What does "Intent" refer to in the colour settings of Acrobat? 1

[junglist]

I'm not familiar with this term, so what does it mean and what are the implications of the various options this function gives? (defualt, perceptual... etc.)

I'm having some problems with how monotone images appear on screen and print once they have been distilled, and think that the solution lies within the colour settings of acrobat.

Thanks in advance.

 

[tgreer]

"Rendering intents" are an aspect of CIE-based colors. I can't explain CIE-based colors here in full. It's just another way of specifying color, for example: RGB defines colors in terms of light of specified wavelengths. CMYK defines colors in terms of theoretically "perfect" inks. CIE defines colors according to an international standard, and defines colors based on human visual perception.

What you end up with is a mathematical description of a color that isn't immediately useful to a printer, press, or screen display. The CIE-based color must be "rendered down" into a "real" colorspace. Since much can be lost in such interpretation, CIE-based colors need a "rendering intent" so that good choices can be made.

Intents mainly affect how in-gamut and out-of-gamut colors are rendered. The image may contain for example, a red that your printer can print, but also a very bright yellow that it cannot. How will the yellow be printed? Will it just leave the red alone but "adjust" the yellow? Or will it adjust the level and then adjust the red so that the relative difference between the colors is maintained?

That's what intents do, manage what is know as "colorimetric accuracy".

The choices are -

AbsoluteColorimetric: colors are represented only with respect to the LIGHT SOURCE, not to medium. For example, if you print on grey paper vs. pure white paper, so what, the light source is the same. No color adjustment is made. In-gamut colors are reproduced example, out-of-gamut colors are mapped to the nearest in-gamut color.

RelativeColorimetric: Colors are reproduced with respect to the LIGHT SOURCE and the MEDIUM. Otherwise, color gamuts are handled the same as above.

Saturation: here the relative values of colors are taken into account... everything is based on the "saturation" of colors, so in-gamut colors may not be reproduced exactly.

Perceptual: Colors are modified as necessary to provide a "pleasing" appearance. Yep, that's subjective.

Rules of Thumb: Use AbsoluteColorimetric for logos, solid colors. Use RelativeColorimetric for vector graphics, line art. Saturation for business graphics, and Perceptual for photos, scanned images.

Hope this made some sense!



Thomas D. Greer

http://www.tgreer.com

Providing PostScript & PDF
Training, Development & Consulting

 

[tgreer]

When you say "monotone images" are you referring to "black and white" scanned/photographic images?

If so, you could try "Leave Color Unchanged" with Intent set to "Perceptual".

Depending on how you produce the source file, their might already be color management profiles in the source file.

You can also, for PDFs intended primarily for on-screen viewing, select "Convert All Colors to sRGB", with perceptual intent, and load a specific RGB color profile in the "Working Spaces" section.

Thomas D. Greer

http://www.tgreer.com

Providing PostScript & PDF
Training, Development & Consulting

 

[junglist]

"When you say "monotone images" are you referring to "black and white" scanned/photographic images?"

No, it is a one colour Pantone image.

Thanks for your very in depth response. I have heard the term CIE colours before, but it is something that I do not deal with on a day-to-day basis, hence it has remained one of those "mystery" terms. Until today.

Some questions:

1.) What do you mean by the term "business graphics"?

2.) I think I may have misunderstood the distinction between Absolute Colormetric and Relative Colourmetric. How does Distiller know what the light source and medium is going to be and therefore adjust accordingly?

3.) Also, is there any reason you would use the "default" intent? This is the setting I've used in the past, as I did not want to choose settings which I was unfamiliar with. (those which you have outlined)

Thanks once again for sharing your considerable knowledge on this topic.

 

[tgreer]

Most of my knowledge comes from the PostScript Language reference, and is just that: knowledge. Knowledge is a different thing than "understanding".

Distiller would know about the media by the use of color profiles. Look in your list... you'll see a lot of monitors. Have you noticed how when monitors display "white" it is actually slightly blue? Ever catch a peek at someone's TV from the street? The overwhelming impression (human visual perception!) is "blue". Distiller tries to adjust for that when you load a color profile and use RelativeColormetric intent. The adustments will vary depending on the color profile.

AbsoluteColormetric doesn't care. It will use the exact same color values, mixtures of pigment or light, regardless if you're printing or displaying or whatever.

