Sie sind auf Seite 1von 35

The G7 Specification 2008

Final Working Draft

2008/09/12

G7 is a Registered Trademark of IDEAlliance

The G7 Specification 2008, Working Draft_B

Copyright and Legal Notices Copyright (c) International Digital Enterprise Alliance, Inc. [IDEAlliance] (2001 2008). All Rights Reserved. This document and translations of it may be copied and furnished to others, and derivative works that comment on or otherwise explain it or assist in its implementation may be prepared, copied, published and distributed, in whole or in part, without restriction of any kind, provided that the above copyright notice and this paragraph are included on all such copies and derivative works. However, this document itself may not be modified in any way, such as by removing the copyright notice or references to IDEAlliance, except as needed for the purpose of developing IDEAlliance specifications, in which case the procedures for copyrights defined in the IDEAlliance Intellectual Property Policy document must be followed, or as required to translate it into languages other than English. The limited permissions granted above are perpetual and will not be revoked by IDEAlliance or its successors or assigns. NO WARRANTY, EXPRESSED OR IMPLIED, IS MADE REGARDING THE ACCURACY, ADEQUACY, COMPLETENESS, LEGALITY, RELIABILITY OR USEFULNESS OF ANY INFORMATION CONTAINED IN THIS DOCUMENT OR IN ANY SPECIFICATION OR OTHER PRODUCT OR SERVICE PRODUCED OR SPONSORED BY IDEALLIANCE. THIS DOCUMENT AND THE INFORMATION CONTAINED HEREIN AND INCLUDED IN ANY SPECIFICATION OR OTHER PRODUCT OR SERVICE OF IDEALLIANCE IS PROVIDED ON AN " AS IS" BASIS. IDEALLIANCE DISCLAIMS ALL WARRANTIES OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO, ANY ACTUAL OR ASSERTED WARRANTY OF NONINFRINGEMENT OF PROPRIETARY RIGHTS, MERCHANTABILITY, OR FITNESS FOR A PARTICULAR PURPOSE. NEITHER IDEALLIANCE NOR ITS CONTRIBUTORS SHALL BE HELD LIABLE FOR ANY IMPROPER OR INCORRECT USE OF INFORMATION. NEITHER IDEALLIANCE NOR ITS CONTRIBUTORS ASSUME ANY RESPONSIBILITY FOR ANYONE'S USE OF INFORMATION PROVIDED BY IDEALLIANCE. IN NO EVENT SHALL IDEALLIANCE OR ITS CONTRIBUTORS BE LIABLE TO ANYONE FOR DAMAGES OF ANY KIND, INCLUDING BUT NOT LIMITED TO, COMPENSATORY DAMAGES, LOST PROFITS, LOST DATA OR ANY FORM OF SPECIAL, INCIDENTAL, INDIRECT, CONSEQUENTIAL OR PUNITIVE DAMAGES OF ANY KIND WHETHER BASED ON BREACH OF CONTRACT OR WARRANTY, TORT, PRODUCT LIABILITY OR OTHERWISE. IDEAlliance takes no position regarding the validity or scope of any intellectual property or other rights that might be claimed to pertain to the implementation or use of the technology described in this document or the extent to which any license under such rights might or might not be available. IDEAlliance does not represent that it has made any effort to identify any such rights. Information on IDEAlliance's procedures with respect to rights in IDEAlliance specifications can be found at the IDEAlliance website. Copies of claims of rights made available for publication, assurances of licenses to be made available, or the result of an attempt made to obtain a general license or permission for the use of such proprietary rights by implementers or users of this specification, can be obtained from the President of IDEAlliance. IDEAlliance requests interested parties to disclose any copyrights, trademarks, service marks, patents, patent applications, or other proprietary or intellectual property rights which may cover technology that may be required to implement this specification. Please address the information to the President of IDEAlliance.
G7 , GRACoL , and SWOP are registered trademarks of IDEAlliance

The G7 Specification 2008, Final Working Draft for Public Comment

Table of Contents
1 STATUS ..................................................................................................................................................... 1 1.1 DOCUMENT STATUS .............................................................................................................................. 1 1.2 DOCUMENT LOCATION .......................................................................................................................... 1 1.3 VERSION HISTORY ................................................................................................................................. 1 2 G7 SPECIFICATION DOCUMENTATION STRUCTURE .............................................................. 2 2.2 REQUIREMENT WORDING NOTE ............................................................................................................ 2 2.3 ADDITIONAL G7 DOCUMENTATION ....................................................................................................... 2 2.4 ORIGIN .................................................................................................................................................. 2 2.5 GOVERNANCE........................................................................................................................................ 2 2.6 ACCESS TO G7 DOCUMENTATION.......................................................................................................... 2 2.7 SPELLING ............................................................................................................................................... 2 3 INTRODUCTION TO THE G7 SPECIFICATION ............................................................................ 3 3.1 3.2 3.3 3.4 4 PURPOSE.......................................................................................................................................... 3 EXCLUSIONS .................................................................................................................................... 3 SCOPE.............................................................................................................................................. 3 ABOUT THE G7 SPECIFICATION ....................................................................................................... 3

WHAT IS G7?....................................................................................................................................... 4 4.2 RELATIONSHIP TO ISO STANDARDS AND PRINTING SPECIFICATIONS .............................................. 4 4.2.1 ISO 12647-2 Printing Standard.............................................................................................. 4 4.2.2 Traditional ISO-Based Print Specifications ........................................................................... 5 4.2.3 G7 and ISO TS 10128............................................................................................................. 5 4.3 G7 VS. OTHER CALIBRATION METHODS ........................................................................................... 5 4.3.1 G7 vs. TVI............................................................................................................................... 5 4.3.2 G7 vs. TVI-Based Print Specifications ................................................................................... 6 4.3.3 G7 vs. 4-D CMYK Transforms................................................................................................ 7

5 THE G7 DEFINITION FOR GRAY BALANCE (NEUTRALITY) ................................................... 8 5.1 TRADITIONAL GRAY BALANCE ....................................................................................................... 8 5.2 G7 GRAY BALANCE ........................................................................................................................ 8 5.2.1 Example .................................................................................................................................. 9 5.3 G7 CMY TRIPLET VALUES ............................................................................................................. 9 5.3.1 G7 Gray Balance at 300% CMY............................................................................................. 9 5.4 BENEFITS AND LIMITATIONS OF G7 GRAY BALANCE ...................................................................... 9 5.4.1 Example of G7 Gray Balance on Blue-Biased Paper ........................................................... 10 5.4.2 Example of G7 Gray Balance on Yellow-Biased Paper ....................................................... 10 5.4.3 Limitations of G7 Gray Balance........................................................................................... 11 6 THE G7 DEFINITION FOR TONALITY ..................................................................................... 12 6.1 NPDC (NEUTRAL PRINT DENSITY CURVES).................................................................................. 12 6.1.2 Dynamic Range Adaptation.................................................................................................. 13 6.2 ORIGIN OF NPDC CURVES ............................................................................................................. 14 6.2.1 SiCoTVI Formula ................................................................................................................. 14 6.3 NPDC RELATIONSHIP TO TRADITIONAL DENSITY-BASED TVI..................................................... 14 6.4 MAXIMIZING ISO COMPATIBILITY ................................................................................................. 15 7 THE G7 DEFINITIONS FOR HR, SC AND HC .......................................................................... 16 7.1 NOMINAL HR, SC AND HC ND AIM VALUES ............................................................................... 16 7.2 G7 HIGHLIGHT RANGE (HR) ......................................................................................................... 16 7.2.1 Measuring HR_cmy .............................................................................................................. 16 7.2.2 Measuring HR_k................................................................................................................... 16

The G7 Specification 2008, Final Working Draft for Public Comment 7.3 G7 SHADOW CONTRAST (SC)........................................................................................................ 17 7.3.1 Measuring SC_cmy............................................................................................................... 17 7.3.2 Measuring SC_k ................................................................................................................... 17 7.4 G7 HIGHLIGHT CONTRAST (HC) ................................................................................................... 17 7.4.1 Computing HC_cmy ............................................................................................................. 17 7.4.2 Computing HC_k .................................................................................................................. 17 7.5 HR, SC AND HC AIM VALUES VS. DYNAMIC RANGE .................................................................... 17 7.6 EXPRESSING HR, SC AND HC IN L* .............................................................................................. 18 8 G7 COMPLIANCE .......................................................................................................................... 19 8.1 8.2 8.3 8.4 G7 GRAYSCALE ............................................................................................................................. 19 G7 TARGETED ............................................................................................................................... 19 G7 CHARACTERIZED ..................................................................................................................... 19 G7 EXTREME ................................................................................................................................. 19 CHARACTERIZATION DATA SETS BASED ON G7 ....................................... 21

APPENDIX A: A.1 A.2 A.3 A.4 A.5 A.6 A.7 A.8

ABOUT THE METHODOLOGY ......................................................................................................... 21 A NEW APPROACH FOR DEVELOPING REFERENCE CHARACTERIZATIONS..................................... 21 DEVELOPMENT OF THE GRACOL2006_COATED1 DATA SET ........................................................ 22 DEVELOPMENT OF THE SWOP2006 DATA SETS - OVERVIEW ...................................................... 23 DETAILS FOR GENERATING SWOP2006_COATED5...................................................................... 23 DETAILS FOR GENERATING SWOP2006_COATED3 ...................................................................... 23 EXAMPLE VERIFICATION OF NEW SWOP CHARACTERIZATIONS .................................................. 23 CONCLUSION ................................................................................................................................. 24 FRACTIONAL PERCENTAGE CALCULATIONS.............................................. 25

