group of people who do not own a
copy of
Photoshop suddenly buy one, so there is absolutely no
business case
for Adobe to spend any engineering resources on this.
Baloney.
Every business that has ever been in existence has also
been concerned with KEEPING their current customers. Adobe does not make
their money on some sort of "monthly Photoshop user fee," they
make money when someone either buys Photoshop or UPGRADES their current
version. Microsoft, for example, isn't as worried about gaining new
customers as it is about getting their current users to pay for *future*
versions. Once you reach market saturation that's what you focus on.
That's where Adobe is right now, except that there is a
bit of room for new users (unlike Windows or Office). Coming up with
compelling reasons for current users to upgrade is VERY important to Adobe.
Certainly if there isn't a ground swell of support for
a new Custom CMYK, it will NEVER happen. I haven't heard anything about the
beta for Photoshop CS5, but if I am allowed to participate that's going to
be part of my input. But I'm just one person - Adobe's going to need
thousands of people asking for this to make any changes.
That's why it is a bad idea to discourage people from
expressing their concerns to Adobe.
J Walton
___________________________________________________________________________
.
Re: What exactly does Custom CMYK need?
Posted by: Dan Margulis
Fri Jan 30, 2009 1:35 pm (PST)
Jeremy writes,
I?ve been following a little of the discussion on
Custom
CMYK, and I want to present a list of needed Custom
CMYK
improvements to those in charge at Adobe. Since I
haven?t
followed the discussion closely, can those who have
posted
about it please recap what they would want?
This has been tried many times in the past by many
parties and has resulted only in insults from the Photoshop engineering
team, statements that you are the only person ever to have asked for such a
feature, a long lecture on why no one should use Custom CMYK and why it
should be deleted from Photoshop without a replacement, and the inevitable
offer that perhaps, if you are very, very good, we will allow you to have a
crippled version of Custom CMYK in the future.
Obviously your heart is in the right place, but you
might want to read the entirely similar post, where another list member
suggested the same thing in 2005, at
http:
//www.ledet.com/margulis/2006HTM/ACT-Asking_Adobe.htm
At that time, I recommended against doing it and told
him he was wasting his time, just as I do today. He did not agree, and went
ahead with approaching the Photoshop team. Some weeks later, he came back
with this:
Date: Thu, 18 Aug 2005 18:13:30 -0400
Subject: Possibility of changes in Photoshop (Dan
right again?)
I really don't want to believe Dan's negative
viewpoint, but [long description deleted]...I went out of my way not to
throw mud, but clearly nothing I said could have made any
difference."
We have recently seen further evidence that they still
wish to delete Custom CMYK without a replacement, let alone providing an
improved substitute, and further evidence that the motivation is not to
improve the user experience but rather to punish a group that they don't
happen to like. It is certainly correct to try to reason with someone with
whom we disagree--provided that the persons are capable of being reasoned
with. The management of the Photoshop engineering team has shown that it is
not. If you want to address the complaint to somebody, it should be to
upper management at Adobe.
Dan Margulis
___________________________________________________________________________
Re: What exactly does Custom CMYK need?
Posted by: "Stephen Ramirez"
Fri Jan 30, 2009 5:33 pm (PST)
Apparently the members of this list are not the only
ones concerned about Adobe's business model of concentrating on gaining new
customers...
Adobe Systems shares fall after analyst downgrades
Associated Press, 01.13.09, 07:19 PM EST
"Adobe Systems Inc. shares fell Tuesday after an
FBR Capital Markets analyst downgraded the maker of graphics, publishing
and design software, saying Wall Street estimates for the company are too
high. FBR analyst David Hilal lowered his rating to
"Underperform" from "Market Perform."
He said Adobe (nasdaq: ADBE - news - people ) depends
too much on new sales rather than recurring revenue from existing
customers, which could put the company's growth rate toward the bottom of
the business software sector as the recession continues to limit technology
spending."
regards,
Stephen Ramirez
___________________________________________________________________________
Re: What exactly does Custom CMYK need?
Posted by: "Jeremy Schultz"
Mon Feb 2, 2009 2:05 pm (PST)
Thanks for the remarks about Custom CMYK and especially
Dan?s link to the 2005 thread. I haven?t pursued it yet, but before I
brought it up on the list I talked with someone relatively high in the
Photoshop hierarchy about my intentions and this person wants to hear what
you say. I don?t think he?s being disingenuous.
Anyway, I am going to see the 2005 thread and go from
there.
Jeremy Schultz
___________________________________________________________________________
Re: What exactly does Custom CMYK need?
Posted by: "Rick Gordon"
Mon Feb 2, 2009 11:15 pm (PST)
I had also spoken to John Nack at MacWorld 2008, and he
sounded geniuinely interested, and even intimated that there had been work
on it. But, as we know, it didn't make it to CS4. But I do still think that
John is not just an Adobe shill.
Rick Gordon
RICK GORDON
EMERALD VALLEY GRAPHICS AND CONSULTING
___________________________________________________
___________________________________________________________________________
Re: Eurostandard inks – anyone?
