RyeTAGA Digital Journal 2014 by TMUTAGA

Technical Association of the Graphic Arts Ryerson University Student Chapter RyeTAGA Journal 2014

TABLE OF CONTENTS Introduction...............................................................................................4 G7 Calibration Methods Comparison of G7 Calibration Methods for Inkjet and Toner-Based Digital Press on Different Substrates.........................................................................................6

Substrate Profiling and Colour Accuracy Test to Determine the Effect of Paper Profiling on Colour Management Accuracy........ 20

M-Score of the Xerox & Epson Technical Association of the Graphic Arts Ryerson University Student Chapter © 2014 No part of this publication shall be reproduced without permission and written consent from the author(s). Published by RyeTAGA http://www.ryetaga.com School of Graphic Communications Management Ryerson University 125 Bond Street Toronto, Ontario, M5B 1Y2 Canada www.ryerson.ca/gcm 2

M-Score of The Xerox DocuColor 7000 & The Epson 4880....................................... 34

Methods of Matching Pantone Spot Colours 4 Methods of Matching PANTONE® Spot Colours on Xerox DocuColor 7000................... 42

ISO 13655 in Press and Proofing Applications Understanding the New ISO 13655 Measurement Standard in Press and Proofing Applications............................................................................... 58

The Print Quality of M&Ms Inspecting a set of Custom M&Ms Under a Microscope to Determine the Print Quality and Consistency .................................................................... 76

Credits...................................................................................................... 86 3

A LETTER FROM THE FACULTY ADVISOR

THE PRESIDENTIAL ADDRESS Dear TAGA,

Dear RyeTAGA Student Chapter, Another student journal has been produced and looks amazing as always. You all put so much work and effort into it. The design is just one part of the equation. The typesetting, editing the articles, producing the printed portions and the finishing always takes the combined effort of many people. All your fundraising efforts to make the trip to Fort Worth possible were a great success and have become a staple at the School of Graphic Communications Management. As you all know I mulled the idea in my head to step down as faculty advisor, but it was just an idea. Seeing the final product and also how proud you are of the journal and what RyeTAGA stands for, makes it all worthwhile.

It is with great enthusiasm we present you with the 2014 Ryerson University Student Chapter Journal. As another year closes, we have the opportunity to reflect back upon what the RyeTAGA team has been able to accomplish with great pride. This year we bring 12 student representatives of the Graphic Communications Management (GCM) program with us to Fort Worth, Texas, USA, for The Annual TAGA Conference. We are excited to be here amongst some of the brightest minds in the graphic arts industry. This journal brought on an extremely valuable learning experience for the RyeTAGA team and student body. With the carefully thought out plan to combine conventional and digital methods, we were able to explore special inks and different substrates to create a traditional yet contemporary feel. This year RyeTAGA was also able to explore new applications in technology, first with the iPhone App and second with a digital journal.

Enjoy the conference and may the best journal win (hopefully ours)! Martin Habekost, Dr. rer. nat.

The opportunities and experiences we have uncovered this y ear will follow each of us throughout our careers, and would not have been possible with generous support. We would like to thank our executive team and general members, our sponsors, the School of Graphic Communications Management, and the funding from the Project Funding Allocation Committee for Students (P-FACS). We could not have achieve our success without the help of Martin Habekost, faculty advisor and Peter Roehrig, print technician.

RyeTAGA Faculty Advisor It has been another successful year and with this journal it will create a deeper understanding of the fantastic work the students of RyeTAGA and GCM can accomplish! Alyssa Szeto and Mark Brejnik

RyeTAGA Co-Presidents

Comparison of G7 Calibration Methods for Inkjet and Toner-based Digital Press on Different Substrates

ABSTRACT G7 calibration method can be used to achieve consistent visual appearance across different printing presses and substrates. As opposed to other calibration methods, such as ink densities or tonal value increase (TVI), the G7 methodology uses the control of gray balance as the main component to reproduce a similar visual appearance among different printing methods and substrates. To calibrate a press to G7 specifications, two custom NPDC (neutral print density curves) are created automatically in the Curve2 software: a combined CMY gray scale and a black-ink scale. This experiment tested the G7 methodology using the Curve2 software, which was applied to two different digital printing methods and on a coated and uncoated substrate. After examining the results under D50 lighting, the neutral gray tones produced in CMY and K proved to simulate a very similar visual match with no visible difference, regardless of the type of substrate or printing method.

G7 CALIBRATION METHODS Lisa Carley Emily Wong

There is a slight noticeable difference in colour; however, after several minutes of viewing under the correct illumination, the eyes experience the effect of chromatic adaptation, thus appearing to be approximately the same. Therefore, this study shows that the G7 method can be applied to inkjet and tonerbased printing process, but will only produce results that visually appear consistent under D50 lighting.

INTRODUCTION Calibration is one of the four most important principles when managing colour with devices. G7 is a calibration method used to adjust CMYK imaging devices in order to simulate the G7 grayscale definition (IDEAlliance, 2011). Calibration happens when the device is brought back to the known condition the colour management is set up for. It is important because all devices are subject to inconsistency; therefore, it is important that they are brought back into its original condition to produce consistent reproduction. G7 allows for a visual match between different printers using one-dimensional curves and also allows shared appearances between different printing devices, substrates and specifications when other colour management is not available (IDEAlliance, 2011. The neutral print density curve (NPDC), gray balance definitions and calibration methodology are the same for all imaging technology regardless of inks, substrates, and printing method (IDEAlliance, 2011). Therefore, this method is device independent. The purpose of this experiment is explore the validity of the G7 printing method by using the Curve2 software to create custom curves for two different digital printing methods, and examine its visual appearance on uncoated and coated substrates under D50 lighting. The two digital presses used in this experiment include the Xerox DocuColor 7000 paired with the EFI Fiery RIP and the Epson Stylus Pro 3880 paired with the ColorBurst 7.3 RIP. A P2P25 target was output through the designated RIP for each printing method without applying colour management. The P2P25 target was measured using the X-Rite Eye-One iSis and ColorPort software to generate