APPENDIX B: B.1 B.2 B.3 B.4 B.5

THE PROBLEM WITH INTEGER PERCENTAGES ............................................................................... 25 CONVERTING INTEGER PERCENTAGE TO INTEGER 8-BIT (EXCEL)................................................. 25 CONVERTING INTEGER 8-BIT TO FRACTIONAL PERCENTAGE (EXCEL) .......................................... 25 CONVERTING INTEGER PERCENTAGE TO FRACTIONAL PERCENTAGE ............................................ 25 G7 FRACTIONAL PERCENTAGE CONVERSION CHART .................................................................... 26 TOLERANCE NOTES .............................................................................................. 27

APPENDIX C: C.1 C.2 C.3 C.4 C.5 C.6 G.7

INTRODUCTION .............................................................................................................................. 27 WHEN TIGHT PRESSROOM TOLERANCES ARE NOT APPROPRIATE ................................................ 27 WHEN TIGHT PRESSROOM TOLERANCES ARE APPROPRIATE ....................................................... 27 G7 TOLERANCES VS. CUSTOM USER TOLERANCES ....................................................................... 27 CHARACTERIZATION VS. PRODUCTION TOLERANCES .................................................................... 28 PRINTING VS. PROOFING TOLERANCES .......................................................................................... 28 WET-TRAPPING - THE BANE OF OFFSET LITHOGRAPHY ................................................................ 28 G7 GLOSSARY .......................................................................................................... 29

APPENDIX D:

ii

The G7 Specification 2008, Final Working Draft for Public Comment

1 Status
1.1 Document Status
The status of this document is: Draft Released for Public Comment Released

1.2 Document Location


The location of this document is: http://www.g7global.org/specification/

1.3 Version History


Version Number Working Draft A Working Draft B Final Working Draft Release Date 071508 091108 09/12/08 Editor Kennedy Kennedy Kennedy Description Developed Draft for G7 as a standalone specification 2008 Additions, alterations and styling by DBH Final additions and edits following review

The G7 Specification 2008, Final Working Draft for Public Comment

2 G7 Specification Documentation Structure


G7 is described in a formal document that represents the G7 Specification.

2.2 Requirement Wording Note


The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT", "SHOULD", "SHOULD NOT", "RECOMMENDED", "MAY", and "OPTIONAL" in this document are to be interpreted as described in [RFC-2119]. The specification also uses the normative term, STRONGLY ENCOURAGES, which should be understood as a requirement equivalent to MUST in all but the most extraordinary circumstances. Capitalization is significant; lower-case uses of the key words are intended to be interpreted in their normal, informal, English language way.

2.3 Additional G7 Documentation


Auxiliary documents support the specification, primarily Calibrating, Printing and Proofing to the G7 Method. [G7 How-To].

2.4 Origin
The G7 Specification was initially developed for commercial printing by the GRACoL Committee and first published within GRACoL 2007. Since that time, both the IDEAlliance SWOP and GRACoL Committees have adopted G7 as the basis their specifications and characterization data sets, as have groups outside of IDEAlliance.

2.5 Governance
As of 2008, ongoing development of the G7 Specification and related documentation is governed by the IDEAlliance Print Properties and Colorimetrics Working Group.

2.6 Access to G7 Documentation


The G7 Specification and additional documentation are all publicly and freely available on the G7 website, www.g7global.org.

2.7 Spelling
Except in the case of proper names, US spelling conventions are used. Note that gray and grey are both legal spellings of the same word. IDEAlliance uses the phonetically simpler version.

The G7 Specification 2008, Final Working Draft for Public Comment

3 Introduction to the G7 Specification


3.1 Purpose
The purpose of this document is to define the colorimetric principles and target values of the G7 Specification. It also defines a multi-tier approach to G7 compliance, including G7 Grayscale, G7 Targeted and G7 Characterized. Because the basic G7 Grayscale definition is substrate- and colorant-independent, this document applies equally to all G7-based imaging processes and specifications, regardless of substrate or colorants.

3.2

Exclusions
How to perform G7 calibration. How to use ICC color management or other tools to match a G7 Characterized Print Specification such as GRACoL or SWOP.

This document does not teach;

Calibrating a device to the values and principles presented in this specification can be found in the companion document Calibrating, Printing and Proofing to the G7 Method [G7How-To].

3.3

Scope
An overview of what G7 is A formal colorimetric definition of G7 gray balance or neutrality A formal colorimetric definition of G7 grayscale tonality expressed as Neutral Print Density Curves (NPDC) Formal definitions of the G7 process control variables, HR (Highlight Range), HC (Highlight Contrast) and SC (Shadow Contrast). Formal definitions of the three recognized levels of G7 Compliance - G7 Grayscale, G7 Targeted and G7 Characterized. Various supporting definitions and appendixes

The G7 2008 Specification includes:

3.4

About the G7 Specification

The G7 Specification is made up of formal colorimetric definitions for neutrality and tonality that together control the visual appearance of the grayscale component of an image. The GRACoL and SWOP specifications are both based on G7 gray balance and tonality. The G7 Specification defines multiple possible levels of G7 compliance G7 Grayscale, G7 Targeted, G7 Characterized and G7 Extreme. G7 is a registered trademark of IDEAlliance. The intellectual property of the G7 calibration process belongs to its inventor, Hutchcolor LLC. HutchColor granted to IDEAlliance usage rights, free-ofcharge, including the right to publish IDEAlliance specifications based on this intellectual property, the right to develop and deliver training materials based on the IP, and the right to use the intellectual property as the basis for IDEAlliance programs. The G7 Specification 2008 was authored by Don Hutcheson with editorial review by Dianne Kennedy the IDEAlliance Colorimetrics and Print Properties Working Group.

The G7 Specification 2008, Final Working Draft for Public Comment

What is G7?

G7 is a specification that defines the factors that can be used to provide consistent gray scale reproduction across multiple devices, processes and media. The main innovation of G7 is the direct measurement of a combined CMY neutral gray scale, instead of separate CMY dot gain (TVI) scales. This allows the well-proven photographic principles of gray balance and exposure control to be applied to CMYK printing, ensuring a close visual match between printed images no matter where or how they are produced. Although developed initially for commercial offset proofing, G7 has been applied successfully to a wide range of processes including coated and uncoated offset, newsprint, gravure, flexography, dyesublimation, ink-jet, electro-photography, screen printing and others. By directly controlling the density, contrast and balance of a neutral gray scale, G7 brings the goal of printing to the numbers one step closer to reality. G7 has enjoyed major adoption in two distinct areas. Firstly, the gray balance and neutral tone scales defined within the G7 Specification have been used as the basis for developing standardized characterization datasets including TR003 and TR005 (SWOP), TR006 (GRACoL) and TR007 (FIRST/Flexo). These and any other printing specifications that are based on G7 exhibit a shared appearance that makes the exchange of image files and proofs between different imaging processes easier and more predictable. See Appendix A for information about developing characterization data sets based on G7. Secondly, the principles of G7 are also being used to calibrate presses and proofing systems worldwide. The techniques for calibrating a device to the G7 values and principles presented in this specification can be found in the companion document Calibrating, Printing and Proofing to the G7 Method [G7How-To].

4.2

Relationship to ISO Standards and Printing Specifications

G7 is a specification, not a standard. IDEAlliance does not create standards but works with other organizations via the TC130 ISO subcommittee to help develop standards. Whenever possible, IDEAlliance specifications such as G7, GRACoL and SWOP are based on the appropriate ISO standards. When a standard does not exist for a new process, or when the standard does not support modern best practices, IDEAlliance, like similar organizations in other countries, may invent appropriate specifications, such as G7, to fill the gap, and then submit them to the standards process for possible future adoption. 4.2.1 ISO 12647-2 Printing Standard ISO is the International Standards Organization. ISO 12647-2 is a standard for "Graphic technology Process control for the production of half-tone colour separations, proof and production prints - Part 2: Offset lithographic processes". Although ISO 12647-2 reflects the best practices of the 1990s, the current version (as of August 2008) is not optimized for new technologies such as ICC color management, CtP (Computer to Plate) and digital proofing, which came into use after ISO 12647-2 was written. For example, ISO 12647 defines multiple TVI (Tone Value Increase) curves based on legacy positive and negative film-based plate making, but contains (at the time of writing) no curve for todays CtP plates. Also, ISO 12647 defines the CIELab values for only a handful of solid ink colors, while todays print specifications like Fogra in Germany and GRACL in the USA are based on Characterization Data Sets (hundreds of CMYK combinations and their CIELab measured data). Finally, ISO 12647 neglects to define one of the most critical metrics in any color process gray balance. Put another way, the existing ISO 12647 standard defines the ingredients, but not the product, of good printing. Todays ICC-savvy users want a standard that defines precisely how the final printed piece should appear. One way to achieve this is to create standards based on Characterization Data Sets similar to those used in ICC color management. 4

The G7 Specification 2008, Final Working Draft for Public Comment At the time of writing, the ISO TC 130 subcommittee is working to define a set of internationallyunified characterization data sets for various offset printing types. Meanwhile, several independent organizations including FOGRA (Germany), IFRA (International), JIS (Japan), GRACoL (USA), SWOP (USA) and others have produced their own specifications based on their own interpretations of ISO 12647. 4.2.2 Traditional ISO-Based Print Specifications An ISO-based print specification is an interpretation of the ISO 12647 standard for a particular process or user group. Traditional Print Specifications (e.g. the original SWOP 1977-2003) defined paper type, solid ink densities (later CIELab values) and a single TVI (dot gain) number for each ink. These parameters helped multiple users approximate the appearance of each other, but did not guarantee a match between any two devices. 4.2.3 G7 and ISO TS 10128 ISO TS 10128 Methods for calibrating a printing system with digital data is presently an ISO work item. This new document includes three options for achieving the aim values in ISO 12647: the near-neutral method as exemplified by G7 and submitted by the US TAG, a density TVI method as exemplified by the FOGRA/ECI/bvdm process, and a device link method submitted by the International Color Consortium (ICC). All of these are viable, interchangeable and compatible with one another as long as good printing is practiced.

