Fogra Softproof Handbook

8/20/2019 Fogra Softproof Handbook

1/43

Fogra

Softproof Handbook

Fogra Graphic Technology Research Association

Streitfeldstraße 19 · 81673 München

www.fogra.org

L*

**

ab

sRG BG amut D50

L C D ?


2/432

Tips and background information

for colour reliable work on screen

This handbook is the result of current investigations from the Fogra research project, “Creation and investigation of asoftproof workstation“. It reflects the most recent experiences and test results. Due to new software versions and relevantfindings, it has been planned to produce an updated version of this handbook in the near future. In the meantime, updatedcontent can be found on the website at: research.fogra.org. This website has information on current projects, and in thedownload centre, programs and further instructions on softproofing.

The publication of this handbook, especially in parts requires prior consent by Fogra. In general we welcome the circulationof the complete handbook.

We would like to especially thank the project acompanying committee for assisting us in this research project:

Michael Adloff twentyfour seven digitale pre press services gmbhThorsten Beermann K-Flow GmbHDr. Ralf Biering Mediahaus Biering GmbHNial Coady Targetcolour UKArmin Collong Eizo / Avnet Technology Solutions GmbHMarkus Cornely NEC NEC Display Solutions Europe GmbHChris Edge KodakFlorian Fejfar MAN Roland Druckmaschinen AGDetlef Fiebrandt Colour ConsultingMichael Gall JUST Normlicht GmbHStephane Georges Dalim

Johannes Haas Meyle-MüllerFranz Herbert ICS ColorDr. Patrick Herzog Onevision Software AGMarkus Hitzler Colour SolutionsDr. Hanno Hoffstadt GMGAndreas Holst NEC Display Solutions Europe GmbHStephane Jamin LaCieOliver Kammann K-FlowKevin Kotorynski KodakRaimar Kuhnen-Burger Quatographic Technology GmbHBodo Langkowski GTIWolfgang Maske DalimDaniel Mayer LaCie

Roland v. Oeynhausen Otterbach Medien KG GmbH & CoChristopher Parker NEC NEC Display Solutions Europe GmbHThomas Richard Richard EBV Bernd Rückert CGSCristina Stoll Rochester Institute of TechnologyFlorian Süßl MetaDesign AGRoland Thees IFRAAndy Williams IFRAShoichi Yamaguchi Eizo

Fogra Softproof Handbook Version 1

The project “Creation and investigation of a softproof workstation“ (Fogra Nr.10,047) was developed by the means of the federal ministry for economy andtechnology by the team of industrial research unions “Otto of Guericke“ e. V.(AIF) promoted (AiF Nr. N07316/06).


3/433

Contents Take Home Message ..............................................................................................................................................5

1. Parameters for Softproofing ..........................................................................................................................61.1 What is a Softproof? - Practical Scenarios .....................................................................................61.2 Fundamentals of Softproofing ............................................................................................................81.3 Setting Up a Softproofing Work Station ....................................................................................... 121.4 MAC or Windows: What is there to consider? ............................................................................ 151.5 Measurement Devices for Softproofing ......................................................................................... 171.6 Displays (Monitors) .............................................................................................................................. 19

1.7 Viewing Conditions: Evaluation of ”Standard Lighting” ........................................................... 24

2. Softproofing Settings in Software Programs ........................................................................................ 26

3. Calibration: Step-by-Step ........................................................................................................................... 30

4. Hands on: Softproofing of RGB and CMYK Images with Adobe Photoshop ................................ 34

5. Fundamentals and Concept Explanation ................................................................................................ 375.1 Description and Measurement of Light (Photometry) ............................................................... 375.2 Illuminant, Chromaticity, and Correlated Colour Temperature ............................................... 395.3 Gamma and its Disguises ................................................................................................................... 405.4 Matrix or LUT (Lookup Table Profile) ...............................................................................................41

6. Annex .................................................................................................................................................................42


4/434

“It works,

if you know what to do!“


5/435

Take Home MessageTarget Settings for Calibration*

RGB

Gamma: According to the

Working Space

Colour Temperature: 5000 - 6000 K

Luminance: 160 cd/m2

CMYK

Gamma: L*

Colour Temperature: 5000 - 6000 K

Luminance: 160 cd/m

Default Settings in Photoshop*

Example for RGB Printing Example for Conventional Printing

Print Preview*

* The indicated values are practical recommendations. They can be used as a working example in many cases. However, they do not represent the best values for all situations, andtherefore, are not to be regarded as a the “Fogra recommendation“. This guide provides extensive background knowledge and many tips for the optimal values to use.

+ visualadjustmenttowards theprint

Hint!Absolute colorimetric provi-des the correct CIELAB datawith the Info palette. Pleaseuse “Convert to Profile“ forimage separation, becausechanging the image mode (e.g.

Image > Mode) doesnot give you a pre-

view and control of

the conversionoptions.

Take Home Message


6/436

Today, even complex jobs are constantlybeing produced at a quicker rate in theprinting industry. Due to time and cost,high-quality printed product is moreand more required. Therefore, even asearly as the creative agency or dataentry, preparation with respect to thecolour results is to be carefully judged.For digital proofs, the Ugra/Fogra MediaWedge CMYK V.2/V.3 serves, nationallyand increasingly internationally, as acontrol strip. With this control strip,

it can be guaranteed that the proofis colour reliable for the embeddedprinting condition, such as FOGRA39.Additional information is found in Fograextra No. 16 „How about the colourreference in digital printing“ (http://fogra.org/products-en/download/Extra-16web.pdf). If there is no control wedgeon the proof or the tolerances areexceeded, a colorful print is consideredto be non-binding. This digital proofenables advantages over the classic

on-press proof, in terms of time, quality,and cost. Proofers, proofing material,colour measuring instruments, andsoftware are required for such a system;moreover, it is important to calibrateregularly.

In the daily comparison of proof andprint, it is assumed that the print willbe subjected to production tolerances.Even so, it is required that the techno-logy related to the simulation is at leastequal to the process it is simulating,

in particular, compared to the colourgamut to be simulated. With softproo-fing, this expense could be diminished,which could provide further cost andtime advantages. Softproofing can bemore concretely defined as the follow-ing:

“Representation of an image producedusing a monitor with the purpose ofshowing the results of the preparation(colour separation) process in such a

way that closely simulates the resultsof the intended output (printing) condi-tion.“ (ISO 12646)

The question remains about the mea-ning of “close“.

Colour-Matching Practice -Three ScenariosThe practice of colour-matching inthe graphic arts industry is very multi-faceted, but should be represented herein three typical scenarios. These differessentially through the object beingexamined and/or the respective viewingconditions.

Scenario IColour-Adjustment on Screen -“Photographer‘s Workflow“The colour-matching at the beginningof the production process is not yetdefined by a particular standard. Thecolour-matching object in this sce-nario is a digital image; regardless ofwhether it came from a digital cameraor was developed on screen, the task isto guarantee a colour accurate screen

representation of this digital file on ascreen that has been correctly adjustedand calibrated on a regular basis.

1. Parameters for Softproofing1.1 What is a Softproof? - Practical Scenarios

Figure 1: Typical viewing scenario in “Photographer‘sWorkflow“

Parameters for Softproofing - What is a Softproof? - Practical Scenario


7/43


8/438

Difference between colour conversionof data to the target colour space and

colour conversion for displayIt is important to understand the colourconversion chain in softproofing whichis explained in the following. All digitalfiles have a starting point, for example,from databases, digital photography, orcreation on screen. They normally usea device dependent colour space. Thisusually includes the use of greyscale(K), RGB, and CMYK colour values (e.g.driving values). To understand colourconversion, it is very important to diffe-

rentiate between the colour conversionof the file (e.g. separation from an RGBworking colour space to a CMYK colourspace for output) and displaying this“separated“ file on screen. The firststep is usually done by means of oneof two different rendering intents, eit-her perceptual or relative colorimetricusing black point compensation. The soconverted (separated) images are thento be converted to the display profile.This twofold conversion is illustrated

by the “Proof Setup“ dialogue box inPhotoshop.

First step: Data colour conversion totarget colour spaceThe settings in the upper black box inFigure 5 pertain to the colour conver-sion of the data to the target colourspace. This will often be an output pro-

file i.e for offset printing. In the exa-mple screenshot below, the perceptual

transformation is made with respect tothe printing condition FOGRA39, (e.g.offset printing in accordance with ISO12647-2). For this printing condition,we normally recommend data prepara-tion for typical subjects with the indus-try profile of the ECI: ISOcoated_V2.icc.

Second step: Colour conversion to thedisplayThe resulting CIELAB tristimulus valuesafter the first conversion are then sub-

ject for the conversion to the displayprofile. The CIELAB values are result ofan intermediate step and can not be„seen“ for them self. In a typical soft-proof setup one wants to simulate thepaper white of a substrate on screen.Therefore the conversion to the displayprofile will make use of the absolutecolorimetric rendering intent. This isachieved by selecting the option “simu-late paper color“ in the second blackbox in Figure 5. When no simulation of

paper white is needed the the relativecolorimetric intent will be choosen forthe display on the monitor. Deselect„Simulate Paper Color“ in this case. Inaddition deselecting „Simulate BlackInk“ in „Display Options (On-Screen)“will make use of Black Point Compen-sation for the colour conversion to thedisplay.