the data brought into the Curve2 software to generate the custom NPDC G7 curves. The curves were exported and applied to the P2P25 target in Photoshop. The new target was output using the RIP once again, re-measured and brought back into Curve2 to compare the calibrated target with the generated G7 curve for each digital press. A test form, including several pictures that display gray balance, was printed on both presses with and without the G7 curve applied. These test forms were compared to each other in order to determine the similarity of gray balance produced on both presses when calibrated with G7, as well as the differences between the original press settings and the G7 calibrated settings under D50 lighting. This study examines whether an image reproduced on a toner-based press and an inkjet press can be perceived as visually similar without colour management, and how the print density and gray balance is controlled and adjusted under the Curve2 software. The main objective is to determine if applying the G7 curve can simulate similar visual appearance by producing consistent gray balance control across any printing process, ink and substrate. The curves produced by the Curve2 software will adjust the digital pressesâ&#x20AC;&#x2122; performance in order to achieve a desired gray balance and tonal value. Custom curves are created for each press based upon the lightest and the darkest points (IDEAlliance, 2007). As opposed to manually drawing out curves on graph paper, Curve2 should enable consistent calibration that conforms to the G7 methodology. If the test targets appear visually similar using this software, it can eliminate the need for colour reproduction changes completed through the CTP, press or other colour management options, thus reducing significant costs.

HYPOTHESIS

SOFTWARE USED

It is predicted that when the custom NPDC curves are applied to the custom G7 test form, then the print will conform to G7 ideal curve and methodology, thus visually simulating a similar appearance across all substrates, inks and printing processes under D50 lighting.

When calibrating using gray balance, solid CMYK ink L*a*b* aim points are used and the CMY tone reproduction is adjusted to create a balanced neutral gray axis (known as the NPDC). If the solid ink L*a*b* aim points are met and the tone reproduction is balanced, all other colours can be predictably reproduced. G7 allows for harmonized colour balance and similar visual appearance in various printing technologies (RPImaging, 2013). It is important that grays are balanced because human vision is most sensitive to subtle differences in grays or colours closer to the neutral axis (Williams, 2002).

EQUIPMENT USED

Adobe Photoshop CS4 ColorPort 2.0 ColorBurst RIP 7.3 EFI Fiery EXP8000 Curve2 - Version 2.3

Epson Stylus Pro 3880 Xerox DocuColor 7000 X-Rite Eye-One iSis

MATERIALS TESTED Opus Gloss Digital Text, 80lb, 120 g/m2 Williamsburg, 70lb, 105 g/m2

PROCEDURE

G7 Ideal

Averaged

Selected

CMY Neutral Print Density Curve (NPDC)

01. Print the P2P25 target to the Xerox DocuColor 7000 through the Fiery RIP. Ensure colour management is turned off. 02. Measure the P2P25 target using an X-Rite EyeOne iSis with ColorPort. Select UV included (M0 measurement mode). 03. Save the file as a CGATS Data File, and drag the file into Curve 2.3. 04. Under ‘Create Curves’, export as a Photoshop Curve (.acv) file and save it into the Curves folder, located under the Adobe Photoshop Presets folder. 05. Open the P2P25 target in Adobe Photoshop. 06. In Adobe Photoshop, go to Image > Adjustments > Curves and load the saved G7 curve (.acv) in order to apply the curve to the P2P25 target. 07. Print the new target with G7 curve applied using the Fiery RIP, with colour management off. 08. Measure the new target using the X-Rite Eye-One iSis and ColorPort software. 09. Drag measurements into Curve 2.3 and observe the differences between the graphed curve before and after the G7 curve was applied. 10. Print the “G7 Form v66” and apply the G7 curve as indicated in step 6. 11. Save the new test form with the curve applied and print the test form through the Fiery RIP to the Xerox DocuColor 7000 with colour management turned off. Run 10 copies on Williamsburg (uncoated paper) and 10 copies on Opus Gloss Digital Text (coated paper). 12. Print the P2P25 target to the Epson Stylus Pro 3800 using the ColorBurst RIP. Ensure “Enable ICC Color Management” is turned off.

13. Repeat steps 2-12 and print each P2P25 target and test form to the Epson Stylus Pro 3800, through the ColorBurst RIP. 14. Compare the test forms under D50 lightning and take note of differences in neutral grays, and differences in CMY and K between the G7 Form v 66 on Williamsburg vs Opus Gloss Digital Text.

RESULTS

Max Density: 0.98 1.0

0.8

0.8 0.6

0.4

0.2 0.0

0.2 0

90 100

CMY Gray Balance

0.0

90 100

TVI (Dot Gain) from CIEXYZ

density 0.88 0.94 0.81 1.09

delta E 22.27 16.71 14.00 18.79

OverPrints R G B CMY

delta E 17.05 22.87 17.03 17.82

Paper

delta E 8.71

-5

-10

0 0

90 100

Overall 0

Avg E 17.18

Max E 22.87

90 100

Figure 1: Ideal vs. Measured CMY NPDC and K NPDC - Run 1 (without G7 curve applied) Williamsburg, 70lb, 105GSM G7 Ideal

CMY Neutral Print Density Curve (NPDC)

Averaged

Selected

K Neutral Print Density Curve (NPDC)

Max Density: 0.987

The output curves window with the curve applied (Figure 5) shows that very minor ink adjustments were made in order to conform to the ideal G7 curve for the Xerox DocuColor 7000. Overall, magenta and yellow were most significantly decreased from 10-40%, which suggests the quartertones were too red in appearance. However, the adjustments generated a very smooth and similar relationship between the curves for black, cyan, magenta and yellow.

Inks C M Y K

0.6

0.4

Xerox DocuColor 7000 Upon comparing the NPDC curves before and after the G7 curve was applied, it is evident that the measured neutral print densities in both CMY and K were very close to the ideal G7 curve. In order for both NDPC curves to appear similar, the overall density of cyan was reduced by 0.01, magenta reduced by 0.02, yellow remained the same, and black ink increased by 0.02.