4.3

G7 vs. other calibration methods

Prior to G7 the two most common methods of calibrating a CMYK device to match another device or reference print condition were; The TVI method - adjusting the TVI values of standard CMYK colorants using four separate 1-D curves. The 4-D CMYK transform method - converting CMYK pixel values via 4-dimensional look-up tables based on ICC profiles or device-links. These two methods are fundamentally different from each other in that the TVI method (like G7) simply controls each colorant individually, while the device link method controls the whole color space. Each of these methods can be either replaced, or enhanced by the G7 method. 4.3.1 G7 vs. TVI Most traditional CMYK calibration methods measure the TVI of each colorant and create correction curves to produce TVI curves similar to some reference curve. Advantages of the TVI method; TVI is easy to measure - most hand-held densitometers already calculate TVI Variations in solid density or paper density create no calculation difficulties Only requires a densitometer Disadvantages of the TVI method; TVI is density based, with no direct relationship to colorimetric data TVI does not control gray balance errors caused by ink hue variations TVI does not control gray balance errors two- and three-color ink trapping TVI does not control neutral density variations caused by 3-color trapping variations TVI produces different mid-tone darkness depending on density range Different TVI curves are needed for consistent appearance on multiple devices Advantages of G7 over TVI; G7 replaces multiple TVI curves with a single 'NPDC' (Neutral print Density Curve) G7 results in predictable gray balance

The G7 Specification 2008, Final Working Draft for Public Comment G7 results in predictable tonality and contrast G7 controls mid-tone darkness independent of density range G7 eliminates much of the work normally done by ICC color management or device links Disadvantages of G7 over TVI; NPDC curves are harder to calculate when solid density varies Requires use of a spectrophotometer 4.3.2 G7 vs. TVI-Based Print Specifications To put the difference between G7 and TVI calibration into perspective, the following illustration shows how closely two different specifications can look when the only difference is G7 or TVI calibration.

Interpretations of ISO 12647-2 by the TVI method (left) and the G7 method (right) and their relative CIEXYZ-based TVI curves (below). (Color and density may vary depending on viewing and reproduction conditions.) SCID images used by permission of CGATS. NOTE: The TVI curves shown above are calculated from CIEXYZ data - NOT density, and are for comparison only.

The G7 Specification 2008, Final Working Draft for Public Comment 4.3.3 G7 vs. 4-D CMYK Transforms Most proofing proofing systems today manage color accuracy with four-dimensional (4-D) CMYK transformations based on ICC profiles or custom device links. Advantages of the 4-D transform approach All image aspects are controlled together including tonality, gray balance, and colors Relatively simple to implement using ICC or proprietary solutions Allows relatively good matching using widely different media, colorants, etc. Disadvantages of the 4-D transform approach Requires excellent profiles of both the imitator and reference devices Good press profiles are hard to make and a bad profile causes more problems than it solves If the output device is naturally very dark or light or non-neutral, much of the profile will be wasted solving these issues, leaving less precision for subtle color adjustments Device link conversions may have less tonal precision than RIP calibration tables, especially in pastel highlight tones Involves changes to image % values that some users may not want to accept, and/or unwanted side-effects such as scum dots Not universally supported in all RIPs or workflow software Slower processing than simple 1-D calibration curves Depends on device calibration repeatability if recalibration results in different gray balance, a whole new profile or 4-D transform may be required Advantages of G7 compared to 4-D transforms Accomplishes gray balance and tonal calibration more efficiently than device-links A 4-D transform may be unnecessary after G7 calibration. Pre-calibrating with G7 improves the precision of subsequent ICC profiles. For example, in a profile of a device with a strong magenta color cast most of the magenta precision is used to cancel the green cast, leaving less precision in magenta-content colors, while a profile made after G7 pre-calibration maintains full precision in all colors Permits greater tonal accuracy, especially in pastel highlight tones Disadvantages of G7 compared to 4-D transforms Not possible on RIPs that lack CMYK calibration curves

The G7 Specification 2008, Final Working Draft for Public Comment

5 The G7 Definition for Gray Balance (Neutrality)


One of the basic requirements for constant visual appearance is gray balance, or neutrality. The color gray implies an absence of all color, but historically the printing industry has never agreed on an unambiguous definition of what exactly is gray.

5.1

Traditional Gray Balance

Traditionally gray balance was loosely defined as either; The CMY percentages needed to match the color of a black ink tint with the same density, The product of printing with equal CMY TVI values, Matching the color (a* and b*) of the substrate (paper). Unfortunately none of these concepts guarantees a consistent or pleasing neutrality because; The color of a black ink tint is often non-neutral, Equal CMY TVI values do not normaly produce a neutral gray, Matching all tones of a gray scale from white to black to the same a* and b* values as a nonneutral substrate often does not produce a gray scale that looks neutral to the eye.

5.2

G7 Gray Balance

The G7 Specification solves the above problems by defining gray balance as a function of paper color (expressed in CIE a* and b*) where the ideal a* and b* values for each gray scale step are reduced towards zero as the gray scale darkens. For convenience the wanted a* and b* values for any step in a CMY grayscale are calculated by multiplying the a* and b* values of the paper by the inverse ratio of the cyan percentage of that step, where cyan is an approximate indicator of grayscale darkness; Wanted a* Wanted b* = = a*_paper * (100 - C%) / 100; b*_paper * (100 - C%) / 100;

These formulae can be expressed graphically as two straight lines, one for a* (pink in the graph below) and one for b* (blue) vs. Cyan %. When C = 0%, both lines originate at the a* and b* values of paper. When C = 100% both lines end at 0a* and 0b*.

Graphs of wanted a* (pink) and b* (blue) for an ideal gray scale printed on paper with a nonstandard white point of a* = 2, b* = -5

The G7 Specification 2008, Final Working Draft for Public Comment 5.2.1 Example For a paper with a* = 2.0 and b* = -6.0, find the wanted a* and b* values for the 25%, 50% and 75% CMY grayscale patches.. Grayscale C % 25% 50% 75% (100 C%) / 100 0.75 0.5 0.25 Wanted a* +1.5 a* +1.0 a* +0.5 a* Wanted b* -4.5 b* -3.0 b* -1.5 b*

Calculating wanted a* and b* for gray patches on a known paper color

5.3

G7 CMY Triplet Values

To define gray balance for a CMY gray scale, it is also necessary to define the composition of the gray scale in terms of CMY Triplet percentages for each gray scale step. The G7 CMY Gray Balance Triplet Percentages are based on a traditional mid-point ratio of 50c, 40m, 40y, modified non-linearly towards each end of the tonal scale. The following chart shows CMY triplet values for each step of the G7 P2P target. C% 0 1.96 3.92 5.88 7.84 10.2 14.9 20 25.1 30.2 34.9 40 45.1 M% 0 1.38 2.77 4.15 5.61 7.41 11 14.9 18.8 22.91 26.78 30.98 35.48 Y% 0 1.38 2.77 4.15 5.61 7.41 11 14.9 18.8 22.91 26.78 30.98 35.48 C% 49.8 54.9 60 65.1 69.8 74.9 80 85.1 89.8 94.9 98.04 100 M% 39.82 44.71 49.8 54.9 60.16 66.07 71.77 78.06 84.61 92.2 96.86 100 Y% 39.89 44.71 49.8 54.9 60.16 66.07 71.77 78.06 84.61 92.2 96.86 100

Chart of G7 CMY Gray Balance triplets rounded to nearest 8-bit equivalent percentages 5.3.1 G7 Gray Balance at 300% CMY For convenience, gray balance triplets above 75% converge to C=M=Y=100% with a theoretical color of 0 a* and 0 b*. There are several logical reasons for this; In practice a 300% CMY patch (100C, 100M, 100Y) can be virtually any color, therefore 0 a*, 0 b* is probably a fair average color A terminus of equal CMY values totaling 300% is mathematically convenient CMY values over 75% in neutrals are uncommon The color of darker CMY triplets is hidden by black ink in most CMYK images.