Posted by: Dan Margulis
Wed Feb 4, 2009 8:36 am (PST)
Terry writes,
I take that to mean "editable in Photoshop's
Custom CMYK..." because
this profile is certainly editable in a profile editor
You should take it to mean "editable in
Photoshop".
As for the rest of the post, it's astonishing how
closely we agree, except as to the key point of what constitutes a
professionally acceptable profile. As I surmised, you are a lot more
liberal, if that's the word, in terms of what's acceptable than many of us
would be.
As for points of agreement, I asked how much tolerance
you would give a printer to vary from certain given results, and still
consider that he had adequate process control. You replied,
In "G7" terms, we'd give him about +/- .02
density at the standard
midtone (50/40/40) visual density of .54. This would
translate to
roughly +/- 1-1.5% or a 3% maximum spread between
colors. Beyond that,
you'll have gray balance issues. Personally, I'd give
the pressman
about 2-3% midtone shift as long as he kept things in
gray balance.
That's about how I see it, too. This variation is
visible--and that's considering a highly disciplined printer, spending a
lot of time trying to reduce it. Run-of-the-mill commercial printers can
have much more variation that that.
Similarly, with respect to the FOGRA29 profile which
you recommended and which I pointed out has an extremely blue shadow in
comparison to any profile for similar conditions heretofore seen, you
commented,
I think if a person were printing here in the USA on
"typical"
uncoated offset papers, they may want to hedge their
bets and possibly
split the difference between "typical" gray
balance and what this
profile calls for. This is not because the profile is
in any way
"inappropriate" for the printing condition
it describes, it's because
the paper stock "color" that this profile
assumes is perhaps incorrect
for typical uncoated papers that I've seen used for
uncoated sheetfed
printing.
The rationale may be different but the solution is the
same. This profile is radically different from those prepared in the past
for similar conditions. The sensible thing is to assume that everyone else
in the past was probably not completely messed up, and either we should
ignore this new profile or at least move it strongly in the direction of
past practice. Same deal with SWOP v.2's treatment of blues, which is
strongly different from anything seen in the past.
You also agreed with me that USSheetfedCoatedv2, the
profile that is delivered with Photoshop, is GROSSLY different in its
computation of dot gain than the profile you and I recommend.
Finally, we have a de facto agreement (because you did
not fall for my invitation to name one) that it's not really possible to
state tolerances for how far good profiles can be different from one
another, the way we can with two runs on the same press. Certain matches
are much more important than others.
As to the consequences of all this agreement, we
disagree, and I believe it points out a great deal about why it has been so
difficult to interest high-end users in politically correct color
management, as opposed to effective color management.
We are interested in results, not theory. A profile
that gets us good results from a variety of printers is a good one. One
that does not is bad. All the measurements in the world cannot save it.
The requirements for an effective CMYK profile are,
IMHO, astonishingly few:
1) Accurate assessment of dot gain.
2) The fleshtone range portrayed accurately.
3) Good handling of gray balance.
4) Greens not necessarily accurate, but cannot be too
dull.
5) Blues not necessarily accurate, but cannot be too
light or too purple.
6) Black generation cannot invite disaster in the hands
of an incompetent printer.
Pretty much any profile that meets these criteria beats
any that fall short in one or more of them. It isn't hard to make profiles
that *do* comply, whether in Custom CMYK or elsewhere, but it isn't as easy
for those without a lot of experience preparing CMYK files. Consequently,
we get profiles from printing standards organizations, or RGB-oriented
color management consultants, or from Adobe, that are seriously deficient
in one or more of the above.
From what I understand, USSheetfedCoatedv2 is simply a
profile
made from a 3M Matchprint analog/film-based proofing
system.
Which was designed to emulate web, not sheetfed
printing, and hence had a higher dot gain--but nowhere close to the
enormously excessive dot gain presumed by SheetfedCoatedv2.
Here's what's "good" about
USSheetfedCoatedv2 (in my opinion):
* Excellent gray balance characteristics. Gray balance
is within
about .5-1.0% of what most of us "G7"
advocates consider optimum for
midtone: 50/40/40, 1/4 tone: 25/19/19, 3/4 tone:
75/65/65.
* Proper total ink limit, if a bit on the high side,
of 350%. Back in
my scanner days, that was considered reasonable for
sheetfed printing
although these days you see this closer to 320%.
* GCR is on the light-to-medium side (K start of about
35%, midtone K
of about 8%, Kmax of 85%).
I haven't looked at the first and last points but I
agree that many sheetfed printers today would disapprove, or reject work
based on this profile because it exceeds their desired total ink. To solve
that routine, trivial problem, of course, the suggestion is that we should
spend $2,500 or hire a color management consultant.
Here's what's not good about USSheetfedCoatedv2:
* Nominal dot gain of around 25% which is way too high
for typical
linear CtP printing today.
It's way too high for sheetfed printing on coated paper
at any time in history.