Max Density: 1.126

Re s u l t s

Max Density: 1.115 1.0

0.8

0.8 0.6

0.4

0.2

0.2 0

90 100

CMY Gray Balance

0.0

90 100

TVI (Dot Gain) from CIEXYZ

delta E 22.42 17.44 14.07 18.02

OverPrints R G B CMY

delta E 17.39 23.45 17.13 18.18

Paper

delta E 8.65

Overall

-5 -10

density 0.87 0.92 0.81 1.11

0.6

0.4

0.0

Re s u l t s Inks C M Y K

90 100

Avg E 17.32

Max E 23.45

90 100

Figure 2: Ideal vs. Measured CMY NPDC and K NPDC - Run 2 (with G7 curve applied)

G7 Ideal

Output Curve

CMY Neutral Print Density Curve (NPDC) Entry C M Y K 0.0 0.00 0.00 0.00 0.00 9.79 14.67 14.37 9.69 10.0 20.0 20.42 22.85 23.75 21.71 30.0 32.22 35.59 36.25 32.13 40.0 43.46 44.10 47.46 43.26 50.0 52.81 54.12 57.19 55.14 60.0 62.40 65.51 68.38 64.82 70.0 73.62 74.82 76.86 75.62 80.0 83.04 83.56 84.82 84.48 90.0 91.57 90.66 91.45 93.36 100.0 100.00 100.00 100.00 100.00

100 90 80 70 60 50 40 30 20 10 0

100

% 25 50 75

Paper Included Paper Excluded CMY K CMY K 0.25 0.22 0.36 0.33 0.58 0.47 0.61 0.50 0.87 0.88 0.77 0.77

Figure 3. Adjusting curves generated by Curve2 - Run 1 (without G7 curve applied)

Max Density: 0.93 0.8

Selected

Max Density: 1.031

1.0 0.8

0.6

0.4

0.2

0.0

Averaged

K Neutral Print Density Curve (NPDC)

90 100

CMY Gray Balance

0.0

90 100

Inks C M Y K

density 0.88 1.14 0.78 1.13

OverPrints R G B CMY

delta E 17.70 29.96 10.79 11.44

Paper

delta E 4.31

Overall

-15 0

delta E 20.84 13.95 23.99 14.67

TVI (Dot Gain) from CIEXYZ

-20

Re s u l t s

90 100

20 30

70 80

Avg E 17.03

Max E 29.96

90 100

Figure 5: Ideal vs. Measured CMY NPDC and K NPDC - Run 1 (without G7 curve applied) Re s u l t s

Output Curve

G7 Ideal

Entry C M Y K 0.0 0.00 0.00 0.00 0.00 6.52 8.10 8.93 9.13 10.0 20.0 17.95 18.46 18.52 18.61 30.0 28.67 28.12 29.37 28.26 40.0 40.69 38.59 40.00 38.25 50.0 49.10 49.79 51.34 48.58 60.0 59.78 60.34 61.15 59.06 70.0 70.32 72.13 72.40 69.61 80.0 81.46 81.21 83.23 79.19 90.0 90.91 90.18 91.01 89.49 100.0 100.00 100.00 100.00 100.00

100 90 80 70 60 50 40 30 20 10 0

100

% 25 50 75

Paper Included Paper Excluded CMY K CMY K 0.36 0.33 0.25 0.22 0.47 0.58 0.61 0.50 0.77 0.77 0.87 0.88

Figure 4. Adjusting curves generated by Curve2 - Run 2 (with G7 curve applied)

CMY Neutral Print Density Curve (NPDC)

Averaged

Selected

K Neutral Print Density Curve (NPDC)

Max Density: 0.889

Max Density: 1.071 1.0

0.8

0.6

0.6 0.4

0.4

0.2 0.0

0.2 0

90 100

CMY Gray Balance

90 100

TVI (Dot Gain) from CIEXYZ

density 0.89 1.16 0.78 1.12

delta E 17.35 29.86 10.80 10.67

Paper

delta E 4.26

Overall

-15 0

90 100

20 30

70 80

delta E 20.69 13.45 24.02 14.90

OverPrints R G B CMY

-20

0.0

Inks C M Y K

Avg E 16.92

Max E 29.86

90 100

Figure 6: Ideal vs. Measured CMY NPDC and K NPDC - Run 2 (with G7 curve applied)

Epson Stylus Pro 3800 On the Epson Stylus Pro 3880, both NPDC curves are below the ideal G7 curve before and after the curve is applied. Although the NPDC adjusted and appear much closer to the ideal G7 curve after G7 is applied, it is still slightly below ideal at 70-90%. It should also be noted that the results printed after the curve applied on the Epson Stylus Pro 3800 (Figure 6) produced similar results to the Xerox DocuColor 7000 without the curve applied (Figure 1), which was an unexpected result. This suggests that inkjet may require further adjustments, such as colour management, in order to achieve better conformity to the G7 ideal curve. As seen in the output curve window, all CMYK control point values were adjusted in the custom curve to match the curve closer to the ideal G7 curve. In general, K was decreased from the 10-60% control points, and increased at 80% and 90%. Cyan, magenta, and yellow were also decreased after the curve was applied. The test form that was printed without the G7 curve appeared to be low in magenta and yellow causing the photos to look very faint and mute. In the TVI chart produced by Curve2, it is evident that the magenta and yellow inks are low in comparison to the cyan and black. The second run with the G7 curve applied, fixed this issue, which can be seen in the TVI chart produced by Curve2 of the G7 calibrated target. Visual Observations Between G7 Test Forms v 66 Under D50 Lighting Comparison on Opus Gloss before and after applied curve on the Xerox DocuColor 7000: The gray scale images of the landscape

and road produced using CMY and K appeared to have a slightly green hue before the curve was applied. The green hue was removed once the curve was applied and appeared to be more natural once the G7 curve was applied to the test form. The grayscale CMY and grayscale K image appeared darker with greater emphasis on detail and shadow tones before the curve was applied. All images on the test form also had a slight cyan hue. The oval contained inside the rectangle appeared cyan and was visually distinguishable from the surrounding gray produced by black ink. In the test form with curve applied, the oval in the centre looked very similar to the neutral gray rectangle, which indicates that CMY and K produced very similar results. After examining the CMY and K gray patches with the curve applied, there was no visually distinguishable difference. Comparison of Williamsburg after curve is applied on the Epson Style Pro 3880 and Xerox DocuColor 7000: The CMY and K gray patches produced by the Epson Style Pro 3800 visually appeared very similar, in particular in the range at 50-100%. However, the test form produced by the inkjet produced slightly darker results than the tonerbased press, which supports the fact that the inkjet used a higher density in black ink than the Xerox DocuColor 7000 required. Comparison of Xerox DocuColour 7000 on Opus Gloss and Epson Style Pro 3880 on Williamsburg after curve is applied: The CMY and K gray patches appeared nearly identical under D50 lighting. The only difference was the slight magenta hue in the background curtain in the image of the musicians produced on the Opus Gloss by the