5.4

Benefits and Limitations of G7 Gray Balance

The G7 Gray Balance definition has been carefully researched to ensure that on non-neutral papers, for example 'warm' publication stocks, or 'cold' white papers, a G7 grayscale appears neutral to the human eye when the stock is used as a visual reference. For example, on 'cold' (blue-white) paper G7 gray balance produces a slightly cold gray balance which appears neutral compared to paper, as shown in the illustration below.

The G7 Specification 2008, Final Working Draft for Public Comment 5.4.1 Example of G7 Gray Balance on Blue-Biased Paper

Three different Gray Balance formulae on blue-biased paper. In left example above the whole gray scale is reproduced at the same a* and b* values as paper, regardless of darkness, with the result that darker grays appear too blue compared to paper. In the right example grays have been forced to absolute neutral (0 a*, 0 b*), with the result that darker grays appear too yellow compared to paper. In the middle example the G7 formula produces the most neutral-looking gray balance compared to the surrounding paper color and to the black-only reproduction (center). 5.4.2 Example of G7 Gray Balance on Yellow-Biased Paper

Three different Gray Balance formulae on yellow-biased paper. In left example above the whole gray scale is reproduced at the same a* and b* values as paper, regardless of darkness, with the result that darker grays appear too yellow compared to paper. In the right example grays have been forced to absolute neutral (0 a*, 0 b*), with the result that darker grays appear too blue compared to paper. In the middle example the G7 formula produces the most neutral-looking gray balance compared to the surrounding paper color and to the black-only reproduction (center).

10

The G7 Specification 2008, Final Working Draft for Public Comment 5.4.3 Limitations of G7 Gray Balance The only known limitation of the G7 gray balance formula is that a grayscale image printed on two papers of noticeably different color will produce prints with noticeably different gray balances. This is illustrated by comparing the middle reproduction in the two examples above. This difference can be reduced in two ways; Altering the gray balance of one or both image files in an image editing software (e.g. Photoshop) to cancel the paper color, but then gray balance will vary according to tone level, resulting in a multi-colored gray scale. (The top and bottom versions of the right-hand image in the above two examples have been treated this way, and are both identical in middle-tones, but to most observers they do not match each other). Converting images via the absolute rendering intent to a custom ICC profile made with the nonstandard paper, but (a.) this may produce a residual dot in white areas that would normally carry no dots, and (b.) grays will no longer appear neutral compared to the surrounding paper. The only correct way to match gray images between different devices is to use ISO-standard paper of a similar color and perform a normal G7 calibration on both devices.

11

The G7 Specification 2008, Final Working Draft for Public Comment

The G7 Definition for Tonality

As important as gray balance is, an equally important requirement for a constant visual match between colored or monochrome images is Tonality. Tonality is a word used in photography to describe the darkness, lightness and contrast of an image.

6.1

NPDC (neutral Print Density Curves)

The G7 Specification defines tonality with a Neutral Print Density Curve, or NPDC. The G7 NPDC is the relationship between measured neutral density and original halftone percentages of a printed gray scale. The G7 Specification defines two standard Neutral Print Density Curves, one for a combined CMY gray scale and one for a black-ink gray scale, as shown below.

The G7 NPDC Graph for a CMY gray scale. The bold black curve shows the nominal NPDC of ISO-standard commercial printing with a neutral density range of 1.37. Other curves are for higher or lower neutral density ranges.

12

The G7 Specification 2008, Final Working Draft for Public Comment

The G7 NPDC Graph for a black-only gray scale The bold black curve shows the nominal NPDC of ISO-standard commercial printing with a neutral density range of 1.69. Colored curves are for higher or lower neutral density ranges 6.1.2 Dynamic Range Adaptation As the graphs above show, the G7 NPDC is not a single curve but an infinite number of possible curves defined by a non-linear function which adapts to fit the natural dynamic range (a.k.a. density range) of any process. The bold black curves show the nominal NPDC target curves for ISOstandard commercial printing. The colored curves show the NPDC curves for processes with higher or lower neutral density ranges. The two graphs above can be applied to any printing method or specification simply by choosing the colored curve (or average of a pair of curves) nearest to the density range of the device being calibrated. The G7 NPDC fitting function is designed to compress shadow detail before sacrificing highlight detail in a way that is most acceptable to most observers on most images. The compression effect moves progressively closer to the highlight as density range decreases, so that very dark shadow details are never completely lost. The adaptive nature of the G7 NPDC curves ensures that imaging processes with widely differing density ranges, for example newsprint and commercial printing, preserve the same amount of image detail in crucial highlight tones, and maintain the same approximate middle-tone weight (lightness / darkness) and contrast in most images. The NPDC principle ensures G7-based printing specifications share a similar visual appearance, even with significant differences between substrates, colorants or natural tonality characteristics.

13

The G7 Specification 2008, Final Working Draft for Public Comment

6.2

Origin of NPDC curves

To determine the 'natural' NPDC curves of commercial CtP-based printing, G7 research analyzed numerous press runs made with ISO-standard ink and paper, and a variety of plate types imaged on un-calibrated CtP systems (no RIP curves applied, not even to linearize the plate). To define the curves mathematically, the measured data were averaged and smoothed to remove measurement anomalies, then matched by trial and error to a variety of experimental formulae until one was found that fitted the data smoothly and accurately. The following equation shows the fitting formula finally adopted. It begins by calculating a pseudoTVI curve from the CMY gray scale using a combination of Sin and Cosine functions hence the name SiCoTVI; 6.2.1 SiCoTVI Formula For TV (dot%) = SiCoY (cmy) SiCoY (k) SiCoTV = = =

0.0 100.0; 100 - 95.7 * SiCoTV/100; 100 - 98 * SiCoTV/100; TV + SiCoTVI; (((1 - 1/10^ SIN(TV/100)) / 0.8559448 - TV/100) * 2.744 (((1 - 1/10^ SIN(TV/100)) / 0.8559448 - TV/100) * 2.2295

SiCoTVI (cmy) = SiCoTVI (k) =

((0.1 - 1/10^ COS(TV/100)) / - 0.1882025 - TV/100) * 0.594) * 25.2; - ((0.1 - 1/10^ COS(TV/100)) / - 0.1882025 - TV/100) * 1.0395) * 20.1;

30 25 20 15 10 5 0 0 10 20 30

Gray TVI from Sin/Cos

SiCoTVIc SiCoTVIk

40

50

60

70

80

90

100

The Gray TVI graphs produced by the Sin/Cos formula, from which the master NPDC curves were derived Having calculated the theoretical Gray TVI of the CMY gray scale, absolute neutral densities were regenerated by the inverse of the TVI formula, to produce the primary NPDC curve. NPDC ND = IF(SiCoY > 0, LOG10(100/SiCoY), 1000);

6.3

NPDC Relationship to Traditional Density-Based TVI

Extensive practical testing since the NPDC curves were first published confirms that the G7 NPDC formula represents what an average press running to ISO 12647-2 specifications would produce with typical un-calibrated CtP plates, on high quality commercial stock. Note that the SiCoTVIk curve shown above can be compared directly to traditional black-ink TVI curves, however attempting to correlate the SiCoTVIc curve to individual CMY TVI curves is an

14

The G7 Specification 2008, Final Working Draft for Public Comment inexact process, as there is no fixed relationship between grayscale density and individual-ink densities. Approximate CMY TVI curves can be extracted from any G7 Characterized data set such as TR006 using CIEXYZ in place of traditional CMY density-derived reflectance values. The CMY_TVI curves extracted this way typically plot between the old positive and negative curves in ISO 12647-2, but closer to positive than negative.

6.4

Maximizing ISO compatibility

To further align G7 with other interpretations of ISO 12647-2, such as FOGRA and ECI in Europe and SWOP in the USA, the G7 K NPDC curve is deliberately slightly lighter at 50% than the G7 CMY NPDC curve is at 50C, 40M, 40Y. IMPORTANT: G7 NPDC curves are NOT an official ISO standard. Other valid interpretations of ISO 12647-2 may produce slightly different NPDC curves.

15

The G7 Specification 2008, Final Working Draft for Public Comment

The G7 Definitions for HR, SC and HC

HR, SC and HC are three new variables which allow the NPDC of a G7-calibrated process to be monitored quickly without having to measure a complete grayscale. Briefly, HR (Highlight Range) monitors the neutral density of mid-tones, SC (Shadow Contrast) monitors the neutral density of shadow tones, and HC (Highlight Range) monitors the neutral density of highlight tones. For convenience, HR, SC and HC are expressed in neutral density (ND), which can be measured with a densitometers K or 'Visual density' channel. ND can also be computed from CIEXYZ_Y using the following official G7 formula; neutral density (ND) = Log10(100/Y); (where Y > 0 < 100);

7.1

Nominal HR, SC and HC ND Aim Values

HR, SC and HC are expressed as ND values with paper ND subtracted. The chart below shows nominal HR, SC and HC values (with paper density subtracted) for ISO-standard commercial (GRACoL) printing or proofing.

