Hi All,

I'm having a problem understanding how laser printers deal with Pantone
color.

As I understand it, the printer will need to convert the Pantone color
to what it think is the respective CMYK values and create the closest
match.

What I don't understand is what this conversion logic is based on. I
have a PDF that has a logo made in Pantone 321 C. When I print it
locally on a laser printer I get a completely different color in
comparison to another print vendor that is printing the same PDF.

Could someone please point me in the direction if enlightenment, as
there is clearly a piece of the puzzle I'm missing here.

Thanks
Warrick FitzGerald

From:	Waldo
Subject:	Re: Pantone color through laser printers
Date:	Fri, 21 Jan 2005 08:24:28 +0100

Warrick FitzGerald wrote:
> Hi All,
>
> I'm having a problem understanding how laser printers deal with Pantone
> color.
>
> As I understand it, the printer will need to convert the Pantone color
> to what it think is the respective CMYK values and create the closest
> match.
>
> What I don't understand is what this conversion logic is based on. I
> have a PDF that has a logo made in Pantone 321 C. When I print it
> locally on a laser printer I get a completely different color in
> comparison to another print vendor that is printing the same PDF.
>
> Could someone please point me in the direction if enlightenment, as
> there is clearly a piece of the puzzle I'm missing here.
>
> Thanks
> Warrick FitzGerald

The logic is very easy, two cases:

1) A "high-end" RIP will replace spot colors by CMYK values. Tools are
supplied to edit these values.

2) A low-end RIP will take the "alternative" color that is also in the
PDF. Every used spotcolor must have an alternative color (e.g. CMYK
values) in order to be compatible with devices that don't support spot
colors.

Waldo

From:	John Doherty
Subject:	Re: Pantone color through laser printers
Date:	Fri, 21 Jan 2005 01:58:00 -0600

In article <1106292279.762069@news-ext.oce.nl>, Waldo wrote:

> The logic is very easy, two cases:

> 1) A "high-end" RIP will replace spot colors by CMYK values.

FWIW, as far as I can tell, the term "high-end" is so poorly defined
as to be nearly meaningless.

For example, I have some Canon color copier/printers. Are they
"high-end"? Not really -- not by a long shot, from what I can guess
about what people mean when they say "high-end," anyway. But then,
when people say that, I can't usually even tell what they mean.

On the other hand, they do have licensed Pantone color libraries.
Every color printer I've ever used has had those (which is no
coincidence -- we buy/lease/use printers like that on purpose).

--

From:	Elmo P. Shagnasty
Subject:	Re: Pantone color through laser printers
Date:	Fri, 21 Jan 2005 06:15:57 -0500

In article <3639694.5m4Kl1SFyu@dupree.null.not>,
John Doherty wrote:

> > 1) A "high-end" RIP will replace spot colors by CMYK values.
>
> FWIW, as far as I can tell, the term "high-end" is so poorly defined
> as to be nearly meaningless.
>
> For example, I have some Canon color copier/printers. Are they
> "high-end"? Not really -- not by a long shot, from what I can guess
> about what people mean when they say "high-end," anyway.

He didn't say anything about the engine. He merely said that a high-end
RIP will replace spot colors with machine-specific CMYK values.

So Canon put a high end RIP onto a low end engine. That's their
prerogative.

> On the other hand, they do have licensed Pantone color libraries.
> Every color printer I've ever used has had those (which is no
> coincidence -- we buy/lease/use printers like that on purpose).

So every color printer you've ever used has a high end RIP on it.

The $299 color lasers at Staples don't have such capabilities.

From:	John Doherty
Subject:	Re: Pantone color through laser printers
Date:	Fri, 21 Jan 2005 13:47:01 -0600

In article , "Elmo P.
Shagnasty" wrote:

> In article <3639694.5m4Kl1SFyu@dupree.null.not>,
> John Doherty wrote:
>
> > > 1) A "high-end" RIP will replace spot colors by CMYK values.
> >
> > FWIW, as far as I can tell, the term "high-end" is so poorly defined
> > as to be nearly meaningless.
> >
> > For example, I have some Canon color copier/printers. Are they
> > "high-end"? Not really -- not by a long shot, from what I can guess
> > about what people mean when they say "high-end," anyway.
>
> He didn't say anything about the engine. He merely said that a high-end
> RIP will replace spot colors with machine-specific CMYK values.
>
> So Canon put a high end RIP onto a low end engine. That's their
> prerogative.

So what does "high-end" mean, exactly?

--

From:	Elmo P. Shagnasty
Subject:	Re: Pantone color through laser printers
Date:	Fri, 21 Jan 2005 15:43:14 -0500

In article ,
jdoherty@nowhere.null.not (John Doherty) wrote:

> So what does "high-end" mean, exactly?