Output Curve Entry C M Y K 0.0 0.00 0.00 0.00 0.00 10.0 10.29 14.50 10.15 10.25 20.0 21.20 27.39 27.59 21.69 30.0 30.85 40.36 42.21 33.22 40.0 41.97 49.05 51.22 42.94 50.0 52.08 58.52 60.35 53.42 60.0 66.83 68.63 67.70 62.53 70.0 78.04 76.03 74.58 72.95 80.0 85.48 83.39 83.36 83.34 90.0 92.61 93.21 93.22 92.35 100.0 100.00 100.00 100.00 100.00

100 90 80 70 60 50 40 30 20 10 0

100

% 25 50 75

Paper Included Paper Excluded CMYK K CMYK K 0.32 0.30 0.25 0.22 0.57 0.54 0.49 0.46 0.81 0.82 0.74 0.74

Figure 7: Adjusting curves generated by Curve2 - Run 1 (without G7 curve applied)

Output Curve Entry C M Y K 0.0 0.00 0.00 0.00 0.00 9.41 9.54 10.0 10.12 10.28 20.0 20.15 21.55 23.80 18.56 30.0 32.09 31.00 31.78 30.91 40.0 42.04 38.68 39.91 40.68 50.0 48.95 47.65 44.74 51.30 60.0 57.39 56.38 53.67 61.47 70.0 66.71 64.52 63.99 73.02 80.0 75.87 72.66 72.11 84.24 90.0 88.07 89.87 89.93 93.02 100.0 100.00 100.00 100.00 100.00

100 90 80 70 60 50 40 30 20 10 0

100

% 25 50 75

Paper Included Paper Excluded CMY K CMY K 0.32 0.30 0.25 0.22 0.56 0.54 0.48 0.47 0.79 0.83 0.71 0.76

Figure 8: Adjusting curves generated by Curve2 - Run 2 (with G7 curve applied)

Xerox DocuColor 7000. All tonal values in highlights and shadows appeared nearly identical, with no noticeable difference in tonal reproduction. Comparison of Williamsburg and Opus Gloss after curve is applied on the Xerox DocuColor 7000: The test form printed on Williamsburg, which is an uncoated substrate, appears to be duller or slightly fainter in appearance in the CMY and K image areas than on the Opus Gloss, which is a coated paper stock. Comparison of Epson Stylus Pro 3880 before and after curve is applied on Williamsburg: On the test form before the G7 curve was applied, the test form appeared faint and appeared more yellow and heavier in cyan overall. After the curve was applied, there was also a hint of magenta evident in the CMY gray balance image, and was evident throughout the ISO 12467-7 digital control strip.

CONCLUSIONS The Xerox DocuColour 7000 generally reproduced a closer gray neutral balance to the ideal G7 than the Epson Style Pro 3880 did after the curve was applied when viewed in the Curve2 software. This is because the measured curve and the ideal curve were very similar, before the curve was applied and even closer once the G7 cure was applied. The inkjet printing method conformed less to the G7 ideal curve after it was applied to the target because the CMY NPDC curve is slightly higher at 50%90%, but slightly lower in the K NPDC at 60-100%. This may have resulted in the noticeable differences observed throughout the G7 Test Form produced by the inkjet printing method, including the slight magenta hue

and the perceived darker details when comparing the test forms before and after the curve was applied. The only noticeable differences between the test forms were slight variations in colour, but only under close observation. This may be the result of the human eye being more susceptible to subtle changes in chroma as colours are closer to the neutral gray axis (Williams, 2002). The purpose of G7 calibration of printing processes is to achieve the most visually similar results as possible, in gray balance, across different printing processes, inks and substrates (Homann, 2009). Research states that G7 is simply a definition of neutral gray print densities and does not seek to match colour gamuts, but instead are calibrated to neutral to produce images with similar neutral appearances (Ruff, 2010). This is why there were differences in density levels after the curve was applied for both printing methods to achieve the CMY NPDC and K NPDC.

The Epson Style Pro 3880 produced a slight magenta hue throughout the digital control strip, which was an unexpected result. Despite this small deviation, when viewed under D50 lighting for several minutes, the difference was no longer noticeable. The difference can only be noticed if viewed very closely. This suggests that inkjet may require further adjustments, such as colour management, in order to achieve better agreement to the G7 ideal curve. IDEAlliance suggested that depending on the device that is calibrated, additional colour management may be needed to optimize the match to a specific reference print condition (2007). Most inkjet proofing systems require further colour management to achieve good simulation of a reference point (IDEAlliance, 2007). Although the Epson Style Pro 3880 is not a proofer, they both share the similar printing process, which is drop-on demand inkjet printing. It also may not have been conformed as precisely to the G7 method because the Epson Style Pro 3880 utilizes eight inks rather than only process cyan, magenta, yellow and black ink.

The G7 method combines cyan, magenta, and yellow at different ratios according to the substrate and inks developed from standard offset printing and standard inks and paper defined by the ISO 126472 (Rong, 2008). Since inkjet and toner have different properties of ink, and the tests were conducted by two digital printing methods on different substrates, there are many opportunities for variation. The differences between the test forms may also be a result of out-of-gamut colours that are caused by the substrate (Ruff, 2010). The whiteness of paper is often called the â&#x20AC;&#x153;fifth colourâ&#x20AC;?. G7 defines grays using L*a*b* values rather than CMYK, allowing for solids to be printed on specific whiteness of paper with certain ink (Ruff, 2010).

