Publisher and PDF

[Neophyte]

When my commercial printer converts my Publisher file to a PDF file before
printing, errors in my publication always occur. For example, a blue bar in
Publisher will print out as a purple bar or a checkbox in Publisher will
print out as a squiggly line. How could I prevent this from happening?

 

[JoAnn Paules [MSFT MVP]]

Convert the file yourself?

 

[Terje Martinsen]

Blue turning into purple is a "normal" behaviour as printing a clear blue
color is difficult in CMYK (which is used by commercial printing).
If you had your monitor calibrated, and also use Publisher (2003 only!) in
CMYK mode, the output would look much more like what's on your screen.
The check box question I hav no answer to.

Regards
Terje

 

[Mac Townsend]

Blue turning into purple is a "normal" behaviour as printing a clear
blue color is difficult in CMYK (which is used by commercial printing).

not difficult, impossible.

But not entirely because of cmyk. Inkjets and lasers all use cmyk inks/
toners, and many of them can do an excellent (if not perfect) job of
reproducing "computer blue"

However, the colorants used in these devices won't work in offset ink.
And the colorants "we" have are not able to work the way they are
supposed to. For example, cyan is supposed to absorb ALL red light and
reflect only green and blue (thus it has a "greenish-blue" appearance).
Similarly, Magenta is supposed to reflect only green, etc.

The pigments that we have which can be compounded into offset inks,
however, can't do that thus cyan ink instead of absorbing all red,
reflects some of it and instead of reflecting all green and blue, it
absorbes a little of it. So the light reflected is not "pure". Ditto for
Magenta.

Now...to get a "true" blue one would have to absorb all red and all
green and refelct all blue coming from white light. Combining magneta
and cyan should do this, in theory.

The theory works better with toners and some inkjet inks because their
colors are cleaner or purer than can be had with pigments suitable for
use in offset inks.

blues are not the only colors that are truoblesome with process inks:
oranges and browns are also a problem. Very few of the Pantone pallette
can be reproduced well with present offset inks.

 

[Mac Townsend]

on always occur. For example, a blue bar in
Publisher will print out as a purple bar or a checkbox in Publisher
will print out as a squiggly line. How could I prevent this from
happening?

the checkbox problem is a font problem. The needed font is not available
to the application at the time is is "converted" or printed.

Are you providing the necessary fonts along with the file? If providing
a PNG file, then if you are using a Type 1 or PostScript font for this
checkbox it will not be included and you will need to provide both
components separately.

The color issue is explained elsewhere. You can't get "blue" except by
using spot color (Pantone) and this is a whole different sort of thing.
Be happy with purple or use something else.

 

[Mac Townsend]

Blue = Purple.

That's the way it is.

Why?

Because of the way offset printing inks can render color.

I don't know how brief I can be here. Red-Green-Blue (RGB) are the primary
colors of light. These are the colors you see on your screen or TV. You are
looking at light being shined (shone?) right into your face

When you are looking at an image on paper you are seeing red-green-blue
light as it is REFLECTED from that paper, not as it is beamed directly into
your eye. We call the colors that are necessary to create this reflective
"color space" Cyan, Magenta, and Yellow". Theoretically Cyan ink absorbs all
the red light that falls upon it refledcting only the green and blue (which
is why it looks like a greenish blue), Magenta absorbs all of the green
light, and reflects all the red and blue, while yellow absorbs all the blue
and reflects all red and green.

Theoretically this works perfectly. The new CBS Reality Show called "Color
and Friends" (CBS beinbg the Color Broadcasting Serveice, of course; few
cable systems carry it<G>) however manages to point out the fallacies in the
system.

One of these is the brightness/whiteness of the substrate (paper) upon which
the ink is printed. Because ALL paper is dimmer than white light, the colors
will be subdued from what one might expect comparing to a monitor. Because
paper has additives that themselves have specific and inspecific color
absorbtion characteristics, even a blue ink printed on different papers will
appear different. Even with the same paper, the ink film thickness will
affect the reflected color appearance. Then imagine a colored paper as the
substrate. So there's that variance and it must be considered. Sometimes it
is really important.

Then there is the main issue. "Computer blue" becomes purple off the press.
Arghhhh! Yup, that's what it does. Here's where we get into more technical
stuff. (if you haven't stopped reading by now, which most people do because
they really don't care and don't want to know!

Different printing devices usse different colorants--pigments--with a large
part of the choice depending on the "stuff" the colorant has to work with.
In the case of inkjet inks, most of which are water-based dyes, we have
found coloring agents that work pretty close to the "theoretical" in that
the cyan dye does a pretty good job of absorbing red light, etc. Laser
toners are also fairly good, but not as good nor as stable as good inkjet
colorants, and cyan toner does a good job of absorbing red and magneta does
a good job of absorbing green, etc. Sigh!

But colorants/pigments suitable for use in the oil-based inks NECESSARY for
offset printing (colorants for dye inkjet or laser toner are not useable,
just like a juicy hamburger's grease content isn't helpful for lubricating
an axle bearing.). The colorants that WILL work for offset are deficient and
do not meet the theoretical requirements. Meaning that Cyan does not absorb
all red light, it reflects some (can't recall the figures but 18% "failure"
stick in my mind, right or wrong) AND magenta is at least 15-20% deficient
in absorbing all green, so some of that gets reflected.

Bottom line. rgb Blue is cmyk Purple (100% c 100%M) off a printing press.
Nothing you or your printer can really do to change that fact. If you find
that your inket or desktop colro laser gives you the nice blue you want, you
will have to repro from that device. Most commercial devices, such as
high-end docutechs, are designed to emulate offset, so you'll get purple.

You can't do anything about the weather, you can't do anything about offset
ink formulations. You you bet your ass (and everything attached to it!) that
the ink companies would like nothing more than to have a "true" cyan,
magenta, and yellow (even yellow reflects some of the blue it is supposed to
absorb)

Too much? Sorry. Life is complicated. And MS Publisher can't make it simple.

 

[Alastair M. Robinson]

Hi,

You can't do anything about the weather, you can't do anything about offset
ink formulations. You you bet your ass (and everything attached to it!) that
the ink companies would like nothing more than to have a "true" cyan,
magenta, and yellow (even yellow reflects some of the blue it is supposed to
absorb)

You may be thinking at this point "Isn't this precisely the type of
problem ICC profiles are supposed to solve?"

The answer is yes, but even using ICC profiles the Blue Turns Purple
problem persists because of the way gamut mapping is done...

For the reasons Mac has explained, a good saturated blue is impossible
to reproduce on a press. Ideally one would substitute a less saturated
blue with the same hue, and ICC profiles generally attempt to do just
that. Unfortunately the gamut mapping is usually performed in the Lab
colour space under the assumption that colours along a line of constant
angle share the same hue. As the diagram on this page shows, this is
not the case. For most shades it's a good enough approximation, but not
for blues.

http://www.brucelindbloom.com/index.html?MunsellCalcHelp.html#BluePurple

All the best,

 

[Mike Koewler]

not difficult, impossible.

I beg to differ, most informed and highly intelligent sir.

I have printed newspapers with very vibrant blues. Not within the
context of CMYK, but with the use of blue ink. I would suggest that if
using blue is a requirement for the OP, he/she includes a spot color and
has the printer use blue for it. Of course, the cost is going to be much
higher.

Mike

 

[Mac Townsend]

differ away.

I was not referring to spot color, but to the conversion of "computer
blue" which is RGB to cmyk.