Evening all -
this is by no means the first time I'm banged my head against the colour management brick wall, and I doubt it will be the last 
I am running Frame 12 and Illustrator CC on Windows 7.
Both these Adobe products ship bundled with Pantone libraries. (Correct me with I'm wrong, but these "libraries" are basically a bunch of look-up tables, mapping named Pantone colours to various CYMK, RGB, etc equivalents?)
Can anyone tell me why these two different Adobe applications appear to give different definitions for the same Pantone colours?
For example, if I choose the Pantone Coated library > 'Warm Red'
In Illustrator CC (left screenshot), it offers this as C0% M87.4% Y79.9% K0%
in FrameMaker 12 (right screenshot), I get C0% M79% Y91% K0%
Why are they different?
Are the library definitions shipped with the products simply different?
Frame's dialog ominously mentions "© Pantone, Inc., 1986, 1988"
And if I go to C:\Program Files (x86)\Adobe\AdobeFrameMaker12\fminit\color and open the corresponding .bcf file in an editor, the header says:
"BCF 2.0PANTONE¨ Coated 1.1 ©Pantone, Inc., 1986, 1988.PANTONE¨* Computer Video simulations displayed may not match PANTONE-identified solid color standards. Use current PANTONE Color Reference Manuals for accurate color. To order publications from Pantone, Inc., in the U.S. please call the toll-free number (800) 222-1149 [within NJ, call (201) 935-5500]. In other countries contact your local Pantone representative. *Pantone, Inc.'s check-standard trademark for color.[1]"
whereas if I go to C:\Program Files (x86)\Adobe\Adobe Illustrator CC\Presets\en_US\Swatches\Color Books and open the corresponding .acb swatch file in there, in amoungst the hex gibberish I note it says:
"�=$$$/colorbook/PantonePlusCoated/title=PANTONE+^R Solid Coated/$$$/colorbook/PantonePlusCoated/prefix=PANTONE *$$$/colorbook/PantonePlusCoated/postfix= CK"$$$/colorbook/PantonePlusCoated/description=Copyright^C Pantone LLC, 2010"
Have Pantone perhaps changed their colour definitions between 1988 and 2010?
(I'm thinking it would be sensible to establish this one way or the other before we get into any convoluted discussions about the Windows GDI etc etc)
24
Replies
Hmpf, well, doing some googling comes up with a bazillion "Why do Pantone colours not match between Illustrator and Photoshop" and "Why do Pantone colours not match between different versions of Illustrator" type threads...
(e.g. Pantone Color Changes )
So I think we can "safely" say that if the Pantone libraries shipping with Frame 12 and TCS5 really are from 1988 (!!?) then they almost certainly do contain different CYMK equivalents compared to the ones in Illustrator CS.
So, basically, we are all banjaxed before we even start, and there's no way whatsoever that any graphic I make in Illustrator and import into Frame is ever going to look the same colour as text or table rulings in Frame, unless I specify both manually as explicity CYMK (or RGB) values, and don't label them as Pantones at all..... which, of course, that no subsequent output device can ever say "aha! a Pantone!" and use colour management to get the best RGB or CYMK conversion for it.
Ho hum.
Sometimes I really think I'd be best just having a big box of crayons on my desk...... 
Thanks Arnis!
Arnis Gubins wrote
You can get new(er) Pantone colours into FM ... by creating a swatch set of the project colours in AI or Photoshop and saving those to an EPS or PDF file. Then import that file onto a Reference page and FM will automatically add those swatch colours to the document color catalog.
Aha! Brilliant. I should've thought of that! (As I'm was well aware that my Colour Definitions list often gets full of spurious entries from imported graphics... some of us may well remember all those RGB colours from paletted PNGs, for instance).
I wonder what Frame will do with these definitions if I defined the Pantone in Illustrator as an L*A*B spot colour though? (I will experiment!)
Understanding colour management is sort of like trying to always get the same colour of light-brown on a piece of toast everywhere. ...
Now come on, don't start going into the "Wooo, colour management! It's complicated!" thing. It's not exactly quantum mechanics. There is nothing "complicated" about how Adobe hasn't deigned to update FrameMaker's Pantone libraries since 1988. (Indeed, since Adobe only acquired Frame in 1995 (?), Adobe have NEVER updated its Pantone libraries!!) I'm struggling to see how this is acceptable, really? What are they playing at?!