It should be noted that both the colour

conversion of the data to the targetcolour space and the following colourconversion to the display profile aredone on the fly and are not saved to thedata. In a later stage although the datanormally will be converted to the targetcolour space which was used in thesoftproofing setup explained here.

From driving values to colour valuesTo achieve a colours reliable displayon screen the data has to be defined

colorimetrically. Similar to the use of adictionary, through the use of ICC pro-files, it is possible to calculate device-independent CIE colour values fromdevice-dependent values such as RGBor CMYK.

1.2 Fundamentals of Softproofing

Figure 5: Photoshop Softproofing Dialouge Box: Data separation (upper black box) and transformation to the display (lower black box)

Note!In this document, for thesake of simplicity, the

driving values willbe designated as

colour values.

Parameters for Softproofing - Fundamentals of Softproofing


9/439

Image Data Representations:When a file is displayed on a monitor,from a colorimetrical point of view,there are four ways that the colour

values can be represented.

1. Monitor representation withoutcolour managementThe first way is to send the RGB valuesof the file without any changes to themonitor. This happens, for example,in programs, such as the WindowsDesktop, which is shown in Figure 6 as“direct representation“. It is suited forsituations in which colour is not bindingbecause based on the different colour

behavior of the devices, the same filewill always look somewhat different. Inthe next version of this handbook, wewill review which applications support acolour managed monitor.

Ergo:Preparation: noneRepresentation: RGB-Data displayed

unaltered („keep numbers“)

2. “Normal View“ in Adobe PhotoshopThis representation is used most fre-quently in the graphic arts industry.Adobe Photoshop is considered thedefacto software standard for photo-graphers. With this, the RGB values areconverted to the display profile with therelative colorimetric intent with blackpoint compensation. For example, theRGB values 227, 0, 15 in sRGB corre-spond to the CIELAB colour values 48,74, 60. In the background, a relativecolorimetric conversion to the screen

profile occurs. Here, the screen profile

of a MacBook Pro “Color-LCD.icc“ withblack point compensation delivers theRGB values 250, 0, 0. These values aresent to the graphics card. This leads

to an optimum („contrasty“) represen-tation in the sense of the utilizationof the available screen contrast. Forthe colour-binding representation ofprint files, this method is not suitablebecause the colour values of the file arerepresented as either very bright and/ordark because the photo contrast is com-pletely scaled to the monitor contrast.Due to this, a representation is displayedthat may possibly contain a darker blackand lighter white than the original data.

However it should be noted that thismethod achieves a colorimetric match ifa hardware calibration set the monitorwhite and black point to the CIEXYZ ofthe printing condition to be simulated.Practically this method is sometimesused to get a fairly accurate representa-tion with the „standard“ mode (normalview) of image represenation in AdobePhotoshop. This method although limitsthe versatility of the monitor calibra-tion, because all applications, regardless

if colourmanaged or not, will show the„simulated paperwhite. This approach isnot recommended in general.

In Figure 7, this effect is shown withthe example of black. On the left anabsolute colorimetric match of the blackpatch (CMYK = 0, 0, 0, 100) is shown.

On the right is the result of the “nor-mal-view“ representation. Due to this,the danger arises that shadow areas inthe photos will be incorrectly displayedand judged and therefore modified orleft untouched.

Ergo:Preparation: noneRepresentation: Relative Colorimetric with blackpoint compensation

Figure 6: Principles of colour conversion

Figure 7: On the left, printed black. Onthe right, monitor black.



10/4310

3. Colour accurate representationwithout printing process simulation:“Photographer‘s Workflow“For RGB retouching, Photoshop automa-

tically uses the monitor profile for imagerendering (using relative colorimetricincluding black point compensation).Therefore, no further intervention isnecessary. The monitor should cover asufficient colour gamut (wide gamutmonitors that can better cover AdobeRGB or ECI-RGB V1/V2). In addition, aprint preview can be used as necessary.For example, it can be used for thepaper white simulation using a personalinkjet printer (see figure 8). See Figure

9 to determine where to store settings.If you choose a profile before you openyour photos, it becomes the default (seeFigure 10). The standard printing previewcan be activated and deactivated withkeyboard shortcuts(Command + Y / Control + Y).

Ergo:Preparation: PerceptualRepresentation: Absolute Colorimetric

4. Absolute Colorimetric Representa-tion With Printing Process Simulation:“Reproduction Workflow“For the so called “Reproduction Work-flow“, the ICC profile of the intendedprinting condition is used as a simula-tion profile. For this processing (sepa-ration), there are two practically usedmethods; one is perceptual and theother is relative colorimetric with blackpoint compensation. With respect to thechoice of rendering intents, for example,in Photoshop (see Figure 11), it is to

be noted that the conversion of CMYKapplies virtually in the background.

Ergo:Preparation: Perceptual (see Figure 11) or relative colorimetric with black point compensationRepresentation: Absolute Colorimetric

Figure 11: “Proof Set-up“ dialogue box in Photoshop

Figure 9: Stored setting

Figure 8: Setting without name

Figure 10: Standard setting



11/4311

Proofing of CMYK DataIt often occurs that already prepared(separated) files must be reviewed onscreen. In this case, the processing and

correct representation must be done onscreen. In order to simulate the imagedata as it would be printed withoutcolour conversion, the option to “Pre-serve CMYK Numbers“ in Photoshopmust be activated. An example is shownin Figure 12.

Figure 12: Photoshop setting for the proofing of CMYK colour values



12/4312

The set-up process can be outlined in

the three following steps: calibration/adjustment, characterization, and pro-filing. An additional validation step isconsidered optional and will be descri-bed in Chapter 3.

Calibration/AdjustmentIn the first step, the hardware is adju-sted by using a colour measuring instru-ment and further methods (test files,reference prints, visual finetuning etc.)to meet the desired target (aim) values.

CharacterizationAfter the screen has been optimallyadjusted, test colours are displayed onthe screen and colorimetrically mea-sured. As a result, characterization datahas been produced (typically as tabula-ted data).

ProfilingIn the last step, the calculation of theICC monitor profile is done based on

the characterization data (measuredbefore). Furthermore, user dependentoptions are also considered in the pro-file creation.

In the following, these three steps aredescribed in greater detail.

Calibration: Hardware or Software?

CalibrationCalibration comprises all adjustments

and (active) methods necessary to reachdesired aim values such as white point,gradation, and luminance. Hereby, botha smooth rendition of the lightnessscale and the utilization of the techni-cal possibilities of the pertinent devicemust be taken into consideration.

Two Categories of RenderingThe representation of images on themonitor is influenced by many parame-ters, such as the type of screen, graphicscard, DVI cable, operating system and/ordriving software. Figure 15 shows typi-cal interfaces.In Figure 14, a logic classification of thedriving is shown. This symbolizes thecharacteristic flow of the colour trans-formation from the right to the left.The main components are defined in the

following:

ICC profile:First the image data is passed to the ICCprofile. This is done by the program (e.g.Photoshop) used including the utiliza-tion of the monitor profile made availa-ble by the operating system.

Graphics Card:Most graphics cards allow the individualmodification of the red, green, and blue

channels using different bit depths. Theconcept of how these are used is impor-tant for the overall quality of the entirecolour management chain.

Screen (display, monitor):Modern screens offer the possibilityfor hardware calibration of the displaypanels both with adjustment possibi-lities at the user interface (“On ScreenDisplay“) or via software activation.Hereby, the amount of the steps isimportant (e.g. 8, 10 or 12 bits) see

Figure 13, since it determines the accu-racy of the needed adjustments. It is ofgreat importance to optimize the inter-play of hard- and software (communi-cation for example via DDC) in order toachieve the desired target values (whitepoint, gradation, luminance).

A hardware calibration is, therefore,considered to be the preferred case ifthe necessary corrections (in order toachieve the aim values) can be donewhere the highest accuracy is available

(monitor LUT). Corrections should becarried out with the highest accuracy

Display Graphic Card ICC-Profile

Figure 14: Principle representation of the three most important components for the final image quality.

Original Data

D a t a a f t e r C o r r e c t i o n

Dark Light

Original DataDark Light

D a r k

L i g h t

D a t a a f t e r C o r r e c t i o n

D a r k

L i g h t

Figure 13: Comparison of 8-bit and 10-bit datarepresentation

1.3 Setting Up a Softproofing Work Station

Parameters for Softproofing - Setting Up a Softproofing Work Station

Calibration:Set of operations that establish,

under specified conditions, the

relationship between values of

quantities indicated by a mea-

suring instrument or measuring

system, or values represented by

a material measure or a reference

material, and the corresponding

values realized by standards [ISO

International Vocabulary of Basicand General Terms in Metrology]


13/4313

possible, particularly in the LUTs of thescreen.In case no hardware calibration isavailable the necessary correctionsmust be carried out consequently byusing the possibilites provided by the

graphics card or by means of the ICCmonitor profile. In the optimal case of ahardware calibration the transfer func-tions of the graphics card and the ICCdisplay profile are linear. Since graphicscard normally have a lower bit-depth(often 8-bit) than the monitor LUTs, thesoftware calibration is usually qualita-tively not as optimal than the hardwarecalibration (see Figure 16).