Business Graphics would be charts, etc.

So what's going wrong, with your spot color? Does it look one way when you print, another on-screen? Which is worse? To what are you comparing it? Is your monitor calibrated?

I've always just "left color unchanged".



Thomas D. Greer

http://www.tgreer.com

Providing PostScript & PDF
Training, Development & Consulting

 

[junglist]

"So what's going wrong, with your spot color? Does it look one way when you print, another on-screen? Which is worse? To what are you comparing it? Is your monitor calibrated?"

Firstly, no, my monitor isn't calibrated. As I don't do much work which requires high-end photo-imaging (hence pin-point colour accuracy), I usually work in Pantones, or with the assistance of a CMYK chart to help me anticipate how things will print.

With regards to what's going wrong with the monotone image, here goes: I placed a monotone (pantone) image in Quark, then saved that page as an .eps. Using Distiller, I open the .eps and then distill a PDF.

When I view the PDF on screen, the colour of the placed monotone image looks a bit pale, compared to how it should look.

However, the major problem is that when I got to print the PDF, the colours of the monotone image inverse (ie. negative space assumes the colour of positive space, and vice versa)
The rest of the page is unaffected and looks fine.

I've since got around this problem by colorising the image as a greyscale tif in Quark, and then proceeding to make an .eps then PDF.

But this still doesn't explain why the colours would inverse when printing the original PDF with the monotone image. I'm baffled!

Any ideas?

Thanks once again for you time, you've been very helpful.

 

[tgreer]

Wow. Well, where to start. There are several points in your process where things can go wrong:

1) Quark to EPS. Look at all of your EPS settings here. Quark might be trying to convert the Pantone to CMYK or RGB, it might be putting in color management profiles... lots can go wrong right here at the outset.

2) EPS to PDF through Distiller. These are the color settings we've been discussing. If all is done correctly in step 1, then you don't need to do anything special here, just use "default" and "leave color unchanged"... in other words, Distiller should just keep its hands off your color!

3) Printing... I don't know your printer or RIP, but things could go wrong here, too.

Lastly, what is the source of the image? There could be problems with it itself, prior to it being placed in Quark. Test it this way: if it's an EPS, Distill it to PDF and print it... run the image by itself through your process. If it still has problems, we need to look at the image itself. You can email it to me (my email is on my website) or post a link so I can retrieve it.



Thomas D. Greer

http://www.tgreer.com

Providing PostScript & PDF
Training, Development & Consulting

 

[junglist]

OK, I've tried distilling the original photoshop .eps image directly through Distiller, with settings on default, and I still get a paler image once PDF'd.

At least, that eliminates the possibility of the problem being in Quark, or the creation of the .eps from Quark.

On closer inspection of the final PDF image, only the areas of 100% colour have gone paler, wheras the tones between 100-0% have become shades of light blue. (the original pantone colour is brown). This is very strange indeed.

I've tried almost every single variation of colour options in Distillers job settings, and I still get the same result.

With regards to the actual project I'm working on, I've found a way around this problem (in Quark).

Nonetheless, I never knew that Acrobat did such things to monotone images, and am still very puzzled. (particularly about the blueish tint in the monotone PDF! Huh?) Still curious to know if any one can shed some light on this strange phenomenon!

Thanks for all your help once again.

 

[tgreer]

Could you post a link to the file in question? Or email it to me? I'd like to see how Photoshop has defined the Pantone color.

Which version of Photoshop? Have you updated the Pantone "Books" that ship with Photoshop?

PostScript has a method of defining spot colors called the "Separation ColorSpace". There is also a "/DeviceN" colorspace. Both of these work essentially the same way: first, the color has a name, such as "Pantone 456 CV". Then there is a formula for creating that color in an alternative colorspace, for example CMYK, for devices that do not have knowledge of this named color.

What can happen sometimes, if the PostScript/PDF producer has an older or different set of Pantone colors than the PostScript/PDF consumer, they may both refer to these slightly different colors with the same NAME. So when you view the file in Photoshop, you see color1... when you print it, your device says "I know color1" and yet it comes out slightly different.

That's why I'd sure like to examine your EPS.




Thomas D. Greer

http://www.tgreer.com

Providing PostScript & PDF
Training, Development & Consulting