Gray Patch 25% (HC) 50% (HR) 75% (SC)

ND (minus paper) CMY K


0.25 0.54 0.90 0.22 0.50 0.90

Nominal gray patch ND aim values (minus paper) for normal GRACoL ink densities

7.2

G7 Highlight Range (HR)

HR is a single measurement that quickly tests the mid-tone density of a G7-calibrated device during production, for example while a press is running or after a proof is made. HR replaces individual TVI readings as the primary measure of G7 print weight or lightness / darkness. HR is computed twice, once for CMY ('HR_cmy') and again for black ('HR_k'). HR gives a more reliable indication of lightness / darkness and contrast than traditional TVI (dot gain) measurements, because HR is an absolute measurement, directly related to visual appearance, while TVI is a relative calculation with only indirect correlation to visual appearance. 7.2.1 Measuring HR_cmy HR_cmy is computed by measuring the neutral density (ND) of a combined CMY gray patch (50c, 40m, 40y) and subtracting the neutral density of paper. HR_cmy = ND(50c,40m,40y) - ND(paper) For devices with a dynamic range of about 1.3 ND or higher, HR_cmy is effectively a constant of 0.54. For devices with significantly lower or higher dynamic ranges, HR_cmy may be slightly lower or higher. (See 6.4 HR, SC and HC Aim Values vs. Dynamic Range.) 7.2.2 Measuring HR_k HR_k is computed by measuring the ND of a 50k patch and subtracting the neutral density of paper. HR_k = ND(50K) - ND(paper) For devices with a dynamic range of about 1.3 ND or higher, HR_k is effectively a constant of 0.50. For devices with significantly lower or higher dynamic ranges, HR_k may be slightly lower or higher. (See 6.4 HR, SC and HC Aim Values vs. Dynamic Range.)

16

The G7 Specification 2008, Final Working Draft for Public Comment

7.3

G7 Shadow Contrast (SC)

SC is a, quick way of checking the NPDC in neutral shadow tones during production. SC replaces traditional CMYK Print Contrast readings as the primary measure of G7 shadow-tone weight or lightness / darkness. SC is computed twice, once for CMY ('SC_cmy') and again for black ('SC_k'). 7.3.1 Measuring SC_cmy SC_cmy is computed by measuring the neutral density (ND) of a combined CMY gray patch (75c, 66m, 66y) and subtracting the neutral density of the paper. SC_cmy = ND(75c,66m,66y) - ND(paper) SC_cmy is nominally 0.90 for commercial printing but the exact value depends on the dynamic range of each device. (See 6.4 HR, SC and HC Aim Values vs. Dynamic Range.) 7.3.2 Measuring SC_k SC_k is computed by measuring the ND of a 75% Black patch and subtracting the neutral density of the paper. SC_k = ND(75k) - ND(paper) SC_k is nominally 0.90 for commercial printing but the exact value depends on the dynamic range of each device. (See 6.4 HR, SC and HC Aim Values vs. Dynamic Range.)

7.4

G7 Highlight Contrast (HC)

HC is a quick way of checking the NPDC in neutral highlight tones during production. HC is computed twice, once for CMY ('HC_cmy') and again for black ('HC_k'). 7.4.1 Computing HC_cmy HC_cmy is computed by measuring the ND of a combined CMY gray patch (25C, 19M, 19Y) and subtracting the neutral density of the paper. HC_cmy = ND(25c,19m,19y) - ND(paper) HC_cmy is a constant of 0.25 (paper subtracted) for all printing processes. 7.4.2 Computing HC_k HC_k is computed by measuring the ND of a 25% Black patch and subtracting the neutral density of the paper. HC_k = ND(25k) - ND(paper) HC_k is a constant of 0.22 (paper subtracted) for all printing processes.

7.5

HR, SC and HC Aim Values vs. Dynamic Range

The graph below shows that HR and HC are effectively constants for normal offset density ranges, but SC (and to some extent HR) vary as a function of density range.

17

The G7 Specification 2008, Final Working Draft for Public Comment

Graph showing HR, SC, and HC aim values for devices with different dynamic ranges

7.6

Expressing HR, SC and HC in L*

Instead of expressing HR, SC and HC in neutral density (ND) they could theoretically be expressed as CIELab L* values. This is convenient if gray balance and tonality are to be measured simultaneously with one CIELab reading. The following chart shows the nominal L* values for HR, SC and HC for a paper with an L* of exactly 95 L*.

Gray Patch 25% (HC) 50% (HR) 75% (SC)

L* (including paper) CMY K


75.7 57.5 39.5 77.7 59.9 39.7

Nominal gray patch L* aim values for paper with a 95 L* value Unfortunately it is not easy to re-compute the L* values for HR, SC or HC for paper having nonstandard lightness without some dedicated software such as IDEAlink Curve (which calculates modified L* values automatically in the Create Curves window). Individual users wishing to calculate L* values for HR, SC and HC values on papers other than L* 95 must first convert CIELab_L to CIEXYZ_Y, then scale the CIEXYZ_Y value according to the differences in paper CIEXYZ_Y, then convert the new CIEXYZ_Y value back to CIELab_L.

18

The G7 Specification 2008, Final Working Draft for Public Comment

G7 Compliance

When calibrating an offset press with non-standard inks, or a non-offset processes (e.g. flexo or screen printing) G7s normalization of the gray scale may not be enough to produce a good match to a standard color space, and may have to be augmented by some form of color management, for example ICC profiles. For this reason this specification proposes four levels of G7 compliance: G7 Grayscale, G7 Targeted, G7 Characterized and G7 Extreme, which are explained in this section.

8.1

G7 Grayscale

A device or process in G7 Grayscale compliance when it is calibrated to the basic G7 definition of constant neutral grayscale appearance. The G7 Grayscale specification defines a constant neutral grayscale (in colorimetric terms) relative to substrate color and relative to the maximum available neutral density (minimum neutral lightness) of the process. The G7 Grayscale specification defines the appearance of neutral image components, e.g. grayscale images, regardless of colorant hue or saturation levels, but for that reason does not specify the appearance of non-neutral components, e.g. colored images.

8.2

G7 Targeted

A device or process is in G7 Targeted compliance when it is G7 Grayscale calibrated and meets standard colorimetric measurements for substrate and colorants, such as those in ISO 12647-2. The G7 Targeted specification not only defines neutral appearance, but also the appearance of solid colors (100% each of cyan, magenta, yellow, and their two-color overprint combinations, red, green, and blue.) G7 Targeted does not define the exact appearance of a printed color image. Other variables such as ink trapping, ink opacity, ink film variations, screening system, etc., may cause differences in colors that exist between the defined neutrals and solid colors, even when the G7 Targeted specification is adhered to exactly.

8.3

G7 Characterized

A device or process is in G7 Characterized compliance when it is G7 Grayscale calibrated and is controlled (for example with ICC profiles) to match a print specification that is itself based on G7. A G7 Characterized process should closely match the appropriate print specifications intended appearance. A typical G7 Characterized process would be a GRACoL- or SWOP- certified proofing system that matches the CIELab values of an IT8.7/4 target within the tolerances defined by ISO and IDEAlliance.

8.4

G7 Extreme

A device or process is in G7 Extreme compliance when it is G7 Grayscale calibrated, and has a significantly greater natural color space than any G7 Characterized specification, and can be optionally controlled (for example with ICC profiles) to match any G7 Characterized color space. G7 Extreme refers to the application of G7 calibration to a device with significantly expanded color gamut compared to normal commercial offset printing. For example, many ink-jet printers have much greater natural color gamuts than GRACoL and would be unnecessarily restricted if forced to match GRACoL or any other G7 Characterized color space, just to comply with G7. Similarly, when an offset press achieves an expanded gamut through more saturated inks, extra plates, higher ink densities, etc., that press can still be calibrated to the G7 Grayscale requirements, and the resulting condition can be referred to as G7 Extreme.

19

The G7 Specification 2008, Final Working Draft for Public Comment In many cases a G7 Extreme device will produce pleasing results from standard GRACoL or SWOP files without further color management, but colored areas will tend to be more saturated than a standard proof. When a closer match to a G7 Characterized color space is required, ICC profiles or some other means of color management can be used, either locally for specific elements, or globally for the whole document.

20

The G7 Specification 2008, Final Working Draft for Public Comment

Appendix A: Characterization Data Sets Based on G7


In addition to providing a new set of process controls to control the tonality and neutrality of a printed page, G7 is being used to develop a family of characterization data sets (i.e. profiles) that for the first time have a shared visual appearance. The new IDEAlliance characterization data sets for GRACoL (TR006) and SWOP (TR003 and TR005) were the first data sets developed using G7 gray balance and print curves as their basis. When compared visually, GRACoL and SWOP prints have very similar visual appearances because the core neutrality and the tonality of the two color spaces are virtually identical. The FTA (Flexographic Technical Association) has proposed a new characterization data set (TR007) for packaging which also is built upon G7. A new global grassroots movement has been initiated to develop a set of global characterization data sets that are built on G7 and also require constant hue angles. IDEAlliance enthusiastically supports such efforts. As the family of G7-based characterization datasets grows, print buyers will experience easier file exchange and image matches across different printing types and substrates, thanks to the shared visual appearance inherent in G7 specifications.