Figure 16: Comparison of hardware and software calibration

Hardware or Software calibration?

Hardware

SoftwareCurrent

Aim

R G B0

255

8745 K

R G B0

255

5000 K

R G B0

255

8745 KLost (nor more

usable)

All adjustments via

software (8-Bit)

R G B0

255

5000 K All adjustments via

hardware (8-Bit+)

v

The Shapes and types of video connector

Connectors on the monitor

PC Video Conectors D-Sub 15 DVI-I DVI-D

Macintosh D-Sub 15 pin

ADC1 2

DVI-I

Windows D-Sub 15pin

DVI-I

DVI-D

v

v

v

Figure 15: Form and types of different video connectors

Software calibration: Ona laptop there is mostlyno access to the screenLUTs.

Expert Tip:There are several DDC versions:In the simple version DDC1, theknown data of the monitor istransmitted unidirectionally andpermanently as so called EDID datato the graphics card. The informa-tion contains records as long as128 bytes, which standard solu-tions and DPMS modes support.The screen size and the manufac-turer statement are also known.The version DDC2 supports bidirec-tional communication and enablesinteraction (e.g. based on colourreading during the calibration, andtherefore, hardware calibration, thedisplay behavior can be altered).



14/4314

CharacterizationCharacterization links the device depen-dent control (driving) values to theircorresponding measured device ind-

pendent colour values. This means thatfor the screen measurement, the RGBvalues to be sent via the application tothe graphics card are compared to themeasured CIEXYZ or spectral data (seeFigure 17). This is the basis for the crea-tion of the monitor profile. The amountof the test patches to be used dependson the characteristic of the screen.Older cathode ray tubes (CRT) behaveroughly in a linear function, in the senseof the additive colour mixture. For this

reason, less test colours are necessaryfor a sufficient colorimetric descriptionin comparison with modern LCD screens.Without correction performed by themanufacturer, many LCDs show worselinearity, so for a sufficient characteri-zation several test colours are necessary.

ProfilingAt the last step, with the goal of anoptimum colour description (fingerprint)of the screen, an ICC profile is genera-

ted based on the characterization data.This means that both the forwards (RGBto CIEXYZ) and the backwards direction

(CIEXYZ to RGB) must contain a highquality transformation. The ability tohave both transformation directions ata high accuracy level is of great impor-

tance for softproofing. This is shown inFigure 18 by an example. An RGB filewith an ECI source profile is supposedto be represented in an absolute man-ner on screen. In the first preparationstage, the data has to be transformedto CIEXYZ. To transform the colourvalues corresponding RGB values mustbe found for all colour values that leadexactly to the CIEXYZ colour valuescalculated before. For that reason, thebackward transformation (CIEXYZ to

RGB) is necessary. On the other hand, tosimulate the representation of an RGBfile on different screens, the forwarddirection is required.

Matrix or LUT?The ICC profile format specificationgives references of how to “store“ acolour transformation digitally. Twodifferent methods are used:

- Transformation with tone value

reproduction curve (TRC) based on amatrix [MATTRC]- Multi-stage, Look-Up-Table [LUT]

based on transformation with multi-dimensional tables, matrices and tonevalue reproduction curves for everyinput and output channel

As a rule of thumb, the user shouldfollow the recommendations of themanufacturer of the calibration system.In case no recommendations are given,both a LUT and a matrix profile shouldbe created and evaluated visually.Generally, LUT profiles can better mimicnon-linear system behavior. A technicalvalidation is a valuable recommendationand should be performed in addition tothe final visual evaluation of the display

calibration. The validation should berepeated in regular intervals in order tobe able to recognize a variation of thesoftproofing system.

Figure 18: Photoshop setting for proofing of CMYK colour data

Figure 17: RGB characterization data generated by ProfileMaker



15/43


16/4316

Using More Than One DisplayIt is common in the graphic arts to workon more than one screen (see Figure24). The second monitor is often desi-

gnated as a palette monitor where thetoolbars are desposited, and therefore,the total area of the main monitor isavailable as a work area. But the seconddisplay might also be a Plasma TV thatshould match the primary display. Fora correct representation, both screensshould be calibrated.

Windows XP permits only one ICC pro-file per graphics card. There is, however,three possibilities for several monitors

to operate and the ICC profiles to be setcorrectly.

1. Two Graphics Cards - Connectionof the Second Monitor at Second Gra-phics CardIf the cards are the same or come fromthe same chip manufacturer, this willbe much more simple. Under Windows

Vista, the graphics card chips must befrom the same manufacturer becausegenerally Vista cannot simultaneously

work with different graphics card dri-vers.

2. Dual-head graphics card simulatingto the OS to behave like two separategraphics cardsWith this, an ICC profile is allowed forevery output connector. This seems tobe the best possibility, as it offers morescreens for use. There are relatively low-priced and simple graphics cards (e.g. byMatrox) that offer this possibility.

3. Microsoft Color Control PanelApplet for Windows XPThis is available free of charge underhttp: //www.Microsoft. com/downloads/details. Aspx?FamilyId=1E33DCA0-7721-43CA- 9174-7F8D429FBB9E&displaylang=en. The Color Control Applet isshown in the system control and allowsto assign ICC profiles to certain devices,also several monitors. Aside from this, itpermits a simple administration of theICC profiles and has a gamut viewer.

For multi-monitor use, one should readthe documentation of the program andshould proceed accordingly. The disa-dvantage of this solution is that in theOS colour settings, only one profile isshown. Though it is not easy to check

if and for which monitor a profile isactive; in this case, so-called test pro-files can be used to gather that infor-

mation. This is a simple no-cost solu-tion, but error prone and therefore onlyrecommended for experienced users.

Figure 23: Screenshot of the optimal operational settings

Operating system

Display 1 Display 2

Figure 24: Example of double screen setup

Hint!If only one screen iscalibrated, there is adanger that the different

white points changeadaptation state

of the eye.

Parameters for Softproofing - MAC or Windows: What is there to consider?


17/4317

For high-quality softproofing, the use of

a colour measurement devices is neces-sary. It has been proven that the humaneye has an excellent gift for seeingcolour difference, however, it is very badin the absolute estimation of the colour(based on colour memory). In this sec-tion, after a short survey on the subjectof colour measurement, the use of prac-tical colour measuring instruments isintroduced.

Brief Introduction to the Colour Mea-

surement of Self-Luminous ObjectsThe colour measurement of self-luminous objects, like a monitor, isdone without a white standard, whichis necessary for the reflection mea-surement as a reference, see Figure 25.Basically, the emitted light is weightedby the Colour Matching Functions (CMF)of the eye (e.g. 1931, 2 degree, normalobserver) followed by a suitable norma-lization (e.g. Y = 1 for the white point).

Colorimeter and SpectrophotometerFor colour measurement, there arebasically two different methods: thecolorimeter and the spectrophotometer.Colorimeters have built-in filters, aimingfor an optimal fit against the CMF (seeFigure 26). Therefore, they are designa-ted as filter measuring instruments. Theaccuracy of the filter adaptation to theCMF indicates quality of the instrument.A typical colorimeter is shown in Figure29.

To determine the CIEXYZ colour values,the transmitted light is separateddirectly into single spectral areas andapplied with further correction pro-cedures, such as 3 by 3 matrices. Thelatter are dependent on the screen,and according to the monitor type (andbacklight) used, different correctionsmight be necessary.

Based on the different spectral cha-racteristics of modern screens with lar-

ger colour gamuts, a different correctionis necessary for different LCD screens.The resulting CIEXYZ colour values canbe roughly interpreted as follows:

- the X component corresponds with

the red channel- the Y component corresponds with

the green channel and simultane-ously, the luminance

- the Z component corresponds withthe blue channel

Spectrophotometers function underanother principle. The transmitted lightis dispersed by means of a prism ora diffraction grating into its spectralparts. The weighing with the CMF is

often done by means of a softwareprogram in the measuring instrument.A spectrophotometer is shown in Figure28.

Figure 25: Colour measurement: Self-luminous and

object colours.

Figure 26: Schematic representation of the functionalprinciple of a colorimeter.

Self-luminouscolours

Object colours

light

Human visual

system

object

Bildquelle: Konica Minolta

Filter (3 or more channels)Display

Figure 27: Measuring instrument configuration withProfileMaker.

1.5 Measurement Devices for Softproofing

Hint!A definition and theexplanation of the diffe-rence between spectral

radiometer, and spec-tral photometer can

be found on page37.

Hint!For information onemitted light and emis-sive measurement, see

Figure 27.

Expert Tip:Compliance of the Luther Condition

Expert Tip:The spectral characteristics includethe interaction of narrow-bandemission of the background illumi-nation (CCFL or LED), the spectraltransmission characteristics of thecolour filter within the LCD panelsand the respective filter curve sensi-vities of the measuring instrument.

Expert Note:In order to maximize saturation thespectral radiances become spectrallymore and more narrow and „peaky“.Strong problems arise, when thepeaks overlap the steep parts of theCMFs. This strongly challanges thecorrect measurement of flat paneldisplays.