An even bigger problem I think is the fact
that the dot gain actually peaks around 37% which
would result in 1/4
tones that print way too "full". The
resulting separation using this
profile would result in very weak/flat 1/4-to-midtone
detail/shape if
printed linear CtP.
The words "way too" as opposed to
"slightly too" deserve emphasis.
As a comparison, the "GRACoL2006_Coated1"
profile
has a nominal dot gain of roughly 15% which peaks
further up the scale
at around 45-47%. Call the difference between
USSheetfedCoatedv2 and
"GRACoL" to be around 10%....which is HUGE.
Exactly. HUGE. Both supposedly describe the same
condition. If one is OK, the other is grossly unacceptable. Dot gain is one
of the critical areas described above. Viewers have very little tolerance
for error in it. If the GRACOL profile is to be accepted as a reference,
then SheetfedCoatedv2 is at least five times further away from its midtone
than I would consider acceptable variation.
Plenty of sheetfed printers still make plates the
old-fashioned way. Those that use CTP often try to emulate the old way. Any
suggestion that this profile describes some ancient printing condition is
exploded by the fact that it was released at the same time as a twin
profile for web coated. Anybody who knows *anything* about printing knows
that other things being equal, sheetfed dot gain is significantly less than
on web. Therefore, if one profile is to serve for each, the sheetfed
profile must deliver a considerably darker separation in the midtone than
the web one does.
Instead, we get 49c38m38y7k (sheetfed, the one that's
supposed to be darker) vs. 53c44m44y10k. Any knowledgeable person looking
at these simultaneously released profiles can only conclude that one or
both is seriously defective.
So the bottom line is that profile is really meant for
a different era
in printing, the days prior to CtP printing.
No, the bottom line is that the profile is and always
has been unusable for quality- oriented sheetfed printing. The bottom line
is that anyone with more than a month of pressroom experience could take
one look at this profile and dismiss it out of hand. The bottom line is
that a beginner, using Custom CMYK with no measurements at all, could make
a more pleasing and more accurate profile than this without difficulty.
It's unfortunate at this point that
we don't have what I would consider a profile that's
appropriate for
"G7" printing conditions on uncoated offset
stock. When and if that
happens, I will change my recommendation from FOGRA 29
to something
else but for now it's the best we have in terms of a
true uncoated
printing characterization for sheetfed offset
printing.
Unfortunate indeed. After eleven years, "for now
it's the best we have."
For nine of those years, we didn't even get *that* for
uncoated papers--we got a single profile that Adobe somehow believed was
suitable for both web and sheetfed printing simultaneously. For nine years,
we got the decidedly fifth-rate SheetfedCoatedv2. For all this time, those
defending these miserable excuses for profiles have somehow had the nerve
to criticize Custom CMYK as being low-quality, obsolete. I do agree that
with a decent modern profiling package, a skilled user will be able to
achieve superior results to Custom CMYK at least part of the time. Whoever
made these profiles did not.
Also, it should be pointed out that in the last two
times this topic has hit the list, people have pretended to misunderstand
my request for something that can handle third-party profiles with at least
as much functionality as Custom CMYK did in 1998. In each case, one person
accused me of advocating putting in a full-featured $2,500 editor, while
another accused me of saying that Custom CMYK was already better than the
aforementioned $2,500 editor.
The foregoing discussion of profile issues illustrates
the type of editing that somebody serious about CMYK reproduction wants.
*Adjust dot gain and black generation to make a
conservative separation for an unknown printer.
*Adjust deep blues in SWOP v2 so that they won't turn
out so purple.
*Adjust CMY overprint in FOGRA profiles so that the
shadow won't turn out so blue.
*Change dot gain in USSheetfedCoatedv2 so that it has
some relation to reality.
*Change ink limit (sheetfed coated mentioned, but
really any profile is implicated) so that a commercial printer who
rigorously enforces ink limits will accept the job.
None of these tasks requires an expert. None requires
an instrument. All were available in Photoshop in 1992--the 1998 revision
of Custom CMYK did not add them. If the Photoshop engineers had decided in
1998 or at any time thereafter to give even a 1992 level of functionality
to third-party profiles, it would have been endorsed immediately by me and
everybody else, and we would not have had an eleven-year period of
stagnation as far as color management is concerned.
Dan Margulis
Receive Schedule
and Pricing updates.
If you let us know how to contact you, we will enter you into our monthly drawing for a free 2-day class (A $895 value).
- Get free video training
- Get free tips, tricks and shortcuts
- Get special sale pricing and free seminar invitations
Sterling Ledet & Associates, Inc. 4100 Perimeter Park South Atlanta, GA 30341
work sales@ledet.com work administration@ledet.com work webmaster@ledet.com
P work 770.414.5007 P work 877.819.2665 F fax 312.896.9094
- Authorized Training Provider
Copyright 2012. Sterling Ledet & Associates. All Rights Reserved.
Design by Digital Scientists, a Conversion Optimization Company.