When the test forms with the G7 curve applied were compared side by side under D50 lighting on a neutral gray background, there was no noticeable visual difference between the neutral gray tonal areas, and the CMY and K gray patches produced. Therefore, a consistent neutral gray balance was achieved using the G7 calibration methodology on the Xerox DocuColour 7000 and Epson Style Pro 3880 on Williamsburg and Opus Gloss. This supports that the G7 method is capable of producing the same image across different printing methods and substrates that visually appear the same to the human eye under D50 lighting.Therefore,theresultsconfirmthe hypothesis, which is that there will be a visually similar appearance across all substrates, inks and printing processes under D50 lighting after the custom NPDC curves are applied to the custom G7 test form.

Weaknesses of the Test In order to minimize error, the P2P target was printed and measured with the eye-one iSis several times. This offered a better idea of the press behavior before G7 calibration and allowed the Curve2 software to create a G7 curve based upon the average of the values of these P2P targets. However, running the samples at a higher quantity may have resulted in different results, as it would have allowed enough time for the drums to warm up and better simulate the conditions of a typical print job. In addition, the test was conducted over the duration of several days. Therefore, the results may have been skewed, in particular for the inkjet printer, because another user may have adjusted the environment settings. Lastly, the most significant weakness of this test is subjectivity because two individuals conducted the observations and comparison of the test forms with and without the curve applied under D50 lighting. Although this would not have an adverse affect on the results, it would have been more beneficial to interview a larger sample of individuals and their observations. Therefore, the next step would be to survey a larger sample of people, equally of both genders and educational levels, as well as experience within the graphic arts industry to gain a better understanding and general indication as to whether the G7 calibration method is affective or not. Benefits and Industry Applications G7 offers many benefits to printers, such as faster makeready, less waste and better proof matching (Ellis, n.d.). All of these benefits allow printers to finish more jobs each shift at a lesser cost per job (Ellis, n.d.). The G7 calibration method was developed to bridge the gap between prepress

REFERENCES and printers in order to reliably and easily reproduce the visual appearance of the proof on press (Kennedy, n.d.). Not only does G7 optimize the use of a single press, but also makes it easier to match similar printing conditions across different presses. G7 is a method of calibration; therefore, it is easier to adjust the press back to the known (or targeted) conditions by using the G7 tools that are available (Ellis, n.d.). The G7 calibration method establishes a closed loop in terms of visual appearance from print buyer to agency all the way to the press (Kennedy, n.d.). G7 allows proofs to be matched on various proofing systems within an agency by controlling the gray balance from the shadows to the highlights (Kennedy, n.d.). Not only does the use of G7 in prepress benefit proofing, it also benefits the final production. If the proofs are produced using G7 and the printer also employs G7, there is a closed loop of factors that determine visual appearance; thus allowing agencies to be confident that clients who approved the proof will be equally as happy with the final product no matter what device it was printed on (Kennedy, n.d.). In addition, when all presses are calibrated to the G7 method and look as similar as possible, printing companies are able to handle any last minute decisions that clients throw at them (Upton & Herold, 2010). Not only does G7 directly benefit printers and prepress, it also benefits print buyers. G7 allows print buyers to meet challenges, such as maintaining brand identity when buying across different print media types (Kennedy, n.d.). G7 allows for similar visual appearances

across various print media, thus solving this problem. Furthermore, the fact that G7 allows for similar prints across different print media and equipment, allows print buyers to purchase globally while assuring a common visual appearance (Kennedy, n.d.). The quality assurance of G7 means that buyers can take advantage of global purchasing opportunities and be confident that a piece printed locally will maintain the same visual appearance as those printed in China or Europe (Kennedy, n.d.).

ACKNOWLEDGEMENTS Special acknowledgement goes to Dr. Abhay Sharma, who provided the access key to a license for the Curve2 software, which was a critical tool and basis used in this study to produce the G7 curve, and providing preliminary guidance throughout the research process. Acknowledgement is extended to Professor Scott Millward, for teaching the general understanding about NPDC and G7 methodology, and providing the IDEAlliance Guidelines and Specifications for 2007. Lastly, special acknowledgement goes to X-Rite for offering the ColorPort 2.3 utility software free of charge to download on their website.

Color Reproduction. (n.d.). Western Michigan University. Retrieved March 29, 2013, from www. wmich.edu/pci/color/PP3.HTM

RPImaging. (2013). G7 Certified Expert Training and Certification. Retrieved April 2 from http://www. rpimaging.com/services/G7-Certified-Expert.html

Ellis, R (n.d.). G7 Key Adoption Considerations. Retrieved April 2 from http://www.techkonusa. com/knowledge-center/g7-support/bid/28975/G7Key-Adoption-Considerations

Rong, X. (2008). G7 Method for Indigo Press Calibration ad Proofing. NIP24: International Conference on Digital Printing Technologies and Digital Fabrication, 603-605.

Field, G. G. (2004). Color and its reproduction: fundamentals for the digital imaging and printing industry (3rd ed.). Pittsburgh: GATF Press.

Ruff, M. (2010). Maximize Print Accuracy with G7. Retrieved April 2 from http://www.signindustry. com/computers/articles/2010-07-01-Maximizing_ Print_Accuracy_with_G7.php3

Homann, J. (2009). Digital color management principles and strategies for the standardized print production. Berlin: Springer. IDEAlliance. (2011). What is G7? Retrieved April 2 from http://www.idealliance.org/specifications/g7/ what-g7 IDEAlliance 2007 guidelines & specifications. (2007). Alexandria, Va.: IDEAlliance. Kennedy, D. (n.d.). Introducing G7, the New Proofto-Print Process â&#x20AC;&#x201C; Why G7. Retrieved April 2 from https://www.visionps.com/wp-content uploads/2013/01/why_G7.pdf

Upton, S. & Herold, P. (2010). CHROMiX ColorSmarts. Issue #43. Retrieved April 2 from http:// www2.chromix.com/colorsmarts/smartNote. cxsa?snid=50132&-session=SessID:18E2574209fd12C 370uqI1D96E4E Williams, A. (2002). Newspapers & Technology. Search the Newspapers & Technology Web site with Google Custom Search . Retrieved October 10, 2013, from http://www.newsandtecharchives.com/ issues/2002/02-02/ifra/02-02_greybalance.htm

SUBSTRATE PROFILING AND COLOR ACCURACY Test to Determine the Effect of Paper Profiling on Colour Management Accuracy

Victoria Bianchi

ABSTRACT

SUMMARY

The purpose of this test is to determine the effect of paper profiling on colour management accuracy. Creating ICC profiles for output devices is a must when trying to maintain accurate colour reproduction across multiple substrates. Different substrates will have varying degrees of what is referred to as “paper white,” and because ink is translucent, the “paper white” of a substrate affects the colour of an image. For this test, let “paper white,” refer to the varying degrees of blue or yellow (b*) in a substrate. This test will compare the dot percentage of ink required for output of a Lin - i1iSis IT8.7/4 target on both a white sheet of Williamsburg paper, and a sheet of Domtar Opaque with a more yellow cast. The data gathered will allow for a greater understanding of the absolute rendering intent and the importance of paper profiling on output devices.