Problems arise if no energy is emit-ted in those „gaps“ where the CMFhave sequential overlaps. A slight

change in the emitted light mightcause a dramatically different colourappearance.

Parameters for Softproofing - Measurement Devices for Softproofing


18/4318

Please note that the distinction bet-ween hand measuring instruments and/or laboratory measuring instrumentsis not made here. Essentially, the

laboratory measuring instruments areindicated by improved optics, bettertemperature stability, and higher qua-lity of filters and/or spectral resolution(sampling).

How do spectrophotometer and spec-troradiometer differ?In this section, the difference between aspectroradiometer and a spectrophoto-meter will be explained. The differencebetween both device types is that they

measure different quantities. A spec-troradiometer measures radiometric(energetic) quantities. For example, likepower per wave length (watt/nm) on areceiver. Consequently, a costly calibra-tion is necessary.

Spectrophotometers are substaintiallymore simply constructed. They measurerelative quantities (ratios), such as thereflection factor. For this, they requireone (white) reference with a given

reflection factor. This reference canbe the paper substrate (as for relativedensity measurements) or the perfectdiffuser for colorimetric measurements.For example, the i1 Pro is a spectro-radiometer because it measures theradiance emitted by the display. Forambient illumination measurement, thisabsolute calibration is also required.The SpectroEye or SpectroDens are typi-cal spectrophotometers because theyalways require a reference. In colorime-try, the refrerence white is the built-in

white tile, and in densitometry, thepaper or the reference white (relative vs.absolute densities).

Operation in PracticeThe use of the colour measuring instru-ments is relatively easy. Most deviceshave a USB connection, so that aftera driver installation, the device is ope-rational. At the beginning of everymeasurement, according to device type,a black calibration may be necessary.

For this, the instrument is placed on anopaque area, so that no light will reachthe sensor. The calibration software nor-mally will give guidance how to do theblack calibration for a specific device.

Colorimeter Spectrophotometer

Advantages compact spectral information

low-pricedno correction of differentdisplay types necessary

good signals to noise ratiomaintenance through themanufacturer

Disadvantages no spectral information Measurement of dark coloursmight be problematic

Table 1: Basic advantages and disadvantages in the comparison of colorimeter and spectrophotometer

Figure 29: Image of typical colorimeter

Figure 28: Image of a spectrophotometer

Hint!Explanations of XYZ

colour values can befound on page 37.

Parameters for Softproofing - Measurement Devices for Softproofing


19/4319

The term “monitor“ can be summarized

as a device that transfers an electricsignal into an self luminous colourimage. Monitors exist in the middle orfirst stage of each softproof job, and areused to display colours through additivecolour mixture (primaries of red, greenand blue). Based on the decline of CRTsin daily practice, they are not reviewedin this handbook in detail. Rather thefocus of this handbook lies on LCDscreens, whereby only the relevant por-tion of the technology is overviewed.

LCD screens using liquid crystals fluids,which possess, at a certain temperature,characteristics that are only otherwisefound in crystals. Thus, they can showdouble refraction (e.g. different index ofrefraction for different levels of polari-zation). Those optical characteristics canbe controlled by electrical or magneticfields.

If a thin layer (a few micro meter)

of liquid crystal is put between twoglasses, on which transparent electrodesare mounted, the transmittance can bealtered through variation of a controlledelectric voltage. In addition, crossedpolarizers must be mounted as shownin Figure 30. Transparent thin film tran-sistors (TFT) utilize the voltage for eachpixel, and therefore regulate a screen isoften called a TFT screen.

By these filtering functionality the

permanent background illumination isdarkened more or less on a pixel by pixelbasis. If one applies colour filters nowon the surfaces of the individual pixels,then a colour monitor is establishedgoverned by the laws of additive colourmixing.

LCD vs. CRT in a nutshellModern LCD monitors show an advan-tage in respect to their colour rendering,

temporal and local homogeneities,

improved picture distortion, as well asthe maximal luminance in comparisonto the CRT monitors. Further characte-ristics, such as being flicker-free, beingimmune to electromagnetic fields, andthe slight weight, make it increasinglymore popular for the prepress work.

Figure 30: Basic construction of an LCD screen. Source: EIZO

1.6 Displays (Monitors)

Parameters for Softproofing - Displays for Softproofing


20/4320

Format and HomogeneityA basic criterion for softproofing dis-plays is the display format. While large-format inkjet printers already reach the

necessary size for proofing, the maximalformat of screens is limited. Typically,the common screen diagonal has amaximum of 30“ (76 cm). With a 22“(56 cm) flat screen, two A4 pages canbe completely displayed, see Figure 31,and for examples, see Table 2.

For large images, the uniform repro-duction over the entire screen is ofgreat importance. Ultimately, it shouldmake no difference whether a picture

is displayed in the left upper or rightlower corner. In Figure 32, a software,in which the homogeneity of a monitorcan be evaluated, is shown.

The objective assessment of the level

of uniformity is subject to ongoingresearch. Here image distortion on amicro and macro scale come into play

such as Mura. Currently the humaneye is the best judge for any kind ofhomogenity problems. The shown 3x3

assessment is a good starting pointand gives a first, general idea.

Diagonal Aspect Ratio Pixel Matrix Size [cm]

19“ / 48 cm 5:4 1280 x 1024 37.7 x 30.1

20“ / 51 cm 4:3 1600 x 1200 40.8 x 30.6

21“ / 53 cm 4:3 1600 x 1200 43.3 x 32.5

22“ / 56 cm 16:10 1920 x 1200 47.4 x 29.6

24“ / 61 cm 16:10 1920 x 1200 51.9 x 32.4

26“ / 66 cm 16:10 1920 x 1200 54.9 x 34.3

30“ / 76 cm 16:10 2560 x 1600 64.6 x 40.4

Table 2: Typical sizes and accompanying solutions of LCD screens

DIN A4 DIN A4

Figure 31: Example of a monitor with a wide display

Figure 32: Capture of a homogeneity measurement

Hint!Panel: The actual moduleof a flat screen. Mura:Japanese expression for

“mistake“ - whereall is possible



21/4321

Colour GamutAnother important characteristic of amonitor is its colour gamut, which isfundamentally made of the primaries, in

a 3 channel display (red, green and blue)and the white point. These are deter-mined in flat screens through the inter-action of the background illuminationand transmission through the filter. Forall printed colours to be simulated, theproofing system must include the colourgamut to be simulated completely or atleast show a gamut warning.

In a digital proofing system, the gamutis determined mainly by the combi-

nation of ink and paper. With screens,this depends on the related phosphors(CRT) or the interplay of colour crystals(LCD) and the background illumination.Both tube and flat screens can exceedthe print gamut in some colour areas.However, in some cases, not all of the„print colours“ can be represented,especially in saturated yellow and cyan-blue areas.

The colour gamut of offset printing on

paper is indicated in Figure 33 (orangecolour). A typical LCD monitor coversthe offset printing gamut to a largedegree.

Wide Gamut DisplaysThese displays provide an even largerthan sRGB colour gamut, which has the

ability to display allmost all colours ofthe standardized offset presses (e. g.represented by FOGRA39). Some exa-mples of wide gamut display models arelisted in Table 3.

ContrastOne aspect of the colour gamut isthe contrast ratio, which is expressedthrough the luminance ratio of whiteand black point. The contrast shouldbe at least 200:1. This serves as the

standard characterization for monitorsand is based on measurements in adarkroom.

BrightnessThe brightness of a display is deter-mined by the measurement of theluminance and indicated in cd/m2. Inorder to obtain the same colour appea-rance between the illuminated paperin the booth and white area on thescreen, the typical illumination level

should be between 400 and 500 lx, andthe luminance should be between 150and 200 cd/m2. This is clarified in moredetail in Chapter 1.7.

Figure 33: Comparison of a screen (shown in naturalcolours) and an offset printing colour gamut (shownin red)

Hint!Luminance: white = 100cd/m2 and black = 1 cd/m2 makes a flare level of 1%

white = 101 cd/m2 andblack = 2 cd/m2

makes a flarelevel about

2%.



22/4322

Label Image

EIZOCG 221

Diagonal 22“

EIZOCG 241 W

Diagonal 24“

EIZOCG 301 W

Diagonal 30“

QUATOIP Intelli Proof 213

Diagonal 21,3“

QUATOIP Intelli Proof 260

Diagonal 25,5“

Label Image

NECSpectra View 1990

Diagonal 19“


Diagonal 20“

NECSpectra View 21

Diagonal 21“


Diagonal 21“


Diagonal 26“

Table 3: Example monitors for graphics art by EIZO, NEC and QUATO (February 2008)



23/4323

StabilityA second factor that is sometimesunderestimated in a hard- or softproo-fing system is the local and temporal

stability. The reproduced colours mustbe constant over the entire sheet and/or screen within narrow tolerances.The change of the CIExy chromaticitycoordinates during warm-up is shown inFigure 34. Typically, a stable conditionis obtained after about one hour. Accor-ding to monitor type, the warm-up timemay be different. Therefore, it is recom-mended to calibrate the monitor after awarm-up of two hours.

