B+W in CMYK


Lee Varis
 

Hi all,

If I want to reproduce B+W photos as 4-color what do you recommend for custom CMYK setup in PS ? I've been using heavy GCR as opposed to medium or light but I'm wondering if "maximum" would be better or would it be more problematic with regards to dot gain settings, ect.. (the black plate does seem to be a little too heavy - but maybe that's a good thing for B+W images) Are there any special dot gain settings or curves that you use for magazine repro (the 'zine in question tells me to expect 25-30% dot gain and to limit black to 94% - Total Ink of 280%) In some images I have small areas of full color and I have composited those sections from a light GCR source into the overall "heavy" GCR B+W image - do you foresee any problems with this approach?


regards,

Lee Varis
varis@...
http://www.varis.com
888-964-0024


Lee Varis
 

On Monday, May 6, 2002, at 06:26 AM, Scott Olswold wrote:

The best possible way is to find/make/buy a profile based on the print
provider's device and convert to CMYK using that profile. This way, the ink
limit and black generation should work with their setup *and* you don't have
to play mucky-muck with adjusting this or that.
This is not an option - The magazine is printing using a web press. The fact that they are giving me fairly high dot gain specs suggests that they are aware of relatively poor press performance with regards to dot gain and I will tend to "believe" that they know what they are talking about ( most of the time I get a fairly lame response from a print production person like, "we don't have any dot gain" when I ask them about these things.) I don't have the time to profile their press - they don't use a profiled workflow and the press is probably too variable to get a good profile anyway. I will not have the opportunity to get a matchprint made nor do they have enough control at the print run to adjust for individual pages to match the matchprint - this is going into a consumer grade 'zine with lots of other articles and ads besides mine. - in other words this is the real world!

Now, given how you *are* doing things, what I would do is this:

1. Keep GCR set to Medium and set your TIL and Black to what's specified by
the output house.
Given that I'm separating B+W images with no color in them and that I want these "B+W photos" to look as neutral as possible, wouldn't it be better to use a heavier GCR (more black ink) to avoid possible color cast if the press varies in dot gain on any of the color plates?

2. Convert your image *first* to RGB and then move to CMYK. This will
probably introduce more tone in the CMY channels. This is good, because
we're going to play with Black.
These images are in RGB, though they are completely neutral color wise, I'm separating them to CMYK because I have color and B+W on the same page and I don't want a tepid grayscale (100% black ink only) image displayed next to a full color CMYK file that will have a 280% ink coverage for it's black point. I'm not looking for more "color" tone - I'm looking for less color, more value range

3. Observe the Black channel. You can use Levels or Curves to tone it down
just a punch more in the deeper shadows. Since you have some ink in CMY for
those areas, it won't affect your overall image that much (specifically on
'zine stock, unless your 'zine is "upper shelf" stock like Nat'l Geog.).
I think your understanding of what I need is backwards - I'm NOT looking for maximum color gamut - I'm looking for minimum color gamut and maximum value range. I think I'd like to replace more of the color ink with black - I'm just not sure how far I can go without creating problems for the maximum value range that I'm after.

I'd like to hear from anyone who might have a particular trick for dealing with this specific scenario - separating B+W images as CMYK - I won't be able to do duatones or tritones so don't bother suggesting things that rely on custom duo/tritones. Are there things I can do with UCA in combination with some other settings? Any creative ideas are welcome - I understand the basic principles of CMYK and Photoshop's custom CMYK setup. Thanks in advance!


regards,

Lee Varis
varis@...
http://www.varis.com
888-964-0024


Gary Fredrick <fredrick@...>
 

Lee Varis wrote:

This is not an option - The magazine is printing using a web press. The
fact that they are giving me fairly high dot gain specs suggests that
they are aware of relatively poor press performance with regards to dot
gain and I will tend to "believe" that they know what they are talking
about ( most of the time I get a fairly lame response from a print
production person like, "we don't have any dot gain" when I ask them
about these things.) I don't have the time to profile their press - they
don't use a profiled workflow and the press is probably too variable to
get a good profile anyway. I will not have the opportunity to get a
matchprint made nor do they have enough control at the print run to
adjust for individual pages to match the matchprint - this is going into
a consumer grade 'zine with lots of other articles and ads besides
mine. - in other words this is the real world!
at the rate you're going with this printer it seems as though you'd be
lucky if the job was in register. Do you have any dealings with a higher
quality pre-press organization? I ask them because I know the last one I
worked for in NYC had a recipe which spat out the right seps after
taking the destination into consideration.

Gary


Lee Varis
 

On Monday, May 6, 2002, at 12:13 PM, joel wrote:

just wondering...why wouldn't you just load your profile into your grayscale settings in PS, then convert to the k channel and print as grayscale? It creates a black-only plate.
Why is this so hard to understand? I don't want a black only plate! I want a CMYK file of a B+W image with a long-black plate with just enough CMY under the black to give me 280% total ink in the blackest areas and just enough everywhere else to insure smooth transitions in tone with a minimum of color-crossover-cast. I don't want a spectrophotometer measured profile - just some custom CMYK settings in Photoshop so I can create this special purpose profile for separating B+W images.


regards,

Lee Varis
varis@...
http://www.varis.com
888-964-0024


marshyswamp71 <samarsh@...>
 

Lee writes:

Given that I'm separating B+W images with no color in them and that
I
want these "B+W photos" to look as neutral as possible, wouldn't it
be
better to use a heavier GCR (more black ink) to avoid possible
color
cast if the press varies in dot gain on any of the color plates?
Sound reasoning, I agree. I have used heavy GCR in cases like these.
I just try to factor in a little more K dot gain since the K is so
high, pay special attention to the three quarter to shadow
transitions so they don't turn to mud (if this is a concern). I will
also use selective colour to pull some CMY from the K and run with a
heavier K with whatever CMY is needed to fill out the TIL.

The art directors of the magazines I edit images for always want more
contrast - and deeper blacks, even at the expense of some detail when
the job is on press (they only seem to care about the proof without
taking into account the press). The easiest way to get this is to
jack up the K higher than you would usually and to pull the
supporting shadow tints back until the TIL reading is legal.

These images are in RGB, though they are completely neutral color
wise,
I'm separating them to CMYK because I have color and B+W on the
same
page and I don't want a tepid grayscale (100% black ink only) image
displayed next to a full color CMYK file that will have a 280% ink
coverage for it's black point. I'm not looking for more "color"
tone -
I'm looking for less color, more value range
'Solid' black ink density appearance can be a major concern, where
only 100K or 70K > + CMY areas of any size appear near each other. I
would probably shoot for around 80 > 90% K and a TIL of around 260 >
280% depending on the image and how large an area the shadow was.

One cheap rag I used to supply images to stated 280% - but any
significant sized shadow seemed to have ink saturation problems with
260 TIL or higher.

I'd like to hear from anyone who might have a particular trick for
dealing with this specific scenario - separating B+W images as
CMYK - I
won't be able to do duatones or tritones so don't bother suggesting
things that rely on custom duo/tritones. Are there things I can do
with
UCA in combination with some other settings? Any creative ideas are
welcome - I understand the basic principles of CMYK and Photoshop's
custom CMYK setup. Thanks in advance!
Lee, with a high K shadow value - the supporting screens are probably
less critical apart from adding relative density and being legal when
combined with the heavier K. So by UCA tricks - rather than adding
supporting screen values in the deep shadows, I would probably be
concerned with plugging things up too much and might back off a
little on the suggested TIL by say 10%, since the heavy K is doing a
lot for visual density, which can have more impact.

Heavy GCR does not concern itself with the highlights to
quartertones, so if this is a concern for the image content...