This test has determined that even the slightest differences in the “paper white” (the varying degrees of blue or yellow) of a substrate can cause a significant difference in colour reproduction. Profiling each substrate prior to the final output of a job improves colour accuracy when printing the same image across a variety of substrates. In this test, two different substrates were used, a white sheet of Williamsburg paper, and a sheet of Domtar Opaque with a more yellowish cast. ICC profiles were created for each substrate, and a target was converted using each of the profiles in Photoshop. After converting the target, L*a*b* values and dot percentages were measured on each profile. In the end, it was determined that by using ICC profiles with an absolute rendering intent, an image can be successfully printed on two different substrates displaying consistent colour reproduction. From these results, it is recommended that colour management be taken into consideration in regards to colour critical work. This will maintain colour consistency and the accurate reproduction of spot colours across various substrates. Attention to detail is of the utmost importance in maintaining consistent colour throughout a workflow. In regards to logos, national brands must be assured that their corporate identity is consistent no matter where it is printed or what it is printed on, whether it is business cards, product, or advertising.

INTRODUCTION The objective of this test is to demonstrate the effect that paper profiling has on color management. By using two papers with a noticeably different colour cast, it will be demonstrated that a difference in paper will produce a significant shift in the colour of an image. Using E the L*a*b* values of a Lin - i1iSis IT8.7/4 target converted to the ICC profile for each the Domtar Opaque and the Williamsburg substrate will be compared. Remember both of these substrates are considered to be “white,” it is just that the sheet of Domtar Opaque is more “yellow” than the Williamsburg sheet, which is more “blue”. The eye will adjust to either one as “white” depending on how it is viewed. The results will prove that by profiling each substrate prior to final output, the colour accuracy will be improved when printing the same image across substrates. The dot percentage of yellow ink required for printing on each substrate will also be examined. This test is an accurate and valid means of determining the effect of paper on colour management because it replicates common industry characterization practices performed when profiling a printer. To create a custom profile three items are needed: a test chart, a measurement instrument, and a profiling program to create ICC profiles.

THEORY An ICC profile is a file that ensures when a specific colour is selected on the computer screen, that designated colour is consistently and correctly delivered on the substrate (Lamb, 2010). They allow for the correct colour to be rendered with the printing conditions

that are exemplified by the ICC profile, creating a link between the colours on screen and the colours an output device is capable of producing. Using an absolute rendering intent allows for an ICC profile to simulate the “paper white” of the substrate being used, it does not scale the white point of the file (Adams, Sharma, & Suffoletto, 2008). The white point is the location of the purest and lightest white in a colour space, it changes when different light sources and substrates are used (Cambridge in Colour, 2013). The relative colourimetric rendering intent however, shifts the white point to match the substrate in use, creating a new white point, which displaces the colours in relation to it. ICC profiles do not change the colour of an image; their purpose is to ensure the correct output for a given input (Lamb, 2010). To create a custom profile three items are needed: a test chart, a measurement instrument, and a profiling program. “The profile serves as a map between the device-dependant values (e.g. RGB or CMYK) and a device-independent colour space such as L*a*b*” (Adams, Sharma, & Suffoletto). The test target, which is a set of device-dependant instructions, and the corresponding measurements from a reproduction of this target on a specified device, normally in L*a*b*, are used to form the profile (Adams, Sharma, & Suffoletto). To illustrate the importance of an output profile, consider the following example in relation to figures 1 and 2. In regards to this test, there are two substrates used, one substrate is a cool white with a bluer cast, and the other has a more yellow cast. Both substrates have been profiled for a specific printer, so each profile knows what “paper white” is, and how the dry ink responds to the substrate and the environment. Now suppose an

Figure 1: Image of a banana on a cool white substrate. image of a yellow banana was printed on each substrate. If this image were to print using the same output profile on each substrate, the yellowish substrate would show more yellow, as inks are translucent, and paper white is added to the colour. Therefore, the yellow in the image of the banana would be different on each substrate. However, if the image of the banana were to be printed using the correct profiles for each substrate, the profile for the more yellowish substrate would determine that less yellow ink is required to achieve the correct colour of yellow in the banana. Through this example, the importance of profiles is apparent. Although exaggerated, even the slightest differences in paper white, among other factors such as texture, humidity, and coating, can cause significant shifts in colours (Xerox Partners, 2010). Profiling each substrate prior to final output improves the overall colour accuracy when printing the same image across a variety of substrates. To prove this theory on a monitor, the L*a*b* values from two original Lin - Eye One iSis targets converted to each profiled substrate, will be compared in a cloud to the L*a*b* values from an original Lin - i1iSis IT8.7/4 target with no profile applied. In addition, the dot percentages in and out of yellow ink required on each substrate will be charted on a line graph. Through these figures, it will be apparent in the image of the

Figure 2: Image of a banana on a substrate with a yellow cast. cloud that the L*a*b* values of the white and more yellow substrates will be fairly similar to each other, yet different in comparison to the L*a*b* values of the original target with no profile applied. Secondly, it will be seen in the graphs that the charted dot percentages of yellow ink required in printing the target on both white, and more yellow substrates will be different. Less yellow ink will be required when printing on the more yellow substrate.