The visual appearance of a proof/soft-proof must be the same when evaluatedto a later time (e.g. days later). Problemscan be avoided by daily checks of thecalibration. If needed, a recalibrationshould be performed. In practice, youshould review and/or repeat the monitorcalibration at least once a month.

Viewing Angle Dependency(not viewing angel ;-) )An fundamental disadvantage of LCD

monitors is still its dependence on vie-wing angle. Due to the display techno-logy, luminance and colour can changewith the viewing angle. This deviation ofthe perfect reflecting diffuser is to thisextent critical because even in a fixedviewing position, as the outer cornersof the screen are already seen under anangle of about 20°. When head move-ments or several observers that view theimage simultaneously come into playthe viewing angle dependency becomeseven more critical, see Figure 35.

Measurements by the Fogra show thatcurrent high-quality LCD screens, suitedfor softproofing, show colour deviationsof ∆E 2.5 that exist at about 20° to 30°viewing angle, see Figure 35.

It is important to note that the statedcontrast ratio values found in productbrochures are determined for a minimalcontrast of 10:1. However, this statementgives no information about the colour

variation with varying viewing angle!

One can judge the viewing angle verywell with different test images; forexample, a process grey vignette, orsubjects with skin tones.

Further characteristicsFurther important characteristics arenot focussed on in this handbookinclude additive mixture failure as wellas the reflectivity of the screen surface.

Figure 34: Variation of the CIExy colour values in the warmingup of a screen (EIZO CG221)

Figure 36: Viewing angle situations in the softproof area

Figure 35: Display of colour deviations due to viewing angle

Tip!

Shiny monitor surfaces are notsuitable for softproofing.



24/4324

Requirements for Ambient Illumina-tion

Illumination plays an important role inthe colour comparison between origi-nal and reproduction. The perceptionis determined by the spectral radianceof the light and the adaptation of theeye. The adaptation is influenced bythe background and surrondings of theobject which are in the field of view.The background is the area adjacent tothe sample. Again, colour stimulus ischaracterized through the interplay ofthe spectral distribution of the light and

the reflectance factor of the object. Theinfluence of different types of light onthe colour appearance is displayed inFigure 37.

ISO Standards Define Guidelines forvisual appraisalFor the colour comparison of print andproof, as well as for the comparisonof print and monitor basically two ISOstandards have emerged. These stan-dards serve as the guide for the daily

colour appraisal practice in the graphicarts industry. These are ISO 3664:1996“viewing conditions for the graphictechnology and photography“ and ISO12646:2008 “graphic technology – dis-plays for colour proofing – characteri-stics and viewing conditions“.

The ISO standard 3664 determines thecriteria and corresponding tolerances

for the viewing booths used for criticalcolour appraisal in the graphic artsindustry. It should be noted that ISO3664 defines two levels with respec-tively adapted demands and tolerances.On the one hand, the critical compari-son between two prints, “ISO viewingcondition P1“, and on the other hand,the practical appraisal for less criticalwork, “ISO viewing condition P2“. Thecriteria of “P1“ is considered to be a“superset“ of “P2“. That means that each

viewing booth which fulfills the narrowtolerances for “P1“, also automaticallyfufills that for “P2“. Items evaluatedinclude:

- Chromaticity- Visible range metamerism index MI

vis

- Ultraviolet range metamerism indexMI

UV

- Minimal colour rendering index- Minimal illuminance and homo-

geneity

- Neutral background and diffuse sur-face reflectance- Requirements for practice such a

hour meter indicator when the lampsshould be replaced

Figure 37: Identical test prints under three different light sources (left: commerical flourescent, middle: D50Simulator, right: incandescent)

1.7 Viewing Conditions: Evaluation of ”Standard Lighting”

Hint!For Fogra scrutiny of ISO3664 compliance, see

page 42.

Parameters for Softproofing - Viewing Conditions: Evaluation of “Standard Lighting“


25/4325

Importance of the Background (CeilingIllumination)Colour appraisal with standard lightbooths seldom take place in totally

darkened rooms. Normally, there is ageneral level of illumination. At thesame time, this concerns also the colourappraisal without use of standard lightbooths. The resulting mixture of light(booth illumination and room illumina-tion) affects the sample area as well asthe screen and therefore, influences thefinal colour rendering. The latter is par-ticularly relevant for light booth on orat press stations, which is not typicallyequipped with a light shield.

Neutral Wall and Ceiling PaintThe wall paint next to the ceiling lightsplays an important role. Only a smallportion of the light output by the lampencounters the print sample directly.A large portion is reflected by multiplesurfaces, where a colored (spectrallyselective) surface will change the lightcolour corresponding to the reflectionproperties. The floor and pieces of furni-ture can also have a negative influence.

Ergonomics?The assembly of softproof workingstations in the vicinity of outside illumi-nation (or even window) should be avo-ided. A perfect shielding of the chan-ging daylight phase (from the sunrise tothe sunset) is ideal. The monitor shouldbe alligned parallel to the window. Thisassures that no light from outside willdisturb the user (no psychic blending/dazzling). Also reflections are avoidedon the monitor screen (see Figure 38).

Figure 38: Optimal monitor placement:The parallel alignment of the monitor to the windowis good practice, but should be not near a window ifpossible. Blinds to stop outside light are not shown inthis illustration, but are needed.

Tip! Viewing distance should be aboutas large as the screen diagonal.

Parameters for Softproofing - Viewing Conditions: Evaluation of “Standard Lighting“


26/4326

In the following, basic recommenda-tions are outlined that should help you

in the use of the pertinent programs.In general, we recommend the useof “branch profiles“. For example, forworking colour spaces, profiles suchas ISOcoated_V2 as a CMYK workingcolour space and ECI RGB as a generalRGB working colour space. These colourmanagement settings are normally setin a way that profile mismatches willresult in a warning message and thenumber of colour transformations areminimized.

Special ICC “checker“ profiles help totest whether a program or operatingsystem honors the monitor profile forthe image to be rendered. Such a testprofile can either swap the colour chan-nels or it can be constructed in such away that all colours are reduced to asingle colour. On the Fogra webpage,you can find a CMYK profile, which,when applied to a picture, will outputonly red colours. A similar RGB test

profile is supplied, which will only showgreen colours. ( http://forschung.fogra.org/index.php?menuid=62 )

To test if your application honors thesource profile, just load this picture“Rot wie die Liebe“ (which is a Germansaying and means “Red like love“ or“Roses are red“). If the source profile ofthe picture is honored, the picture willbe shown in red colour. If not, it will beshown blue (see Figure 39) . Note thatthis test only helps you to determine, if

a monitor profile is used at all. It givesno information how accurate a monitorprofile is.

As noted, these settings can deviate inindividual cases. However, in practice,they have proved themselves to bewell-made. The dialogue box in olderPhotoshop versions can be adjustedaccordingly.

Figure 40: Recommended colour settings in Adobe Photoshop CS4

Adobe Photoshop CS4

Figure 39: False colour profile (so called checker profiles)

2. Softproofing Settings in Software Programs

Attention!The choice of absolute colorime-tric rendering intent (RI) assuresthat the info palette shows thecorrect CIELAB values. Be awarethat colour conversions via modechanges will use the renderingintent defined in the coloursettings too (“Image > Mode >CMYK“). Colour conversions shouldalways be done via „Edit > Con-vert to profile“ to have maximumcontrol. Also when importingdocuments in InDesign the defaultconversion options will be used.

Therefore it might be needed toswitch rendering intent to percep-tual or relative colorimetric (withor without) black point compen-sation.

Hint!This csf file is available

on the Fogra website

for downloading.

Softproofing Settings in Software Programs


27/4327

Quark XPressXPress is being tested at the presenttime with respect to its softproofingcapacities. Basic review and settings

for softproofing with Quark XPress areplanned for the next version of thishandbook.

Adobe InDesign CS3It is assumed that all CMYK files placedin the layout were prepared already forthe intended printing condition andneed no more conversion.

Older VersionsInDesign 3 (CS1) does not offer the

option to preserve numbers underthe colour management guidelines.Therefore, InDesign CS1 was to be setin a way that the source profiles areretained, see Figure 42. An unwantedCMYK-to-CMYK transformation is avo-ided when the placed CMYK contentsare already separated (converted) forthe intended printing condition and thesame profile is assigned to the docu-ment.

Figure 41: Recommended colour settings in Adobe InDesign 5 (CS 3)

Figure 42: Recommended colour settings in Adobe InDesign 3 (CS 1). The setting choosen is “Preserve embeddedprofiles“



28/4328

Adobe Acrobat ProfessionalIn Acrobat Professional, there are twoplaces where to alter the colour set-tings. For colour management, the wor-

king colour spaces are determined here,see Figure 43. In order to guaranteeconformity for PDF/X-files, one shouldtake note that the option Output Intentoverwrites the active working space.

Figure 43: Recommended colour settings in Acrobat Professional 8.0



29/4329

Under Page Display, there are settingsthat will also determine the representa-tion of files on screen, see Figure 44.