Well, everyone has a different concept of "high end". However, in
general, if you see one with spot color lookup tables--especially if the
RIP gives the user to alter those tables to the user's
satisfaction--then it's probably high end.

From:	John Doherty
Subject:	Re: Pantone color through laser printers
Date:	Fri, 21 Jan 2005 14:56:49 -0600

In article , "Elmo P.
Shagnasty" wrote:

> In article ,
> jdoherty@nowhere.null.not (John Doherty) wrote:
>
> > So what does "high-end" mean, exactly?
>
> Well, everyone has a different concept of "high end".

Right. In other words, the term is so ill-defined as to be essentially
meaningless.

--

From:	Warrick FitzGerald
Subject:	Re: Pantone color through laser printers
Date:	Fri, 21 Jan 2005 22:55:47 GMT

John Doherty wrote:
> In article , "Elmo P.
> Shagnasty" wrote:
>
>
>>In article ,
>> jdoherty@nowhere.null.not (John Doherty) wrote:
>>
>>
>>>So what does "high-end" mean, exactly?
>>
>>Well, everyone has a different concept of "high end".
>
>
> Right. In other words, the term is so ill-defined as to be essentially
> meaningless.
>
> --

Well fist off, thank you all so much for your help thus far. I think I'm
finally starting to get the big picture here. And although I don't know
enough about RIPs to contribute to this part of the discussion I'll
explain what I've discovered out the process thus far.

It turns out that the vendor who's printing these spot colors is using a
toner-based variable digital press (Xerox DocuColor 2060). The
colors are all screen-mixed using CMYK.

I phoned the print vendor (with my new found knowledge) and asked how he
was doing the print runs. He explained that they were printing
directly from Quark 6.5.

So here's what I don't understand. Surly there is no way that Quark
knows what mix of CMYK on the Xerox DocuColor 2060 gives the closest
match to a specific Pantone color. Unless of course the printer driver
tell the application this of course.

What I have been told thus far is that there is a "process equivalent".
Which seems to for all intensive purposes point to a color space like
CMYK. As DeviceCMYK is device dependent the specific C,M,Y,K equivelant
values can not be stored in the PDF itself, as only the device (or some
peice of software alongside the device) knows what the best mix of CMYK
is to achive the specific color match is.

So I though the entire point of a RIP was to take your art from a
generic format like PDF into a format that is perfect for your print
device (and do all the check at the same time of course). So isn't
printing streight from quark onto a high end printer like a Xerox
DocuColor 2060 a bit of a silly idea, since Quark knows nothing about
the device?

Thanks again
Warrick

From:	Elmo P. Shagnasty
Subject:	Re: Pantone color through laser printers
Date:	Fri, 21 Jan 2005 20:29:38 -0500

In article <41F18872.7000708@crtman.com>,
Warrick FitzGerald wrote:

> So here's what I don't understand. Surly there is no way that Quark
> knows what mix of CMYK on the Xerox DocuColor 2060 gives the closest
> match to a specific Pantone color. Unless of course the printer driver
> tell the application this of course.

No and no.

Quark knows nothing except what it's programmed for. And what it's
programmed for is the Pantone "process equivalents" that Pantone has
defined, using Pantone CMYK inks on a press.

THOSE are the CMYK values it sends down for that color, right after it
sends down the Pantone number directly.

It is up to the RIP in question to decide whether or not to use the
Pantone number and its built in lookup table, or not and use the CMYK
values that Quark sent down. The RIP operator sets that setting. (All
RIPs for the 2060 have a Pantone lookup table to call out
device-specific CMYK mixes that are programmed into the printing system
by the system engineers.) Again, the CMYK values that Quark sent down
are entirely unrelated to the color gamut of the toners on the 2060.

Quark knows nothing about how the device itself handles Pantone colors.
Quark just does what Quark does: "OK, here's the Pantone call out, now
here's the CMYK equivalents that Quark engineers programmed into me,
that came from Pantone assuming you're using Pantone CMYK inks in a SWOP
environment on real presses."

> What I have been told thus far is that there is a "process equivalent".
> Which seems to for all intensive purposes point to a color space like
> CMYK.

(that would be, "for all intents and purposes")

Process printing *IS* CMYK. Pantone colors, though, are NOT CMYK
colors. Pantone colors are NOT made by combining the regular
cyan/magenta/yellow/black inks on a press, whether it's a regular press
or a digital press. Pantone colors can be reproduced ONLY by laying
down Pantone inks. You can do this on a regular press, with a printing
station dedicated to that specific ink. The 2060 in question does NOT
have the capability to lay down a Pantone ink, therefore it cannot print
a Pantone color in the strictest sense. It has to SIMULATE the Pantone
color as best it can by using its CMYK inks--in other words, it prints a
"process equivalent."