To me, this is just yet another example of where colour "management" doesn't work properly because of poor quality software, but the vendor hopes to get around it by going "wooo! colour management! complicated!!" and obfuscating the actual problem.
But anyways, enough moaning - I shall fill in that bug report/feature request form suggesting that they might like to update those libraries. But Arnis' workaround will do very nicely in the meantime
OK, I did some experimenting.
I made a swatch in Illustrator CC, with Pantones define as spot LAB values.
I saved this as an EPS and imported it onto a new Reference Page in Frame 12.
Arnis, it works PERFECTLY in CYMK mode! The colours in the swatch all appear in the my Frame colour definitions, and if I examine their CYMK values, they match exactly the CYMK equivalents Illustrator uses if I take the Pantone out of spot LAB mode and view the CYMK equivalent. Given that Frame doesn't support LAB, we can't really ask for better.
However, it all goes belly up in RGB mode
As per the definitions in the Frame11 Maker.ini reference, in my maker.ini [Preferences] section, I have GetLibraryColorRGBFromCMYK=None, which is meant to "Display and print the RGB values from the library color description."
However Frame12 isn't doing this - it is using its own bogus algorithm to convert the CYMK value to RGB, the same broken behaviour we saw in this thread here https://forums.adobe.com/thread/1485020
when text colour in CHM output was all wrong.
For instance, compare this "Spruce Green" (Pantone 355C) - in Illustrator (right hand side), if I take it out of Spot LAB mode and ask for an RGB equivalent, I get 0-148-77.
Frame obfuscates this a little by expressing the values as percentages, but converting it to values out of 255, we get 0-220-0 (that lurid lime green colour)
Now I don't think I'm being unreasonable in asking how on EARTH is that supposed to be a good equivalent to the original Pantone?!? It's zeroed the blue channel!!! Truely bizarre.
David,
What options did you use for defining the colours internally and what options for output in AI? I suspect that if you examine the EPS file you created from AI for the Panotne 355C, the only values in there are the CMYK ones - no LAB nor RGB. So FM has to use the algorithm mentioned in the other thread and creates the lime green based upon the algorithm. If you have set the AI drawing to an RGB color mode and set the EPS export option for "Include CMYK Postscript in RGB files" to OFF, then you would get the desired RGB values in the EPS file.
If you're so intent on using Pantone Spot colours to get RGB, then you can create your own ACF library file with the exact RGB definitions that you want (you can then use whatever values you get in Photoshop, Illustrator or whatever source that you're trying to match). The structure of the file is as follows:
ACF 1.0
My Color Library <---- name of library
LibraryVersion: 1.0
Copyright: © 2014 <your name here>. All rights reserved.
AboutMessage: User defined Spot colours for RGB
Names: Partial
Rows: 4
Columns: 4
Entries: 1 <---- number of colours in file
Prefix: <---- used for display in FM, e.g. "Panotne"
Suffix: <----- used together with prefix, e.g. "CVC, CVU, etc."
Type: Process <---- Spot, Process, Tint, Mixed (need to add specifier to Data lines)
Models: CMYK RGB
PreferredModel: RGB <---- whatever you want
Data:
0.98 0.11 1.00 0.02 <---- CMYK values (0-1)
0 38143 19967 <---- RGB values as 16-bit, i.e. 16-bit = ( [RGB value (1-255)]*256) + 255; zero is still zero
Spruce Green <---- name of colour
The acf file needs to be installed in the fminit\color folder.
Also, a note of caution, If you screw something up in the acf file (like the incorrect number of entries in the file or you're missing a component value, this will hang FM if you use the View > Color > Definitions... option. You'll need to kill FM, fix or remove the ACF file and re-start. Also, FM only reads the libraries at the start, so if you need to make any changes to the ACF definitions, you have to r-start FM for those changes to take place.
[I know, FM doesn't make it easy, but it can be done...]
Thanks Arnis, there's some excellent info in there!
If you're so intent on using Pantone Spot colours to get RGB...
Well it's not so much that, as that I'm intent on using FrameMaker 12 for advertised purpose: single sourcing and outputting in multiple formats. And some of those formats (HTML Help) are in the RGB domain.
Frame ought to be able to capable of this (by referring to it's bundled Pantone libraries, which should contain good CYMK and RGB equivalents for the Pantone colours), but it's ending up pretty woeful in practice, because not only are Frame's Pantone definitions a quarter of a century out of date, but it keeps doggedly refusing to use the RGB definitions! So at present I'm pretty much having to have a seperate Frame source files and template for my HTML Help work. Obviously this is a PITA.