Figure 44: Recommended items to review for the display in Acrobat Professional 8.0

Softproof Settings in User Programs


30/4330

The explanations in Chapter 1.3 aboutcalibration, characterization, and

profiling can be supplemented bythis section in a more detailed, butgeneral form. Here, we explain thegeneral concepts, while on the Fograweb page manufacturers of monitorcalibration solutions describe the spe-cific settings and options accordingly(http://forschung.fogra.org/index.php?menuid=19). In this chapter, weexplain all possible parameters thatcould be altered. If a calibration pro-gram does not provide the ability to

alter a specific attribute explained inthis chapter (e.g. very few programsand monitors allow measurement andthen adjustment of the gamma settingmanually), the setting has to be ignoredby the user.

Target Settings: Luminance, WhitePoint (Colour Temperature) and Gra-dation (tonal scale)A basic goal of a calibration is a percep-tual uniform differentation of available

lightness levels. Also, the technicalfeatures of a device should be usedin a way that the target settings areachieved, and the colour rendering ofthe device is not limited in any way.The best choice of target settings isdependent on the actual use case. Thethird use case “softproof at the presscontrol station“ is not explained here,since more experiments have to beconducted.

Luminance

In the case of the comparison of a printsample to a screen, a luminance matchbetween the paper white in the boothand white displayed on the screen isthe goal. This is obtained through aninteractive adjustment of the luminancecontrol of the booth (dimmer) and theluminance control of the screen (thecontrol is often labeled “brighness“). Fortypical illumination levels about 500 lx,a luminance of about 160 cd/m2 is theprerequisite of a good lightness match.

The calibration can be done visuallyand/or by means of measurements. Itshould be noted that the monitor shouldnot be adjusted to its maximum lumi-nance because over time the maximumachievable luminance of all displays will

degrade.

In a “Photographer Workflow“, it is goodpractice to adjust the luminance to alevel of 160 cd/m2. According to theambient illuminance level, however, alower or sometimes a higher luminancecan also be appropriate. In a “Photo-grapher Workflow“ a good monitorluminance is achieved when white areason the desktop or in an image don‘tlook greyish, and also, the lightness ofthe image is not unpleasant or “blin-dingly“ high. If the ambient illumination

level changes, the monitor luminanceshould also be adjusted accordingly. Bydimming the background illuminationof flat screens (in the OSD, On ScreenDisplay menu) the luminance can beadapted by many monitors withoutseverely affecting the accuracy of themonitor ICC profile.

White Point (Colour Temperature)The white point is usually adjusted first.For a hardware calibrated display, the

white point and other parameters areset automatically. In a software calibra-tion, red, green, and blue are regulatedas the three channels by gradationcurve corrections on the graphics card.

To find the best target white point, acti-vate the softproof in your applicationand use an absolute rendering intent forthe display (in Adobe applications this isachieved by choosing the option “Simu-late Paper White“). Then, test whichwhite point setting will give you the

best match of the white in the softproofto the white of the paper stock. To findthe best setting you can try to calibrateto different target colour temperaturesor target white points (xy). Some moni-tor calibration solutions offer a tool tovisually fine-tune and recalibrate thewhite-point.

A colour temperature (CCT) in the rangebetween 5000 to 6000 K might beoptimal. Many users prefer values bet-

ween 5300 and 5800 K. The calibrationto a target white point which differsfrom 5000K is not wrong, but takesinto account that the eyes perceptionof a self-luminous monitor might beperceived somewhat different to an

illuminated unprinted substrate havingan own coloration. This is due to the

partly unknown „mechanics“ of chro-matic adaptation. A tolerance measuredof ±100 K is often not critical for thewhite point. In the “Photographer‘sWorkflow“, the white point can beselected somehow freely because theeye adapts itself to the white point.However, here, the white point shouldfit the environment. As a rule, a whitepoint between 5000 and 6500 K is per-ceived as neutral.

3. Calibration: Step-by-Step

Hint!Further information onthe gamma settings can be

found on page 37.

Hint!

A low illumination level isrecommend. This helps toachieve a stable adaptionto the monitor white point.Furthermore it minimizes

distracting influences ofa somewhat colored

illumination (ifpresent).

Calibration: Step-by-Step


31/4331

GradationNext, the gradation of the colour chan-nels is adjusted by red, green, and blue.The functional relationship between

digital input value and resulting outputvalue (e. g. luminance) is often designa-ted as a gamma function.

It is common to use a gamma settingbetween 1.8 and 2.4 (see Table 4). Here,also lies the so-called CIEL* calibration,whereby the above corresponds to thelightness function defined in the CIELABencoding space.

“Photographer‘s Workflow“ “Reproduction Workflow“

ColourSettings

Luminance(brightness)

160 cd/m21. Luminance adjustment with blank paper in cabin2. 160 cd/m2

White point (colourtemperature)

5800 K1. Colour adjustment with blank paper in booth2. 5000 K

Gradation (tone valuecurve)

1. CIEL*

2. Gamma = 1.83. Gamma = 2.24. Gamma = sRGB

1. CIEL*2. Gamma = 1.8

MonitorSettings

Type LCD and/or CRT

Calibration Hardware calibration (if not possible, software calibration)

ProfileSettings

Type (size) 16 Bit (large)

Chromatic Adaptation CAT02 (from CIECAM02)

Table 5: Target values for typical softproof applications.Please be aware that those target values should not be seen as „the Fogra recommendation“, but to be used as a starting point if one is unsure which values to choose. See theFundamentals section of this handbook and the recommendations in the text how to find the best target values for your setup.

Colour Temperature Gamma/Gradation Comment

Adobe-RGB 6500 K 2,2

Apple-RGB 6500 K 1,8

ECI-RGB 5000 K 1,8

ECI-RGB V2 5000 K L*

Photogamut 5000 K 2,2 optimized for print gamut

ProPhoto-RGB 5000 K 1,8 extremely large

sRGB 6500 K sRGB (~ 2,2) Standard for office documents and amateur devices

ISOCoated_V2 P1 (ISO3664) CMYK standard colour space for offset printing

Table 4: Colour temperature of white point and tone response of typical colour encodings.



32/4332

General calibration procedure

1. Examination of the SurroundingsTo examine this, both neutral grey wall

paint and general illumination with acolour temperature of about 5000 K arerequired. Furthermore, there must notbe any colorful items or other objects,such as post-it notes for comments, inthe visibile field of the viewer. Somecalibration programs offer the function-ality to measure the ambient light. Withthis, the colour temperature and theillumination level (measurement of theilluminance in lux) can be estimated.

2. Warm-up Phases and Cable Con-nectionsLet the monitor sit for about twohours after it is turned on. Be sure thatthe optimal signal connection exists(digitally rather than analog) and thecommunication between computer andmonitor is guaranteed, for example withDDC (by means of a suitable DVI cable)or USB cable. If required, calibrate themeasuring instrument when the soft-ware asks for it.

3. Definitions of the Target SettingsStart the calibration software, anddetermine specifications in accordancewith Table 5 and/or the ambient lightmeasurements. Start the calibration.

3.1 Hardware CalibrationWhen monitors with incorporated cor-rections are used (e.g. correction curvesfor red, green, and blue), no furtherintervention of the user is necessarybecause all correction steps needed

have been automatically performed bythe software.

Definition of White Point:The lightest neutral colour that mightserve as a reference for the colouradaption. Expressed in CIEXYZ (where Yindicates the luminance).

Definition of Black Point:The darkest neutral (CIExy colour coor-dinates of the white point) colour. Thismay require an average of several mea-

surements due to measurement noise.

3.2 Software Calibration (ManuallySupported)If the monitor does not support hard-ware calibration, all parameters shouldbe adjusted by means of the controllerand tuners at the monitor. The remai-ning inaccuracies will be compensatedin the further calibration with the helpof corrections to the graphics card.

3.3 Software CalibrationIf the screen offers no substanial controlpossibilities, a software calibration mustresult, which the user does not havecontrol of.

4. Storing the Monitor ProfileAfter the calibration and profiling iscomplete, name the monitor profile witha suitable filename. It is recommendedto include the following: monitor type,date, and target values in the form of

luminance, white point, and tone valuecurve. For example, 160_D50_Lstar or120_D55_18.

5. Visual ExaminationA visual examination is recommended.Some calibration solutions offer optionsto fine-tune the calibration. This canbe achieved by means of the monitorLUT or the graphics card LUT or by mea-suring a visually optimized whitepointwhich is used in a second calibration asthe new aim white point. To fine-tune a

calibration use a display program withICC colour management (e.g. AdobePhotoshop), and review the represen-tation based on a reference image (e.g.offset print sample with its specificsimulation profile). Then, by means ofthe available correction possibilities(gradation curves and/or white point),match the screen image with the refe-rence print sample in the viewing booth.Finally, depending on the options of thecalibration software, recalibrate and/or

replace the original profile with the newprofile.

Tip 1!For the mixture of white and black

wall paint, it should be noted thatthe resulting colour should betested for its neutral colour appea-rance. Example: Measure a paintsample with the EyeOne Pro. Targetvalue: L* = 38 - 82, C*


33/4333

Advice for visual adjustmentsIf you perform a fine visual djustment ofthe calibration, it is very important thatyou are confident that your reference

print is of high accuracy. In many casesa visual fine-adjustment is not needed.But for highest quality an experienceduser can make use of visual fine-adjust-ments. Be aware that altering the moni-tor or graphics card LUT can introducebanding in gradients which should looksmooth. If possible a recalibration to anoptimized target white point should bepreferred.