EDUCATIONAL GAINS By completing this test, a greater knowledge will be gained about the effect paper profiling has on colour management accuracy. In this context of creating output profiles, colour accuracy is the ability to take a CMYK input value, and, based on the input and output profiles print a colour-managed version of the patch (Global Graphics, 2013). Before beginning the process it is identifiable that on a run of the test targets with colour management removed, the toner based inks will reproduce better on the white coloured Williamsburg test sheet, opposed to the Domtar Opaque with a more yellow colour cast. When measuring the test targets with an X-Rite i1iSis XL Spectrophotometer, and reading the L*a*b* values, it is predicted that the

colours reproduced on the Domtar Opaque test sheet will appear to be more yellow, as dry ink is translucent. Therefore, the same coloured test targets will not be reproduced on each paper. By using i1Profiler, however, to create correct profiles for both the Williamsburg and the Domtar Opaque, the profile for the Domtar opaque will determine that less yellow ink is needed to achieve a colour match on the image printed on both substrates. Creating profiles will take into account the spectral reflectance that is reflected by the paper. If the colour profiles are created correctly, the L*a*b* values of the targets converted to the ICC profiles will be similar to one another, while different from the L*a*b* values of the original target before being converted to a colour managed profile. This will indicate a near perfect match. For the purpose of this test, two papers looking visibly different were selected to exaggerate the effect of paper profiling in regards to color management. Educational gains include the knowledge that the slightest difference in paper white, among other variables of paper (not measured in this test) including texture, humidity, and coating will cause a change in the color of an image. In addition, a greater knowledge will be gained regarding L*a*b* values and dot percentage when profiling for correct color management.

DEFINITIONS Absolute colourimetric rendering intent: In colour management for proofing, an absolute rendering intent does not scale the white point of a file, it simulates the paper as well as the colours converted to the proofing space (Adams, Sharma, Suffoletto, 2008).

CIELAB colour space: The 1976 CIE colour space transformation with the dimensions L*, a*, and b*, in which equal distances in the space represent approximately equal colour differences (Field, 2004). Colour management: The act of bringing consistency to colour across all the devices used, from input, through the image-editing environment, to the final output (Daalder, 2011). Delta E ( E): Is the mathematical difference between two samples, and is computed as the distance in the CIELAB space between two colours (Adams, Sharma, Suffoletto, 2008). ICC Profile: ICC profiles capture the colour reproduction characteristics of each device using an industry-standard file format. They relate device-dependent colour to a mathematical model of human vision known as CIELAB, which is device-independent colour (Adams, Gilewicz, Habekost, Lisi & Seto, 2009). Paper white: The varying degrees of blue or yellow (b*) in a substrate. Relative colourimetric rendering intent: This will map the white point of the image reproduction 'relative' to that of the original. The white point of the original colour space will be matched to that of the output (typically a printer profile) and other colours scaled in relation. It can be used in proofing situations where accurate reproduction of the printing press 'paper white' isn't required (Cruse, 2013). RIP: An acronym for raster image processor. It is the process by which the postscript language, contained in a graphic artwork file, is converted into printed dots (Anderson, 2007). Spectrophotometer: An instrument for measuring the relative intensity of radiation throughout the spectrum that is reflected from or transmitted by a sample (Field, 2004). White point: The location of the purest and lightest white in a colour space, it changes when different light sources and substrates are used (Cambridge in Colour, 2013).

TESTING PRINCIPLES The methods used in this test are the same as those used in industry when characterizing a press. Therefore, the information gathered in this test is applicable to others who wish to profile a press specific to actual printing conditions, when output the same image across multiple substrates. Key equipment and software that was used in creating the ICC profiles included ColorPort, the X-Rite i1iSis XL Spectrophotometer, and i1Profiler. ColorPort is a software utility that allows targets to be created, saved, and measured with a spectrophotometer for use in creating ICC profiles. A spectrophotometer measures “the relative intensity of radiation throughout the spectrum that is reflected from or transmitted by a sample” (Field, 2004). “The spectrum is the most complete description of a colour, and can be used to calculate all other metrics, such as XYZ, L*a*b* and even density” (Adams, Sharma, & Suffoletto, 2008). To generate profiles, i1Profiler, a professional ICC profiling suite from X-Rite is used. i1Profiler has the capability of creating ICC profiles for workflows such as monitors, projectors, scanners, and printers. ColorPort will be used to create the IT8.7/4 target, which will be printed with colour management removed, on a white coloured substrate, and again on a substrate with a yellow cast, using the Xerox DC 7000 with the EFI Fiery RIP and the colour management settings turned off. The two substrates will be measured with the i1iSis spectrophotometer and ColorPort measurement software, and two profiles will be created that take into account the “paper white” for each substrate. The L*a*b* values and dot

percentages in and out on a Lin - i1iSis IT8.7/4 target will be measured in Photoshop, and the target will then be duplicated. The original will be converted to the “white” paper profile, and the duplicate to the “yellow” paper profile. The L*a*b* values and dot percentages in and out of the two targets will be then measured. The L*a*b* values of the three targets (the original, the “white” paper profile, and the “yellow” paper profile) will be displayed in a cloud on ColorThink Pro, and the dot percentages will be charted in a series of line graphs in Excel. Through these figures, it will be apparent in the cloud that the L*a*b* values of the white and yellow substrates will be fairly similar, where as on the line graph the dot percentage out of the yellow ink required on the white and yellow substrates will be different, as less yellow ink will be required on the yellow substrate. The test will be carried out in this method because it will provide the clearest results for the purpose of this test. By converting a target to the created ICC profiles and measuring the L*a*b* values and dot percentages with the use of the eyedropper tool in Photoshop, clear results can be obtained in an efficient manner to be charted and compared. Alternatively, the created ICC profiles could be applied in the RIP and the target could be printed a second time with colour management on, measured with the i1-iSis spectrophotometer and ColorPort measurement software again, and then the data from both sets of measured targets before and after being converted to the ICC profiles could be used to calculate L*a*b* and be compared to see if the values are a close match for the different substrates.