Quark sends down the Pantone number (for if you're using a regular press
and you want a plate for the Pantone ink itself) and also its own
programmed "process equivalent" CMYK formula, which is not device
specific and which is based on what Pantone says a process equivalent
would be. This process equivalent number that's in the print stream is
there just in case the printer chooses not to use a separate printing
tower for the Pantone color (that can get expensive), but instead
chooses to print the "process equivalent" using the same CMYK inks he's
already using on the press (a cheaper method, but it's not really the
Pantone color).

The 2060 has an in-between solution. In the end, it can only simulate
the Pantone color with its CMYK inks; however, its CMYK inks have a MUCH
BROADER color gamut than regular press CMYK inks, and therefore they
stand a much better chance of simulating the Pantone color. So the 2060
system has the lookup tables in it, and when the system sees a Pantone
number come down, it ignores the Quark-generated CMYK process equivalent
formula and substitutes its own, 2060-specific CMYK process equivalent
formula.

That's not to say the 2060 formula will be able to match the Pantone
color itself; it's just that it stands a better chance of doing so than
the Quark-generated process equivalent formula. Some Pantone colors
cannot be represented by ANY mix of CMYK inks in existence today.

Let me repeat that: Some Pantone colors cannot be represented by ANY
mix of CMYK inks in existence today.

> As DeviceCMYK is device dependent the specific C,M,Y,K equivelant
> values can not be stored in the PDF itself, as only the device (or some
> peice of software alongside the device) knows what the best mix of CMYK
> is to achive the specific color match is.

Ah, but the PDF or Postscript stream DOES contain CMYK values.

You are 100% correct in saying that CMYK is device-dependent. You are
also correct when you imply that a given CMYK formula printed on one
device will look different when printed on another device.

That's why it's important to know that when Quark sends down its CMYK
process equivalent numbers, Quark is simply using the numbers that
Pantone supplied them--numbers that ASSUME a regular press, using
Pantone CMYK inks, probably in a SWOP environment.

And that's EXACTLY why the 2060 intercepts the Pantone call, throws out
Quark's CMYK process equivalent formula, and uses its own lookup tables
to come up with device-specific CMYK formulas for the 2060 system.

And the user can turn that off, if the user wants the 2060 not to be as
accurate about representing Pantone colors. In that case the 2060
system will ignore the Pantone call and will use the Quark-generated
CMYK process equivalent. This is a good way to proof or simulate what
happens on the regular press when the print shop chooses not to use a
printing tower for the Pantone spot color itself (again, which is more
expensive and some people don't want to pay for it).

> So I though the entire point of a RIP was to take your art from a
> generic format like PDF into a format that is perfect for your print
> device (and do all the check at the same time of course).

In general, yes.

> So isn't
> printing streight from quark onto a high end printer like a Xerox
> DocuColor 2060 a bit of a silly idea, since Quark knows nothing about
> the device?

Quark knows nothing about ANY device.

By your statement above, it's silly to print straight from Quark to
anything.

No, Quark does not know anything about any given printing system.

You either have to tell it, or you have to manage your images to be
proper for the specific output device, or you leave your images in some
non-device-specific color space (L*a*b*, for example) or you leave your
images in RGB, and then you let the device in question convert the
images to its own device-specific CMYK space.

So in other words, I'm not sure what you think you're saying by that
comment above. Care to elaborate?

From:	John Doherty
Subject:	Re: Pantone color through laser printers
Date:	Fri, 21 Jan 2005 21:12:19 -0600

In article <41F18872.7000708@crtman.com>, Warrick FitzGerald wrote:

> What I have been told thus far is that there is a "process
> equivalent".

In a PostScript file that uses spot colors, you will find comments
similar to these:

%%DocumentCustomColors: (PANTONE 294 CVC)
%%CMYKCustomColor: 1 0.56 0 0.18 (PANTONE 294 CVC)

From that, we can tell that the color named "PANTONE 294 CVC" is
(roughly) equivalent to a CMYK color of 100-56-0-18. The values for
the CMYK components typically come from a color library licensed from
Pantone by the vendor of the application that created the PostScript
and are meant to represent the color as closely as possible using
conventional CMYK inks.

An application that processes this PostScript file can use those
values to render the color in CMYK, as when, for example, converting
spot colors to process during color separation. And those are the
values that will often be used by a color printer rendering the color
in CMYK.