Presuming Photoshop custom CMYK profiles (like the old tables setup):

Heavy GCR does not treat the light tonal range with any real GCR - so
you are still getting UCR style hightlights to quartertones with a
balance of CMY with no K. I have used an approach in the past where I
do a second separation to Max GCR and then blend this image as a new
layer over the original separation. The layer options blend if
sliders can be used to add the Max GCR K into only the highlight to
quarter tone range (opt/alt split the triangle slider to refine the
blends). Advanced blending options can be set to exclude CMY from
this layer blend too. You can then drop the opacity to taste...set
some fixed samplers in key points in this tonal area then toggle the
Max GCR layer on and off so you can see a visual and numbers preview
of the _minimal_ addition of K to this tonal range as some insurance.

Another method may be to simply apply a good true grayscale into each
channel of a blank CMYK document...then use some quick linear curves
to only introduce the CMY into the mid to shadow range of the image -
making sure the TIL is legal and that the support screens are
neutral. The K is the true image, but you get some manual UCA applied
in the darker areas to add density. I have not used this manual
method though...just a random thought (I guess Max GCR would be
easier if it came to this).

I do not personally use Photoshop's UCA command during separation, I
just use the selective colour command to manually play with the
shadows when needed (which is every image, so why can't it sticky
default or be action set to blacks and not defaulting every time to
reds? <g>).

Hope this helps,

Stephen Marsh.


Lee Varis
 

On Monday, May 6, 2002, at 05:46 PM, marshyswamp71 wrote:

Lee writes:

Given that I'm separating B+W images with no color in them and that I
want these "B+W photos" to look as neutral as possible, wouldn't it be
better to use a heavier GCR (more black ink) to avoid possible color
cast if the press varies in dot gain on any of the color plates?
Sound reasoning, I agree. I have used heavy GCR in cases like these.
I just try to factor in a little more K dot gain since the K is so
high, pay special attention to the three quarter to shadow
transitions so they don't turn to mud (if this is a concern). I will
also use selective colour to pull some CMY from the K and run with a
heavier K with whatever CMY is needed to fill out the TIL.

The art directors of the magazines I edit images for always want more
contrast - and deeper blacks, even at the expense of some detail when
the job is on press (they only seem to care about the proof without
taking into account the press). The easiest way to get this is to
jack up the K higher than you would usually and to pull the
supporting shadow tints back until the TIL reading is legal.
Yes, actually this is something I do most of the time as a matter of course.

... I
would probably shoot for around 80 > 90% K and a TIL of around 260 >
280% depending on the image and how large an area the shadow was.
OK...

...Heavy GCR does not treat the light tonal range with any real GCR - so
you are still getting UCR style hightlights to quartertones with a
balance of CMY with no K. I have used an approach in the past where I
do a second separation to Max GCR and then blend this image as a new
layer over the original separation. The layer options blend if
sliders can be used to add the Max GCR K into only the highlight to
quarter tone range (opt/alt split the triangle slider to refine the
blends). Advanced blending options can be set to exclude CMY from
this layer blend too. You can then drop the opacity to taste...set
some fixed samplers in key points in this tonal area then toggle the
Max GCR layer on and off so you can see a visual and numbers preview
of the _minimal_ addition of K to this tonal range as some insurance.
Yes... this was exactly what I was looking for... I might try this!

Another method may be to simply apply a good true grayscale into each
channel of a blank CMYK document...then use some quick linear curves
to only introduce the CMY into the mid to shadow range of the image -
making sure the TIL is legal and that the support screens are
neutral. The K is the true image, but you get some manual UCA applied
in the darker areas to add density. I have not used this manual
method though...just a random thought (I guess Max GCR would be
easier if it came to this).
I tried playing around with this approach but I was finding it difficult to get the density right and the tone neutral throughout the range