A.1

About the Methodology

Defining reference characterization data for common print conditions has become universally compelling with advances color management tools for digital proofing and plate-making systems. The complement of colorimetric references and software tools has made it possible to produce pre-press proofs that consistently predict the intended print process. Likewise, with computer-to-plate, printing systems can be calibrated to these references to optimize the quality and efficiency their color reproduction processes as never before. The success of the CGATS-TR001-1995 characterization data set in representing SWOP printing on coated grade #5 paper has been instrumental in demonstrating this value and has motivated the creation of new data sets for other print conditions. In 2004, GRACoL under the auspices of IDEAlliance, provided an open forum for industry experts to develop a new characterization for commercial sheet fed printing on Grade #1 paper. This would replace a proposed characterization (DTR004) that failed to meet industry acceptance. This initiative was extended by SWOP to update (or replace) TR001 for SWOP printing on Grade #5 paper, and to develop a new characterization for SWOP printing on Grade #3 paper. Between 2004 and 2006, during the course of performing press tests and analyzing data, new approaches to developing characterizations evolved and were adopted based on the need to meet a diverse set of technical requirements and industry needs. This paper will review details of this development, which resulted in the Gracol2006_Coated1, SWOP2006_Coated3 and SWOP2006_Coated5 data sets. In 2007, CGATS SC4 documented these data in TR006, TR003 and TR005 respectively.

A.2

A New Approach For Developing Reference Characterizations

Experience in developing new reference characterizations for print processes has revealed a myriad of challenges in this multivariable estimation problem. The concept of averaging many new press tests to obtain a sort of industry mean is often impractical. Alternatively, attempting to conduct a single new press test that achieves ideal parameter aims invariably includes imperfections due to noise, process variation and biases of individual press systems. The approach described here is an alternative that provides a framework for continuous improvement and controlled transition: (a) Define colorimetric aims for the outer printing gamut. This includes primary solids, two color solids, and the paper and should be based on industry specifications, adopted standards, or actual press data.

21

The G7 Specification 2008, Final Working Draft for Public Comment (b) Define colorimetric aims for 3-color (CMY) and K tone scales. This departs from the usual method of establishing tonal aims (e.g. TVI) for the C, M, Y and K single color scales. The selected aims, termed Neutral Print Density Curves (NPDC), are described in G7 process documentation downloadable at www.gracol.org. Research had shown that the NPDC is a good representation of the 3-color and K tone curves produced by offset printing with linear CTP plates on coated stocks. (c) Obtain an initial colorimetric characterization. This should be derived from press data (existing or new) known to be good representation of the intended print conditions and the aims in (a) and to have reasonably smooth transitions throughout the CMYK color space. (d) Optimize this initial characterization. Use color management tools to achieve a close colorimetric match to the desired aims in (a) and (b) and to provide a high degree of smoothness throughout the CMYK color space. (e) Evaluate effectiveness the optimized characterization. Determine if press data from typical press systems operating under the intended reference print conditions can achieve an excellent colorimetric match using only simple 1-D plate curves adjustments. If clear shortcomings are identified, incorporate improvements into the characterization by repeating steps (a) (d).

A.3

Development Of The Gracol2006_Coated1 Data Set

From 2005-2006, seven press tests were conducted on Grade #1 paper with ISO2846 inks, and at various sheetfed shops. IT8.7/4 data sets were measured with white backing, filtered for outliers and averaged as Lab data. During the course refining the data to have NPDC characteristics, it was noticed that the draft version of the Fogra39 data set had a very similar gamut, with maxdE94 of the chromatic solids under 2. It was also observed that the Fogra39 set had been precisely adjusted to match all solid colorimetric aims of ISO12647-2-PT1 and was very smooth. In the interest of supporting this standard, and since the Gracol press test data had a K density that was much heavier than the aims, it was decided to use the Fogra39 set as a starting point. The procedure is as follows: 1. Initial data. The draft 2006 version of the Fogra39 data set was developed by FOGRA as an improved version of an ISO12647-2-PT1 characterization. Though it had been adjusted to the desired solid colorimetry and had been smoothed, it exhibited TVI characteristics specified in 12647-2 for legacy positive film based plating. This was not consistent with the desired NPDC for the 3-color and K scales. 2. Adjustment to achieve a NPDC. This is the most complex portion of the development process and consists of the following steps using commercial software tools. a. Gretag Profilemaker. Create initial ICC profile with Fogra39 data. b. Chromix ColorThink. Calculate Lab values for P2P CMYK reference list using the initial profile. The P2P reference is a set of near-neutral CMY and K values and is downloadable from www.gracol.org. c. IDEAlliance IDEALink Curve. Import the P2P Lab data file from (b) and calculate 1-D CMYK adjustment curves. These curves could be applied to the CMYK reference values of the original characterization to implement the adjustment. However, since we would like to use standard IT8.7/4 reference values it is necessary to find the corresponding adjustment to the Lab values. This is accomplished in the next two steps. d. Adobe Photoshop. Open 1 pixel/patch image file of IT8.7/4 target and apply 1DCMYK adjustment curves that were derived in (c.) e. Chromix ColorThink. Open adjusted IT8.7/4 image file and apply initial ICC profile from (a) in absolute mode. Then convert this image file to a list and save only the adjusted Lab data set (not the adjusted CMYK values). f. Microsoft Excel. Merge adjusted Lab data set with and IT8.7/4 CMYK reference list to form the adjusted characterization data set.

22

The G7 Specification 2008, Final Working Draft for Public Comment g. Verify. Repeat steps (a) (c) to see if the NPDC has been adequately achieved, i.e. if suggested adjustments are less than 1%. 3. Final data smoothing. Heidelberg PrintOpen 5.1 was used to apply a intelligent smoothing at a medium level to improve tonal transitions throughout the CMYK color space. The overall effect of this smoothing caused shifts with average dE<1.

A.4

Development Of The SWOP2006 Data Sets - Overview

Attempts to define new SWOP reference characterizations began in 2004 with sheetfed tests on #5 and #3 grade paper. These results were somewhat unsuccessful as they did not achieve aims as closely as desired and exhibited gray balance inconsistency compared to previous SWOP press runs. The shortcoming clearly illustrated the importance of incorporating well-defined 3-color tone and neutral balance curve into characterization data development. Web press tests on #5 and #3 sheets were performed at 3 sites (Quad, Brown, RIT) in 2006 with pretests to derive curves for achieving NPDC in the 3-color and K tone scales. Much of the data exhibited the common difficulties encountered with web press testing -- measurement integrity, smoothness, inter- and intra-sheet uniformity, and printing artifacts. Ultimately, attempts to derive characterization data solely from the new press data required significant adjustments and smoothing and yielded data questions of consistency, integrity and repeatability. However, several of the new data sets suggested that the web offset process could indeed achieve a gamut boundary (colorimetry of primary and 2-color solids) quite consistent with TR001. With this observation, it was decided to choose a set of primary and 2 color solids aims for the new SWOP #5 characterization that were close to those of TR001, with adjustments to correct the slight shortcomings known to exist in the TR001 data set. The primary and 2-color solids for the new SWOP #3 characterization were designed as small adjustments (< 4 deltaE) to the SWOP #5 aims based on the observed differences between #5 and #3 gamuts in the new web press tests.

A.5

Details For Generating SWOP2006_Coated5


1. Initial data. It was decided to extract an initial data set from the Adobe USWebCoated(v2) profile since it had been derived from TR001, was known to be smooth and have good separation characteristics. 2. Create initial data set with desired gamut boundary. Heidelberg PrintOpen was used to provide the slight adjustment to the primary, 2-color solids and selected paper aim which was L=90 a=0 b=4. This software provides a smooth, tapered adjustment that from gamut boundary through midtone area. The colorimetry of the C=M=Y=100 point was also adjusted to have a value of a*=b*=0. This is an aim that is recommended in ISO12647-2 and is useful for the purposes for providing an anchor point for incorporating the NPDC into the data. 3. Adjustment to achieve NPDC. This adjustment is performed with the same steps described in the previous section for Gracol2006_Coated1, except that the initial data set is that created in step 2. 4. Final data smoothing. Heidelberg PrintOpen 5.1 was used to apply a intelligent smoothing at a moderate level to improve tonal transitions throughout the CMYK color space. The overall effect of this smoothing caused shifts with average dE<1.

A.6

Details for Generating SWOP2006_Coated3

Details are essentially the same as for SWOP2006_Coated5, except in step 2 the gamut boundary is adjusted to the selected SWOP#3 aims for primary and 2-color solids, and a paper aim selected to be L=92.5 a=0 b=0.