You've given me a lot to work with there, though, thanks! I need to open up my EPS swatches in a text editor and see what Illustrator CS has actually been putting in there. If it's not adding the RGB data, I need to amend my Illustrator settings so that it does! or if that can't be done, I shall have to hack together one of those ASCII Colour File .acf definitions.
(It's difficult to justify doing this kind of laborious research work on the client's time though - particularly when my excuse is "well, you know what TCS5 suite I persauded you to fork out all that money for, because it was the best tool for the job? Well, it doesn't actually work properly, so I'm having to play with technology for the 1980s in a text editor to try and fix it 
)
I will post back here with my findings!
A question:
was there ever a GUI front end utility from Adobe or a 3rd party, for editing ACF files? Maybe even something on DOS or 16 bit windows? I still have a Windows XP installation on a Oracle Virtual Box VM which could probably run something like that....
>> So at present I'm pretty much having to have a seperate Frame source files and template for my HTML Help work. Obviously this is a PITA.
That's pretty much the whole idea. You should have separate templates for the various Publish outputs. Each of the formats (HTML5/ePub/Mobi/CHM) have various differences & limitations compared to a print/PDF workflows. That's why the Set Template button is right there on the Style Mapping page. Use semantic naming for your tags and colours (e.g. instead of Spruce Green, why not CorporateColour or something like that?) and then use an optimized template for each of the required Publish outputs. This is especially true for using the colours. If you know what RGB you need, then in another template, just create that colour as an RGB one in FM in the first place. FM templates coupled with the Style mappings in the STS file settings, give you a lot more control then either route by itself. [other issues notwithstanding... 
]
The EPS swatch that I showed in the previous message was imported from AI, where the options were set to RGB and to not include CMYK in the RGB postscript, to show you that it can be done.
Regarding a GUI for editing the ACF, there never was one, AFAIK. There is a detailed description in the PageMaker 6 User Guide (if you can find one) and a bit of detail in this publication: ftp://ftp.adobe.com/pub/adobe/pagemaker/all/6.x/hificolr.pdf
All I'm saying is that there are (as typical in FM) any number of ways to accomplish what you want (a specific RGB colour), but due to the decrepit state of FM's documentation, this topic is not covered in any useful manner. Hopefully, the information provided here will help you find the path of least resistance to getting your work done.
Arnis Gubins wrote:
What options did you use for defining the colours internally and what options for output in AI? I suspect that if you examine the EPS file you created from AI for the Panotne 355C, the only values in there are the CMYK ones - no LAB nor RGB. So FM has to use the algorithm mentioned in the other thread and creates the lime green based upon the algorithm. If you have set the AI drawing to an RGB color mode and set the EPS export option for "Include CMYK Postscript in RGB files" to OFF, then you would get the desired RGB values in the EPS file.
Hi Arnis, I really appreciate your answers on this.
I haven't been able to work on it for a few weeks (licensing woes: the client I'm working for upgraded me to TCS5, so I lost Illustrator..... it's taken until now to get it back....... the main reason being that this client's purchasing team had no process for dealing with subscription-only-software billed monthly on a credit card.... ARRRGH ... but let's not get sidetracked! )
In this shiney new Illustrator CC 2014 (version 18), if I draw a rectangle, fill it with Pantone 355 C from the built in swatch library, and save to EPS.....
If my Doc Colour Mode is RGB and I leave "Include CMYK Postscript in RGB files" OFF, and open the resulting EPS in WordPad, I see:
%%DocumentCustomColors: (PANTONE 355 C)
%%CMYKCustomColor:
%%RGBCustomColor: 0 0.5859 0.2282 (PANTONE 355 C)
If however, my Doc Colour Mode is RGB and I set "Include CMYK Postscript in RGB files" ON, and open the resulting EPS in WordPad, I see:
%%DocumentCustomColors: (PANTONE 355 C)
%%CMYKCustomColor: 0.9578 0.0376 1 0.0040 (PANTONE 355 C)
%%RGBCustomColor:
I'm guessing that if the EPS were to contain both CYMK and RGB definitions for the Pantone 355, then FrameMaker could use them both in the appropriate contexts?
Is there an easy way to make Illustrator write them both? Can I add them both manually to the EPS file in WordPad?