I do not personally use Photoshop's UCA command during separation, I
just use the selective colour command to manually play with the
shadows when needed (which is every image, so why can't it sticky
default or be action set to blacks and not defaulting every time to
reds?
LOL, yeah... I know what you mean -- this has been very helpful... I think I might try that advanced blending layer technique to extend the black into the highlight/quartertone just a little. Thanks Steve... I wonder if Dan has something to add... this seems to just about cover it!

regards,

Lee Varis
varis@...
http://www.varis.com
888-964-0024


Dan Margulis
 

Lee Varis writes,

I'd like to hear from anyone who might have a particular trick for dealing
with this specific scenario - separating B+W images as CMYK - I won't be able
to do duatones or tritones so don't bother suggesting things that rely on
custom duo/tritones. Are there things I can do with UCA in combination with
some other settings? Any creative ideas are
welcome - I understand the basic principles of CMYK and Photoshop's custom
CMYK setup. >>

Your basic take on how to do this was correct. You want Heavy GCR (not
Maximum, which would give no CMY plates at all).

So much black is needed to prevent a color shift--few printers indeed hold a
gray balance throughout with CMY inks only.

The main problem is that black dot gain is typically 4-5 points heavier than
in the other channels, and Photoshop doesn't compensate. So, if you just use
some kind of printer-supplied generaly figure, the result is apt to be mud.
Instead, go into the black dot gain curve, delete the 50% point, and raise
the 40% point until the general shape of the curve is around 4-5 points
higher than it was previously.

As for how to constitute the shadow, this doesn't vary from color images. If
there’s shadow detailing that you don't wish to lose but which isn't
critical, shoot for 80c70m70y70k. If the shadow detail is critical, go for a
very steep black in the shadows and an endpoint of 65c55m55y95k or something
like that.

Dan Margulis


thi lamaire <bobineinc@...>
 

Hi Dan and all
Dan writes in the article Allowing the color problem
to fade away " In the Curves dialog box,expose .with
command-1, the first channel's curve,in this case the
red. Hold down the mouse button, and drag the cursor
across the pitcher. A small circle appears and moves
about the curve. Note approximately where. Place the
points at both ends of the range that the pitcher
seems to fall in . And twist"
Base on that web document(probably compressed) the
base of the pitcher is near black and the highlights
of the pitcher are near whites. How do you know not to
choose the base of the pitcher and not to choose the
highlights? How did Dan decided to select the points
he selected?
Respect
Jais / Bobine


__________________________________________________
Do You Yahoo!?
Yahoo! Health - your guide to health and wellness
http://health.yahoo.com


Floyd Rominski <fromin@...>
 

I'd like to hear from anyone who might have a particular trick for
dealing with this specific scenario - separating B+W images as CMYK - I
won't be able to do duatones or tritones so don't bother suggesting
things that rely on custom duo/tritones. Are there things I can do with
UCA in combination with some other settings? Any creative ideas are
welcome - I understand the basic principles of CMYK and Photoshop's
custom CMYK setup. Thanks in advance!

regards,
Lee Varis
---------------------------------------

Lee

a usable Sep Set Up is as follows:

CMYK SetUp:
Dot Gain: Curves, enter curves ala Dan M

Sep Options: GCR 100% > 91%
90% > 78%
50% > 45%
Put Tot'l Ink limit where you need it


A much better method is as follows.
This will give you identical Grayscale opacities as your b/w image,
5% b/w = 5% 4c > 100% b/w = 100% 4c

This has been successful for me-

Having your b/w document document open,
File: New, gives new doc. w/same size/res as orig.
Choose Mode: cmyk color
Contents: white

Now back to b/w document,
'Select All'
'Copy'
Then back to new document
'Select' C channel 'Paste'
repeat for MYK

This give a very dark muddy image, not to worry.
You then open a 'Curve AdjustmntLayer' and load curves i can send you if you
want to explore beyond the first suggested Sep SetUp above.

i've done a few of them, to give neutral, cool, and warm b/w 4c's

regards,
Floyd Rominski

fromin@...