But in the case of a printer that has licensed Pantone matching, what
must happen is that the printer recognizes the name "PANTONE 294 CVC"
and substitutes its own values for those in the %%CMYKCustomColor:
line.

Presumably, the values it has stored in its color libraries are
calibrated to represent the color as closely as possible using the
printer's own colorants, which are different than conventional
process inks. So the values used by that printer may differ from
those found in the %%CMYKCustomColor: line, and different printers
with different colorants may well have different values in their own
color libraries.

That's basically a gloss on the issue and overlooks a lot of detail,
but it's the general idea. And so you can see how the application
that created the PostScript need not know anything about the
printer's colorants or how they can best represent a particular
Pantone color. The information as to how that color is best
represented on that particular printer is stored in the printer
itself.

--

From:	Elmo P. Shagnasty
Subject:	Re: Pantone color through laser printers
Date:	Thu, 20 Jan 2005 14:11:42 -0500

In article ,
Warrick FitzGerald wrote:

> I'm having a problem understanding how laser printers deal with Pantone
> color.

They all deal with it differently. Some better, some worse.

> As I understand it, the printer will need to convert the Pantone color
> to what it think is the respective CMYK values and create the closest
> match.

Um, not always. Good, high end printers will read the Pantone number in
the print stream and then go to an internal lookup table to see what
printer-specific CMYK values most closely approximate that Pantone color
(according to the printer manufacturer). Printers that don't have such
a feature simply ignore the Pantone callout and go on to see the
application-delivered CMYK values in the print stream, and use those.
(You were aware, right, that everytime and application sends down a spot
color as part of a composite print, it also sends down what the
application thinks are approxmating CMYK values--right? Both pieces of
information are available to the printer to use or not, however it's
designed to work.)

> What I don't understand is what this conversion logic is based on. I
> have a PDF that has a logo made in Pantone 321 C. When I print it
> locally on a laser printer I get a completely different color in
> comparison to another print vendor that is printing the same PDF.
>
> Could someone please point me in the direction if enlightenment, as
> there is clearly a piece of the puzzle I'm missing here.

You simply have to know the capabilities of the printer in question.
When you send Pantone 321C out of Quark, it simultaneously tells the
print stream some CMYK value that may or may not be of any use to the
laser printer in question. If the printer uses that CMYK mix, you'll
get that printer's rendering of that CMYK mix. If you go to another
printer, you'll get THAT printer's rendering of that CMYK mix (all
depending on how you set your color processing attributes). Given that
CMYK is a device-specific color space, it's obvious that two different
laser printers will print the same mix differently.

And consider this: the CMYK mix that's programmed into Quark as being
representative of that Pantone color? It's based on the Pantone process
simulation book, which offers the companion CMYK mix assuming that
you're using Pantone process inks on a press. These CMYK values have
absolutely nothing to do with your printing on a laser printer.

In other words, you shouldn't be using Pantone spot colors to print to a
process device. Pantone colors are specific colors, created by specific
inks. Each spot color is its own ink. You *may* approximate the ink
color by using the CMYK process, but more likely you won't. What you
need to do is NOT specify Pantone colors in your document, but rather
CMYK colors.

And to do that well, you need to print out a CMYK chart for reference.
I have one; it's 441 pages, 11x11. It's a great reference chart. Print
it out on any given printer, then you can look for the color you want
and KNOW what CMYK mix to define in Quark (or whatever) to get that
color.

From:	Aandi Inston
Subject:	Re: Pantone color through laser printers
Date:	Thu, 20 Jan 2005 18:49:52 GMT

Warrick FitzGerald wrote:

>I'm having a problem understanding how laser printers deal with Pantone
>color.
>
>As I understand it, the printer will need to convert the Pantone color
>to what it think is the respective CMYK values and create the closest
>match.

No. That does exist but is very rare.

In a PostScript file (or PDF), every spot colour has (to oversimplify)
both
1. A name and
2. A process equivalent.

So, when you make separations, the name is used. When you do ANYTHING
else (laser print, proof, preview) the process equivalent is used.

So the printer has no built in knowledge of the Pantone (or any other
system of) spot colours.
>
>What I don't understand is what this conversion logic is based on. I
>have a PDF that has a logo made in Pantone 321 C.

Which will have a specific process equivalent in the file. (Or maybe
Lab).

>When I print it
>locally on a laser printer I get a completely different color in
>comparison to another print vendor that is printing the same PDF.

Different profiles. Or, they have a printer which HAS got the licensed
Pantone library.
----------------------------------------
Aandi Inston quite@dial.pipex.com http://www.quite.com
Please support usenet! Post replies and follow-ups, don't e-mail them.