Or can I make two EPS swatches, containing the same named Pantone colours but on has RGB definitions and the other has CYMK definitions? And then I place *both* swatches on a Reference page and would Frame be smart enough to merge them?
Or is the only sure way going to be to hand-craft a .acf file?
(I guess I can experiment with all this myself, but someone may already know the answer... )
ARRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRRGH 
OK, here's what I tried this time:
I used Arnis' template to create a .acf file in Notepad, containing CMYK and RGB definitions of my Spruce Green colour. (FWIW I set "preferred" mode to RGB, not sure what practical difference it makes).
I restarted Frame and lo and behold, there in the Color Definitions dialog was now my very own library with my one colour in it. I could add it to the document's color definitions, and switch between CMYK and RGB mode, and it had the right definitions and colour for both of them. I could color some text with my new colour, and make PDFs in both CMYK and RGB output modes, and they both looked very similar in Acrobat (the CMYK one was, unsurprisingly, sliiiiiightly different to what I saw on screen in Frame, as Frame had been displaying me the RGB definitions, but Acrobat was converting the CMYK to RGB on the fly to display on screen.)
But then the facepalming moment:
I tried CHM output via the Publish command.
ARRRRRRRRRRRRRRRGH.
I get the horrible lime green again.
Frame totally disregards the RGB definition in has for the colour, and does bogus maths on the CYMK version, resulting in a totally wrong hue.
It's beginning to look like this is a bit of a no win situation.
There's no way I can have a single set of colour definitions across Illustrator and Frame that will output reliably to both CYMK print and on-screen HTML Help.
Any time a colour definition possesses numbers for CYMK, it ruins the CHM output's colour, even if a valid RGB definition is also present.
The only way to get reliable colour in CHM output is to have a purely RGB definition of a library.
This is, of course, conversely, no use for creating PDFs for print.
I basically just need two different swatches: one RGB one and one Pantone one,
Frame documents that will end up as CHMs have the RGB EPS added to a reference page;
Frame documents that will end up as PDFs and might get printed have the CMYK EPS added to their reference page.
Or, another solution is just to keep everything in RGB mode. That way, PDFs look OK on screen and match my CHMs exactly. When printed on paper, the printer will just have to convert RGB > CMYK themselves, which will probably give perfectly acceptable results anyways.
Error7103 wrote:
. Older thread on this mess: ANSI Z535.1 Safety Colors in Frame
I was on that thread! (Guess what my login was back then.....:) )
Since then, the only progress seems to have been the introduction of the new Publish feature, which breaks things in a whole new way )
Thanks Arnis, this is more or less what I've ended up doing.
I'm confused at to why the MIF files can't store the RGB definitions, though. It's just a text file, same as a .acf file is just text - I can't see any technical reason why the MIF shouldn't store those numbers? As CHM is an RGB-only medium, it seems totally daft from Frame to generate it from a CMYK definition.
(I appreciate the answer is "because they kludged the functionality in from RoboHelp, which they bought with Macromedia 9 years ago"... but since then they've sold/rented us TCS1, TCS2., TCS3, TCS4 and now Frame12, all supposed to have ever-greater levels of integration between Frame and CHMs.... how many years is it going to take them to get this workflow right? 19? 29? .... Ah well......)
At least thanks to Arnis, I now know more about this subject than I ever dared dream You haven't quite lived until you've hand-crafted an .acf in Notepad
I'm confused at to why the MIF files can't store the RGB definitions, though. It's just a text file, same as a .acf file is just text - I can't see any technical reason why the MIF shouldn't store those numbers? As CHM is an RGB-only medium, it seems totally daft from Frame to generate it from a CMYK definition.
That's just the way MIF was implemented in the last century, when it was originally designed for print production (and even then it was a kludge to get it to work in the RGB-centric Windows world). FM is not doing the generating of CHM - the RoboHelp routines are and they only read MIF, not the internal FM binaries.
Sometimes the sum is the lesser of the parts when developers use a Frankenstein approach to cobble together features from disparate programs, rather than doing it right the first time. 
(At least it's not a total Chindōgu implementation, as some other marketing-driven features have been.)
Well they change the MIF format for every release, so it's not backwards compatible ... I hope that's not the only reason they change it!! 
they could also make it support some new features.
At least I have enjoyed reading Chindōgu - Wikipedia, the free encyclopedia this morning
AdChoices