6. Proper Technical Examination (Vali-

dation)In examination of the monitor, deviceindependent values (RGB) are sent viathe display profile to be tested to at thescreen. The resulting colour is measuredand compared to the desired values(that the generated ICC profile predicts).From that, colour differences are cal-culated and shown for judgment of thecalibration‘s quality.

7. Regular Calibration

After the calibration of the softproofstation, the question remains of whento test the calibration. In the beginning,at least a weekly inspection shouldresult by means of the visual and/ortechnical test.



34/4334

In this chapter, there are examplesboth for the “Photographer‘s Work-flow“ and “Reproduction Workflow“ inAdobe Photoshop 9 (CS2). While in the“Photographer‘s Workflow“, the soft-proof shows the colours of an RGB filewhich is yet not prepared for a printingcondition, while in the “ReproductionWorkflow“ a softproof simulates a prin-ting condition. Configure Photoshopin accordance with the basic settingsdescribed in Chapter 2.

Dialogue Boxes:The “Softproof Control Center“ whereyou can customize the proof conditionfor the softproof is found under themenu „View > Proof Set-up“, see Figure65 and Figure 66.

Figure 65: The “Softproof Control Center“ in Photoshop

Figure 66: Softproof settings: Image separation („data processing“) options and settings for the rendering to the screen.

4. Hands on: Softproofing of RGB and CMYK

Images with Adobe Photoshop

Hint!Photoshop does not alwaysindicate all Display Options(on screen). This depends onthe profile characteristics of

the involved ICC pro-files and the number

of channels to besoftproofed.

Tip!Please note the separation of„Proof Conditions“ where theconversion options are set and the„Display Options (On-Screen)“ toconfigure the output to the display.

Practice Examples: Softproofing of RGB and CMYK Images with Adobe Photoshop


35/4335

Figure 67: Colour transformation: “How does my data really look?“

Colour Transformation:“Photographer‘s Workflow“ versus“Reproduction Workflow“When the colour (pixel) values of the

file should be shown bindingly — thisis a typical scenario for a photographerwho wants to know “which coloursare in the file“. Therefore the coloursof the file (image encoding) have to beconverted absolute colorimetric to themonitor. This is reached with the cur-rent monitor profile being adjusted tosimulate the device, see Figure 67.

To simulate how the data file looks ona specific output device, the data has

to be converted for this output device.Then, the converted data is displayedabsolute colorimetrically on the display.This task exists both in the “Photogra-phers Workflow“ and “Reproduction

Workflow“. To achieve this, the ICCprofile of the device is selected as thesimulation profile, whose colour result isbeing tested. Please note that this will

only work up to Photoshop CS1 for RGBencoded pictures when using the AdobeCMM (or ACE). For CS2 and greater thisfunctionality can be achieved with theApple or Windows CMM.

Another possibility is to bypass proces-sing (separation). In this case, the RGBor CMYK colour values of the image fileare sent directly to the simulation pro-file and followed by an absolute colori-metric representation. This shows how

a file will come out on a specific deviceif no prior colour conversion has beendone. Figure 69 displays this examplewith a CMYK image.

Hint!Figure 67: Colours out-side of the screen gamut

are cut off (“clip-ped“).



36/4336

Figure 68: Colour transformation: “How does my file look when processed with a perceptual intent in respect toISOcoated_V2?“

Figure 69: Colour transformation without new processing and absolute colorimetric representation.

Hint!Figure 68: Relative colorimetricwith black point compensation

is also common for imagedata conversion (“sepa-

ration“).



37/4337

The measurement of light forms thebase for an exact representation of theoriginal scenery. Herewith, the funda-mental measurements are to be referredto as “luminance“, “illuminance“, “lumi-nous intensity“, and “luminous flux“. Fordigital photography, the illuminance,as well as the luminance, are of impor-tance, which are described in the fol-lowing section. In light measuring, onedistinguishes between light measure-ment and object measurement. Light

measurements characterize the pro-perties of a self-emitting light source.Object measurements describe the pro-perties of the light that is reflected froman object.

Luminance, LThe luminance L, is correlated with thesubjectively perceived lightness andbrightness of self-emitting light sourceor an illuminated area. The unit of theluminance is Candela per m2 (cd/m2).

The luminance should not be confusedwith the lightness. To determine theluminance of a light source, for exam-ple, a floodlight, you must divide theluminous intensity (cd) of the lampsin viewing direction by the projectedarea in m^2 which is seen (light mea-surement). The luminance of an object,which is not self-emitting light, is cal-culated from illuminance E in front ofthe object and the coeffecient of reflec-tion ρ (so-called object measurement).

This simple calculation assumes that theilluminated object reflects the incominglight perfectly and diffuse. By Equation1, it is clear that with a well-knownluminance and illuminance, the coef-ficient of the reflection can be calcu-lated. With the same illuminance, the

luminance of an object with large “ρ“ ishigher than with a smaller reflection.

Illuminance, EIn contrast to the luminance, whichdescribes the light intensity thrownback by the object per area, the illumi-nance, “E“, indicates how brightly theobject is illuminated. Therefore, it con-cerns a light measurement. The unit ofthe illuminance is the Lux (lx).

whereφ is the luminous fluxA is the illuminated area

The luminous flux of a light source indi-cates how much light is issued into alldirections of the surrounding room. Theunit of the luminous flux is the Lumen(lm). The illuminance is divided in halfwith the square of the distance to the

light source (double distance: ¼ of theilluminance). Over how many orders ofmagnitude, the illuminance can vary inthe daily life, is illustrated in Figure 72.

Figure 72: Typical illumination levels.

Figure 71: Illuminance

Figure 70: Luminance

5. Fundamentals and Concept Explanation5.1 Description and Measurement of Light (Photometry)

Equation 1

L =(E · ρ)π

Equation 2

E =φ

A

Fundamentals and Concept Explanation - Description and Measurement of Light (Photometry)


38/4338

Exposure, HThe light quantity which influences overa certain period, t, of an area is calledexposure. It is, therefore, the product

out of illuminance and time or luminousflux multiplied by time divided by area.

The exposure is represented often in alogarithmic scale. The task of the photo-grapher or the automatic exposure con-trol mechanism is to select the exposuretime dependent on the light surroun-ding to the camera and the reflection of

the object to be photographed, so thatall important details can be captured.The exposure is determined by the timeand illuminance. This is adjusted overthe opening aperture (aperture stop orf-stop) in the lens. It is referred to asthe aperture number or aperture stop, K.

Equation 4

k =

wherek is the aperture numberf‘ is the focal length of the lensD is the diameter of the aperture

In the establishment of the aperturenumber, k, it is meaningful to obtain aduplication or halfing of the incominglight with every change of k. Such a

variation of the aperture opening by afactor of 2 makes a required change ofthe aperture diameter by a factor of √2.For this reason, aperture numbers areeven powers of square root of 2 (1; 1.4;2; 2.8; 4; 5.6; ...).

Colour StimulusThe physical spectrum, which is recei-ved by a human observer or a technicalimage capturing device indicates thecolour stimuli.

Equation 3

H = E · t

Fundamentals and Concept Explanation - Description and Measurement of Light (Photometry)

f‘D


39/4339

In supplement to the prior section, thespectral power distribution (SPD) of the

light source is discussed in the following,as well as chromaticity, colour tempera-ture, and colour rendering index.

The illuminant (tabulated data) orlight source is described by the relativespectral radiation distribution of a lightsource, which is comparable to thespectral reflection factor of non-lumi-nous colour. Therefore, it represents acomprehensive colorimetric descriptionof primary light sources. For measure-

ment, a spectroradiometer is necessary,which is expensive and often only foundin larger laboratories. The Figure 73shows a spectra typical for commonilluminants.

The spectral peaks of 410, 450, and 550nm exist due to the gas discharge in themercury steam, which represents thebase for light creation in fluorescenttubes. Dependent on the composi-tion of the gas mixture, there are two

essential types: HMI and HTI. The HMIlamps posess a large colour renderingindex, while the HTI lamps have a highluminance. Table 6 explains the abbre-viations.

Chromaticity and Colour TemperatureThe chromaticity of a light source or anillumated object is specified as chro-maticity coordinates x, y. Tristimulusvalues, which differ only through lumi-nance (CIE Y) from one another, havethe same chromaticity. The plot of all

chromaticity coordiantes of the spectrallights (only emitting at one wavelength)will result in a horseshoe type diagram.The chromaticity coordinates CIEx =0.3457 and CIEy = 0.35854 representfor example the chromaticity of theilluminant (CIE) D50.

Colour TemperatureThe “temperature“ of the colour of alight source is determined by how itcompares to the colour of a “black

body“. This is an idealized hollow body,for example out of platinum, thatabsorbs all light that falls on it. If a“black body“ is heated, it glows andemits light. At the same time, it passes

through a colour scale of dark red, red,orange, yellow, white to the light blue.