A.7

Example Verification Of New SWOP Characterizations

To see if the resultant SWOP characterizations can be realized by actual press data, averaged IT8.7/4 measurement data from two of the press tests (Quad #3 and Quad #5) were smoothed,

23

The G7 Specification 2008, Final Working Draft for Public Comment adjusted to have NPDC via the procedure described previously and then white point normalized using the procedure in Annex 1 of CGATS.5. The resulting adjusted data sets were compared to the final SWOP2006_Coated3 and SWOP2006_Coated5 characterizations respectively using Delta-E76 with results shown in the table. Adjusted Quad#5 vs SWOP#5 Avg dE Full IT8.7/4 Max dE Full IT8.7/4 Max dE 1 and 2-color solids This level of match is comparable to that required for certification of SWOP proofing systems lending support to the notion that this particular web offset systems were capable of closely matching the new SWOP characterizations using appropriate 1-D plating curves. 2.7 3.5 6.1 @ C=M=Y=100 5.2 @ C=M=Y=100 1.5 Adjusted Quad#3 vs SWOP#3 1.5

A.8

Conclusion

The practical constraints of time and resources, the challenge to create a set of characterizations that have common appearance characteristics, and the pressure of an anxious industry in transition to a digital world led to a development that is arguably a departure from a typical standards process. However, results seem to have justified the means. Use of NPDC for designing characterization data sets has promoted: (a) visual similarity across families of similarly derived characterizations (b) common separation behavior from profiles based on such characterizations (c) similarity in plating curves used to match such characterizations. (d) proposals to incorporate comparable methodology into formal standards (e) new software developments by vendors to incorporate near-neutral calibration tools by into color management and process control software. The adoption of the new characterization data sets has led to unprecedented consistency and matchability for color proofs being supplied from preparatory organizations to print manufacturing sites. This is not to say that these references are optimal. However, we appear to have an effective starting point and process by which improvements can be made on a rational, incremental basis that can avoid confusing radical transition.

24

The G7 Specification 2008, Final Working Draft for Public Comment

Appendix B: Fractional Percentage Calculations


B.1 The Problem With Integer Percentages
Most integer (whole-number) percentages have no exact 8-bit equivalent. File values expressed as integer percentages therefore often result in slightly different equivalent percentage value when rendered to an 8-bits per channel image (assuming no error diffusion or 'dither' is applied). To eliminate this problem, all G7 formulae, software and documentation express halftone dot percentages as if they had been rendered into 8-bits without dither using the formulae below.

B.2 B.3

Converting Integer Percentage to Integer 8-Bit (Excel)


Int_8bit = ROUND(255 - int_% * 2.55, 0);

Converting Integer 8-Bit to Fractional Percentage (Excel)


Frac_% = ROUND(100 - Int_8bit / 2.55, n); (Where 'n' is the required number of decimal places)

B.4

Converting Integer Percentage to Fractional Percentage


Frac_% = ROUND(100 - ROUND(255 - int_% * 2.55, 0) / 2.55, n) (Where 'n' is the required number of decimal places)

The formulae (1) convert an integer percentage to nearest integer 8-bit (0 - 255), then (2) convert integer 8-bit back to nearest fractional percentage, with optional rounding. On the following page is a chart showing some common integer percentages converted to nearest 8bit equivalents with various levels of precision.

25

The G7 Specification 2008, Final Working Draft for Public Comment

B.5

G7 Fractional Percentage Conversion Chart


Int_% 0 1 2 3 4 5 6 7 8 9 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 91 92 93 94 95 96 97 98 99 100 Int_8bit 255 252 250 247 245 242 240 237 235 232 230 217 204 191 179 166 153 140 128 115 102 89 77 64 51 38 26 23 20 18 15 13 10 8 5 3 0 (1) 0.0 1.2 2.0 3.1 3.9 5.1 5.9 7.1 7.8 9.0 9.8 14.9 20.0 25.1 29.8 34.9 40.0 45.1 49.8 54.9 60.0 65.1 69.8 74.9 80.0 85.1 89.8 91.0 92.2 92.9 94.1 94.9 96.1 96.9 98.0 98.8 100.0 Frac_% (2) 0.00 1.18 1.96 3.14 3.92 5.10 5.88 7.06 7.84 9.02 9.80 14.90 20.00 25.10 29.80 34.90 40.00 45.10 49.80 54.90 60.00 65.10 69.80 74.90 80.00 85.10 89.80 90.98 92.16 92.94 94.12 94.90 96.08 96.86 98.04 98.82 100.00 (Decimal (3) 0.000 1.176 1.961 3.137 3.922 5.098 5.882 7.059 7.843 9.020 9.804 14.902 20.000 25.098 29.804 34.902 40.000 45.098 49.804 54.902 60.000 65.098 69.804 74.902 80.000 85.098 89.804 90.980 92.157 92.941 94.118 94.902 96.078 96.863 98.039 98.824 100.000 places) (4) 0.0000 1.1765 1.9608 3.1373 3.9216 5.0980 5.8824 7.0588 7.8431 9.0196 9.8039 14.9020 20.0000 25.0980 29.8039 34.9020 40.0000 45.0980 49.8039 54.9020 60.0000 65.0980 69.8039 74.9020 80.0000 85.0980 89.8039 90.9804 92.1569 92.9412 94.1176 94.9020 96.0784 96.8627 98.0392 98.8235 100.0000

Some common percentages converted with various levels of precision

26

The G7 Specification 2008, Final Working Draft for Public Comment

Appendix C: Tolerance Notes


C.1 Introduction
One of the most difficult things to define in a printing specification is the allowable tolerance for each parameter - partly because acceptable tolerances for one user may be unacceptable to another, and partly because realistic tolerances often depend on the type and condition of the printing system in use. A single set of color performance tolerances cannot be applied (fairly) to all devices or processes, if for no other reason than the wide variation of native stability and consistency between devices. Obviously more stable devices or processes have an advantage when high accuracy and repeatability are needed, but frequently this higher stability is paid for in slower production or higher material costs, often negating the original advantage. For example, a typical pre-press proofing system or wide-format ink-jet printer is typically much more stable than offset lithography, and will usually score higher over time against a rigid set of tolerances, but the relative efficiency of offset printing, combined with the use of standard materials, proper operator training and automated press control, regularly satisfies the vast majority of print buyers. One valid concern is that applying blind tolerances for all the possible solid colors to every print run is harder than traditional printing, where only the colors in the particular job had to be matched.

C.2

When Tight Pressroom Tolerances are NOT Appropriate

The overwhelming success of offset printing in relatively high-quality imaging applications, in spite of relatively low predictability, indicates that rigid measurable tolerances may not be a great advantage in certain traditional workflows where either the proof is not a reliable constant, or press conditions or customer requirements are uncertain. If the only available pressroom guide is a customer-supplied 'proof' of unknown origin or quality, pressroom tolerances should be tight enough to get the press close to the aim point, but loose enough to allow room for traditional skill-based proof matching. Having said this, any printer serious about quality and economy should attempt to reduce or eliminate uncertainties from the pressroom, which is probably why you are reading this.

C.3

When Tight Pressroom Tolerances ARE Appropriate

Tighter pressroom tolerances are not only appropriate, but essential for efficient, automated printing, but will only produce optimum results when ALL the following conditions are met; The press uses standard media (paper, ink, blankets, fountain, etc.) Plates have been calibrated with the G7 method Every proof submitted to the pressroom has been created using color management to match EITHER the standard characterization data to which the press was calibrated OR a custom profile of that particular press All CMYK files are separated with an ICC profile from the same characterization data referred to in the previous point The press is equipped with effective automated ink control systems that have been tested to work with G7 gray balance and NPDC target values

C.4

G7 Tolerances vs. Custom User Tolerances

Where possible, G7 uses the tolerances suggested in ISO 12647-2, for example on solid ink Lab values and paper color. For new parameters like gray balance, NPDC, or characterization data where no official ISO tolerance currently exists - G7 provides, or is working on, SUGGESTED tolerances as a starting point, but until these are tested and proven in practice, they should be interpreted (and if necessary adjusted) to fit the capability and expectations of each individual user. NOTE: Print buyers should remember that tighter tolerances equate to higher costs.

27

The G7 Specification 2008, Final Working Draft for Public Comment One way to determine realistic custom tolerances for an individual user or process is to calculate the standard deviation of each parameter over time, when process control is tightly maintained.

C.5

Characterization vs. Production Tolerances

Characterization tolerances should be stricter than production tolerances because a characterization print must be as near to 'perfection' as the device is capable of. This accuracy will usually not be achieved in a single press run, but may be achieved by averaging several prints made on different days. A good general rule is that characterization tolerances should be roughly half as large as production tolerances, or at least as strict as the most stringent production aims. For example, if gray balance tolerances for production work are +/- 1 a* and +/- 2 b*, characterization tolerances for that device should be +/- 0.5 a* and +/- 1 b*. If gray balance (and other parameters) cannot be controlled to these tolerances on the characterization print, the resulting characterization will probably NOT represent the average capability of that press.

C.6

Printing vs. Proofing Tolerances

Most pre-press proofing systems are much more stable and often have a higher color gamut than the typical offset press, partly by avoiding wet-trapping issues (see G.6.1). This alone makes it reasonable to apply higher tolerances to proofing systems than to production offset presses, but another more important reason is that in most pressrooms the proof is still the ultimate 'target' (regardless of measured values), and a 'moving target' will obviously lead to less consistent printing. G7 recommends that proofing tolerances be as fine as possible, regardless of pressroom capabilities, to minimize proof-to-proof variations. Anything less would make the proofing system a moving target. Actual local proofing tolerances should be determined by trial-and-error or statistical analysis, along with an understanding of local process, control capabilities, and limitations. Wider tolerances may be necessary on coarser grades of paper.

G.7

Wet-Trapping - the Bane of Offset Lithography

Modern offset presses print all four (or more) inks in rapid succession, 'wet-on-wet'. The first inks down usually adhere to the paper better than later inks, which are partially repelled by the still-wet inks printed earlier. In some cases the later inks actually remove some of the earlier inks, depending on the relative tack (stickiness) of each ink in the sequence. Either way, the amount of one or more inks remaining on paper is usually less than would be achieved on a 'dry trapping' press, (printing one color at a time, or with a drying system between each unit), and the visual effects are typically unpredictable - often varying as the press warms up over the course of a run. 'Wet-trapping' introduces two problems that are largely absent in slower, single-color presses namely reduced color gamut and unstable performance in darker tones - in each case due to the loss of some wanted ink. In fact the relative instability of wet trapping in 3-color and 4-color areas is one of the main problems in matching multiple presses to a standard characterization data set, even when each press uses the same paper and ink. Another trapping-related problem with many offset presses is that gray balance is often seriously affected by micro register changes from sheet to sheet, which create microscopic area wet trapping differences from sheet to sheet within the halftone pattern itself, and consequent short-term gray balance errors. This makes it dangerous to demand tight gray balance tolerances on an offset press - at least on any single sheet. To measure gray balance reliably on an offset press, or to calculate separate RIP correction curves for CM and Y, be sure to average several sheets in succession.

28

The G7 Specification 2008, Final Working Draft for Public Comment

Appendix D: G7 Glossary
calibration, printer Characterization data Adjusting a printing device to a known and repeatable condition - typically by using consistent substrate and colorants, and by adjusting tonality, gray balance and other metrics to a constant measurable state A file of CIELab or CIEXYZ or spectral measurements of a printed target from which, if the corresponding colorant values of each patch (e.g. CMYK percentages) are also known, an ICC profile can be created (Commission Internationale de l'Eclairage) an international group formed in the early 20th century (still active) to study and numerically define human color vision. Described as follows at www.cie.co.at/cie;
"..an organization devoted to international cooperation and exchange of information among its member countries on all matters relating to the science and art of lighting. The CIE is a technical, scientific and cultural, non-profit autonomous organization. It has grown out of the interests of individuals working in illumination. Since its inception 90 years ago, the CIE has become a professional organization and has been accepted as representing the best authority on the subject and as such is recognized by ISO as an international standardization body."

CIE

CIEXYZ

CIELab (CIEL*a*b*; Lab) CIELCH (LCH)

CMY CMYK Colorimetric CtP D-50 Delta-E (Delta-E*, E*ab)

Density Dot Gain Fogra

G7

GRACoL

GRACoL2006_Coated1 GRACoL 7

A standard CIE color space expressed in variables X, Y and Z, corresponding to the 'visual matching functions' derived from the original CIE color vision experiments A standard CIE color space derived from CIEXYZ where L* = Lightness, a* = the green-red vector and b* = the blue-yellow vector in a mathematically cubic space A more visually-intuitive interpretation of he standard CIELab color space in which a* and b* values are replaced by the radial vector C (Chroma or saturation) and the angular measure H (Hue) Cyan, Magenta and Yellow Cyan, Magenta, Yellow and Black Based on CIE color science (as opposed to densitometry) Computer to Plate - a plate making system which eliminates the need for intermediate positive or negative films Industry - standard viewing light color expressed as a complete spectral energy distribution curve Standard CIE formula for expressing a color difference between CIELab readings. Several improvements exist, including E*94, E*CMC and E*2000, but the original formula; E*ab = (L*2 + a*2 + b*2)0.5 is simplest to calculate and adequate for the purposes of this document. Log10 (1 / reflectance) See 'Tone Value Increase' www.fogra.org- German Graphic Technology Research Association whose purpose is to promote research into, and the development and application of printing technology, and to make the results available to the printing industry. www.g7global.org The G7 specification provides formal colorimetric definitions for neutrality and tonality that together control the visual appearance of the grayscale component of an image. The GRACoL and SWOP specifications are both based on G7 gray balance and tonality. 'G' denotes the focus on 'Grays', while '7' refers to the seven colorimetric ink values of Cyan (C), Magenta (M), Yellow (Y), Black (K), Red (M+Y), Green (C+Y), and Blue (C+M). www.gracol.org- An IDEAlliance Working Group focusing on sheetfed offset commercial printing. GRACoL stands for General Recommendations for Applications in Offset Lithography Official characterization data for commercial offset printing released in 2007 by the GRACoL committee Seventh edition of the GRACoL document and specifications

29

The G7 Specification 2008, Final Working Draft for Public Comment


Gray (Grey) A tone or image element without any visual chroma or color. The universallyaccepted spelling 'Gray' is adopted for this document in preference to the equally-acceptable 'Grey' due to simple phonetic logic. (Samuel Johnson has a lot to answer for!) 1. The relationship of C, M and Y percentages or necessary to make a theoretically neutral gray - usually defined for C = 50, (e.g. 50c, 40m, 40y) but also given for other gray scale steps in this document - (see Appendix E: Gray Balance) 2. The colorimetric definition of gray in CIE units - (typically zero a* and zero b* but depends on lighting conditions and personal preference) 3. The condition in which a color or image is said to be 'neutral' or uniformly gray (with no apparent coloration) when viewed under standard lighting A target of various different near-neutral patches from which the natural gray balance of a printing system can be determined See Highlight Contrast A quick process control check of NPDC in lighter tones. Density of a 25% patch minus paper density. A quick process control check of NPDC in mid-tones. Density of a 50% patch minus paper density. see 'Highlight Range' A device-specific translation table between the PCS (profile Connection Space e.g. CIELab) and the device's native color units, e.g. CMYK or RGB, and saved in the standard file format defined by the ICC International Color Consortium - the organization responsible for defining standard file structures and rules of modern color management www.idealliance.org- International Digital Enterprise Alliance - US graphic industry organization that has been developing, educating and validating best practices in publishing and information technology for 40 years A software utility to simplify G7 calibration without the tedium and uncertainty of manual curve drawing. Using measurements from the P2P target calculates aim values that can be applied to the calibration tables of any RIP or software driver to simulate the G7 NPDC and Gray Balance condition. International Standards Organization A standard for "Graphic technology - Process control for the production of halftone colour separations, proof and production prints - Part 2: Offset lithographic processes". ISO TS 10128 Methods for calibrating a printing system with digital data (presently in draft). This new standard includes three options for achieving the aim values in ISO 12647: the near-neutral method as exemplified by G7, a density TVI method as exemplified by the FOGRA/ECI/bvdm process, and a device link method submitted by the Inernational Color Consortium (ICC). All of these are viable, interchangeable and compatible with one another as long as good printing is practiced. ISO ink standard based on laboratory measurement of samples of a specific inkfilm thickness measured on a specific substrate ISO viewing and lighting standards Most recent version in a series of standard CMYK characterization targets available in randomized and visual layouts See CIELab See CIELCH Neutral Density, or the density of a neutral gray (colorless) sample (see 'Density'). Calculated in G7 as Log10 (1 / CIEXYZ_Y) where 0 < Y < 1.0, or as Log10 (100 / CIEXYZ_Y) where 0 < Y < 100. The relationship between measured neutral density and original halftone percentages on a gray scale.

Gray Balance

GrayFinder target HC Highlight Contrast (HC) Highlight Range (HR) HR ICC profile

ICC IDEAlliance
IDEAlink Curve

ISO ISO 12647-2

ISO TS 10128

ISO 2846-1 ISO 3664 IT8.7/4 Lab (CIELab) LCH (CIELCH) ND

NPDC

30

The G7 Specification 2008, Final Working Draft for Public Comment


Number 1 sheet P2P Target Press2Proof Target Reference characterization RGB High quality coated white commercial offset paper having nominal colorimetry values of 95 L*, 0 a* and -2 b* See 'Press2Proof Target' (Also known as 'P2P Target') - a compact target used for analyzing NPDC and other variables during the calibration process A characterization data set (see Characterization data) representing the desired appearance of a device after applying G7 calibration and / or color management Red, Green and Blue - the so-called 'primary colors' of light. Also the abbreviations given to two-color solid ink patches, namely R = Magenta and yellow, G = Cyan and Yellow, B = Cyan and Magenta. Raster Image Processor Rochester Institute of Technology - provides (amongst other services) ISO 2846 laboratory ink testing See Shadow Contrast A quick process control check of NPDC in neutral shadow tones. Optional replacement for Print Contrast measurement. Density of a 75% patch minus paper density. Abbreviation of the Sin / Cosine formula used to calculate basic G7 NPDC curves A combination Spectrophotometer and Densitometer convenient for measuring the quality of press sheets and/or proofs An IDEAlliance Working Group focusing on publication printing. SWOP stands for Specifications for Web Offset Publications. The SWOP Specification is based on G7 neutrality and tonality. The specification, metric, device or characterization which is being imitated or aimed for Percentage dot area or halftone value on a scale of 0% (no ink printing) to 100% (maximum ink printing) Tone Value Increase (formerly known as 'Dot Gain') - the reflection halftone percentage measured on a printed sample minus the original halftone percentage file value that produced it See Tone Value See 'Tone Value Increase' Without any curves or look-up-tables (LUTs). A LUT or RIP in which input digital values are the same as output digital values

RIP RIT SC Shadow Contrast (SC)

SiCo Spectro-densitometer SWOP

Target Tone Value Tone Value Increase

TV TVI Un-calibrated

31