The corresponding curve in the chroma-ticity diagram is named Planckian locus,see Figure 74. The unit of the colourtemperature is named Kelvin (K), occasi-onally also called “degree Kelvin“.

In practice, it is very rare that the CIExyof light sources (e.g. displays) lie exactlyon the Planckian locus. In this case,one indicates the so-called correlatedcolour temperature (CCT). It is describedin the standard DIN 5033-8. Both the

colour temperature and the correla-ted colour temperatures say nothingabout the spectral power distribution,but rather serve only as a indicationwhether a colour is more reddish, neu-tral, or blueish. The light of lamps of thesame light colour (colour type of thelight type) can have a clearly differentspectral composition and can evoke,therefore, also a clearly different colourrendering.

Colour Rendering Index (CRI)The colour rendering index serves asthe numerical indication of the colourrendering characteristics of a lightsource against a reference light source.With this, a colour test procedure isused, which determines and values thecolour shift between the tristimulusvalues under the reference and the testilluminant. The colour difference is tobe calculated in a perceptual uniformcolour space. The standard DIN 6169defines, for example 14 colours and the

UVW colour system to calculate thecolour difference. Based on the colourdifference found for each test colour,the general colour rendering index, Ra,is calculated after an empirically deter-mined relation, see Equation 5.

For the determination of a specialcolour rendering index, other testcolours can be used, if necessary. Boththe test colours and the colour diffe-rence formula used to have to be stated.

Figure 74: Chromaticity cooardinates of the “black body“(dotted line) and chromaticity coordiantes of a RGBsystem (defining the gamut as shown in the triangle).

Table 6: Symbols of gas mixtures in lamps

Figure 73: Spectra of a fluorescent bulb:

Abbr. Meaning

H Mercury (Hg)

IIodide, Bromide(Halogen compound)

MBase metals (Alkalineearths: Holmium,Dysprosium)

T Daylight

5.2 Illuminant, Chromaticity, and Correlated Colour Temperature

Fundamentals and Concept Explanation - Illuminant, Chromaticity, and Colour of Light

Equation 5

Ra = 100 - 4.6 · ∆E

i


40/43


41/4341

Matrix TransformationThe matrix-based transformation with

a tone value reproduction curve (TRC)is used if the device to be characterized(e.g. a screen) shows a linear behaviorbetween device dependent controlvalues and the resulting colour values.For maxtrix profiles the CIEXYZ colourspace is used as the PCS (“ProfileConnection Space“) instead of CIELAB,because CIEXYZ scales linear withintensity. Matrix-based profiles havea small file size, which is about 4 to 8KB. Consequently, they fit well for the

definition of a working colour space.With the use of matrix profiles, it isesay to achieve that equal RGB combi-nations (grey balance) will also alwaysbe rendered neutrally. The connectionbetween the device dependent RGBcolour value and the PCS is representedwith the matrix model in Figure 76,whereby the calculation is possible inboth directions.

In Figure 76 the RGB values are to be

linearized by gradations curve (TRC) andsubsequently subject to be convertedwith a 3x3 Matrix into the XYZ coor-dinate of the PCS. For the reverse, toachieve linear RGB values the XYZ tri-stimulus values are converted with theinversed 3x3 Matrix . The device specificRGB values result after applying theinversed TRC curves to the linear RGBvalues.

LUT TransformationThe multi-stage conversion with LUTs is

described in the following:

- for every input channel, a tone valuereproduction curve

- a multi-dimensional table- for every output channel, a tone value

reproduction curve- additional matrices in specific V4

profile tags

For example the conversion from acolour encoding space with 3 primaries,

e.g. ECI-RGB, to an encoding space with4 primaries, e.g. ISOCoated, can be ima-gined as a cube with many grid points[Figure 77].

In this example the conversion fromRGB values to CIELAB is done first bymeans of a source profile. For every gridpoint in CIELAB encoding the corre-sponding CMYK aim values are stored.The RGB code values R=G=B=128 areconverted to CIELAB values [L* = 61,a* = 0, b* = 0]. As described above amatrix/TRC transformation by means ofECI-RGB profile is used in this case.

The separation with perceptual rende-ring intent of the ISOcoated profile isspecified in the BtoA0 table. In the exa-mple the following CMYK values result:[C = 38, M = 30, Y = 28, K = 11]. ForCIELAB values that do not lie on a gridpoint, the CMYK values are interpolated.The interpolation is done by the CMM.The CMM is normally integrated intothe operating system. Figure 77 illus-trates the conversion sequence when

the tone reproduction curves (TRC) ofthe input and output are not altered.The CIELAB values are used in the thre-dimensional colour look up table (CLUT).The figure illustrates how the resulting

CMYK values for a sample colour arecalculated.

Figure 76: Schematic representation of the matrix transformation in an ICC profile (Source: Fama Skript)

Greylevels

RGBR‘

G‘

B‘

X

Y

Z

Monitorprofile

Matrix -1 3x3

TRCred

TRCgreen

TRCblue

Rlin

Glin

Blin

Figure 77: Conversion with a multi-stage transformation

5.4 Matrix or LUT (Lookup Table Profile)

PCS: Profile Connection Space

TRC: Tonevalue ReproductionCurve

CMM: Colour Matching Module

Fundamentals and Concept Explanation - Matrix or LUT (Table Profile)


42/43

Kosten

EUR 1.500,– zzgl. MwSt.Für Fogra-Mitglieder[30 % Rabatt]EUR 1.050,– zzgl. MwSt.zzgl. Reisekosten

Preise gelten für maximal4 Kabinen bzw. 4 verschie-dene Beleuchtungsszenarien[z. B. verschiedenen Beleuch-

tungsniveaus, Lichtfarben,variabler UV-Anteil, …]

Übersetzung: zzgl. EUR 200,–

Costs

EUR 1.500,00 excl. VATFor Fogra members[30 % discount]EUR 1.050,00 excl. VATTravel costs to be added

Prices for a maximum of4 cabinets or 4 differentillumination set ups[e. g. illumination levels,illuminants, variable UV-content, …]

Translation: EUR 200,00 extra

Information München, Oktober 2007Überprüfung Ihrer Abmusterkabine gemäßISO 3664 – unabhängig und kompetent

Die kritische Abmusterung von Farbenist für die tägliche Produktion in dergrafischen Industrie von hoher Bedeu-tung. Fehlurteile aufgrund einer falschenBeleuchtung führen fast zwangsläufigzu Reklamationen und damit zu erhöh-ten Kosten im Produktionsprozess. DieISO-Norm 3664 definiert hierfür ver-

schiedene Kriterien hinsichtlich derGütesicherung für die farbverbindlicheAbmusterung. Beurteilen Sie Ihre Far-be korrekt und zuverlässig und lassenSie dafür Ihre Abmusterkabine von derFogra überprüfen.

Gemäß ISO 3664:2000 prüfen wir:Farbwiedergabe [allg. und speziellerFarbwiedergabeindex]Farbwiedergabe [Metamerie-index MIvis]Korrekter UV-Anteil [Metamerie-

index MIUV ]Farbgenauigkeit [D50]HomogenitätUmfeldbeschaffenheitWartungsvorrichtungen

Durchführung & Dauer der Prüfung:½ Tag vor Ort [nach Terminverein-barung]Ergebnisbericht innerhalb derfolgenden 3 Wochen

¬

¬

¬

¬¬¬¬

¬

¬

Impressum/KontaktImprint/contactFograForschungsgesellschaftDruck e.V.Andreas Kraushaar[Dipl.-Ing.]Streitfeldstraße 1981673 München, GermanyTel. +49 89. 431 82 - 335

Fax +49 89. 431 82 - [email protected]

Practical viewing

conditions for softproofing [Color Communi-cator of Just Normlicht].

ISO 3664:2000 Betrachtungsbedingungen für die graphische Technologie und die Photographie.ISO 3664:2000 Viewing conditions — Graphic technology and photographyBezug/How to get: Beuth-Verlag, Berlin, www.beuth.de

Scrutiny of your viewing cabinetaccording ISO 3664

Since deficiencies in light sources andviewing conditions, and inconsisten-cies between colour viewing facilities,can distort the colour appearance ofsubstrates, reproductions and artwork,

they are likely to cause miscommuni-cation about colour reproduction andprocessing. Ask Fogra for objectivelytesting the specifications for illumina-tion and viewing conditions that, whenproperly implemented, will reduce errorsand misunderstandings caused by suchdeficiencies and inconsistencies.

Based on ISO 3664:2000 we check yourcabinet with respect to:

Colour rendition [general andspecial colour rendering index, CRI]

Colour rendition [CIE visible rangemetamerism index – MIvis]Correct UV energy [CIE UV rangemetamerism index – MIUV ]Colour accuracy [D50]HomogenietyAmbient conditionsMaintenance

Duration of the test:½ day at the vendors premise[after appointment]

written report within 3 weeks

¬

¬

¬

¬¬¬¬

¬

¬

Praktische Abmusterungs-bedingung für den SoftProof[Color Communicator derFirma Just Normlicht].

M S

Documents

Fogra Softproof Handbook