MATERIALS USED

Microsoft Excel

Williamsburg 12 x 18" 70 lb 105 g/m Domtar Colors Opaque 12.5 x 18.5" 70 lb 104 g/m2 2

Xerox 006R01199 DocuColor Black Ink Toner

Xerox 006R01200 DocuColor Cyan Ink Toner Xerox 006R01201 DocuColor Magenta Ink Toner Xerox 006R01202 DocuColor Yellow Ink Toner

FILES USED Lin - i1iSis.tif Lab Cloud.txt

EQUATION Eab=√( L*)2 + ( a*)2 + ( b*)2

TOOLS USED 26

ColorPort X-acto knife Cutting mat Ruler i1Profiler EFI Command Workstation Version: CWS 4.5.1.9 Fiery RIP Xerox DC 7000 Digital Press Webster, NY X-Rite i1iSis XL Spectrophotometer SN:005165 Grand Rapid, MI Photoshop ColorThink Pro

PROCEDURE Part I - ColorPort and i1Profiler: 01. Using ColorPort the IT8.7/4 target was created and saved as “IT8_specialproject_Feb11.pdf”. 02. The target was printed colour management removed with the Xerox DC 7000 using the EFI Fiery RIP on a sheet of Williamsburg paper. 03. Step 2 was repeated with Domtar Colors Opaque 04. The IT8.7/4 target was cut out from both sheets 05. The two targets on their respective sheets of paper were measured under UV Included conditions using the i1-iSis spectrophotometer and measurement software ColorPort. 06. The spectral and L*a*b* measurements were saved using the CGATS format ranging from 380nm730nm, scaled from 0-1 and no XRGA conversion. 07. The UV included spectral measurement files were used to generate an ICC printer profile for each paper type using the i1Profiler software, with default separation settings. 08. The two printer profiles were saved as “CGATS Data_White_UVinc_Mar4_isis.icc” and “CGATS Data_Yellow_UVinc_Mar4_isis.icc”. Part II - Photoshop L*A*B* Values and ColorThink Pro: 01. The ICC profiles were placed in the Library > ColorSync > Profiles folder. 02. In Photoshop, the Intent under Conversion Options, in the Edit>Color Settings tab was set to Absolute Colorimetric. 03. The Lin - i1iSis IT8.7/4 target was then opened, and the eye dropper tool was used to measure the L*A*B* values of the test patches. The L*a*b* values were then recorded in a formatted ColorThink Pro plain text document.

04. Two copies of the original Lin - i1iSis target from step 3 were made using the Image>Duplicate tab, and each image was converted to Adobe RGB (1998) using the Edit > Convert to Profile tab. 05. The two copies were saved as “Lin - Eye-One iSis_white_Apr1” and “Lin - Eye-One iSis_yellow_ Apr2,” in order to distinguish between the two profiles for the White and Yellow paper. 06. Using the same process to convert a profile as described in step 4, each target from step 5 was converted to their respective profile created in Part I, Step 8, using an Absolute Colorimetric rendering intent. 07. The eyedropper tool was then used to measure the L*a*b* values of the two test patches converted to their respective ICC profiles in step 6. The measurements from each file were recorded in two separate formatted ColorThink Pro plain text documents, one document for each profile. 08. Using ColorThink Pro a cloud of the recorded L*a*b* values was created by dragging and dropping the three plain text documents into the Graph in 3D function.

Part III - Photoshop Dot Percentage and Excel: 01. Using Photoshop, the Dot Percentage In and Out of the yellow patches from 0-100 in the original CMYK image “Lin - Eye-One iSis” were measured and recorded in Excel (See Table 1). 02. Next the dot percentage of yellow ink needed for output in the yellow patches from the image converted to the ICC profile for the white paper “Lin - Eye-One iSis_white_Apr1” were measured and recorded (Table 1). 03. Step 2 was repeated, with the image converted to the ICC profile for the yellow paper “Lin - Eye-One iSis_yellow_Apr2” (Table 1). 04. Step 1 was repeated for Cyan, Magenta and Black by measuring the Dot Percentage In and Out on the original CMYK image (Table 2). 05. Steps 2 and 3 were repeated recording the Dot Percentage Out for the amount of yellow ink needed for output in each the cyan, magenta, and black patches for both the images converted to the white and the yellow paper profiles (Table 2). 06. Use excel line graphs to display the results.

Yellow Original CMYK Dot Percentage In and Out (%) Yellow Patch

White ICC Profile-Y Dot Percentage Out (%)

Cyan Yellow ICC Profile-Y Dot Percentage Out (%)

Original CMYK Dot Percentage In and Out (%) Cyan Patch

Yellow ICC Profile-Y Dot Percentage Out (%)

Table 1: Part III Excel data table example.

White ICC Profile-Y Dot Percentage Out (%)

Table 2: Part III Excel data table example, repeat for Magenta and Black.

RESULTS AND DISCUSSION The results revealed that the L*a*b* values of the two targets converted to the profiles of each the white and yellow coloured substrates were fairly similar to each other, yet different in comparison to the L*a*b* values of the original target with no profile applied. Secondly, the results displayed that the dot percentage of yellow ink required in printing the target on both the “white”, and “yellow” substrates is different. Less yellow ink is required when printing on the yellow substrate. This indicates colour management practices were used correctly. The cyan, magenta, yellow, and black dots in Figures 3a and b represent the L*a*b* values of the original Lin - i1iSis target. The red dots represent “Lin - Eye-One iSis_white.tif”, the target converted to the profile for the white Williamsburg paper. The green dots on the other hand, represent “Lin - Eye-One iSis_yellow.tif” the target converted to the Domtar Opaque paper

Original CMYK Yellow Patch Converted to White ICC Profile

with the more yellowish cast. Upon examination of the cloud it is apparent that the L*a*b* values of the targets converted to the two profiles are similar, as the dots are overlapping. This demonstrates a close colour match of the image on two different substrates. Figures 4, 5, 6, and 7 represent the dot percentage in and out of yellow ink required in the yellow, magenta, cyan and black patches of the Lin - Eye-One iSis target. The red line in the charts represents “Lin - Eye-One iSis_ white.tif”, the target converted to the profile for the white Williamsburg paper, and the green line represents “Lin - Eye-One iSis_yellow.tif” for the Domtar Opaque paper with the more yellowish cast. These figures show that less yellow ink is required on the Domtar Opaque paper because the ICC profile used in conjunction with an absolute rendering intent does not scale the white point of the file. Using an absolute rendering intent simulates the paper, and as ink is translucent it takes into account the yellowness of the substrate.

Figures 3a and b: Cloud of the values from the Lin - i1- iSis original target, the target converted to the white paper profile (red dots), and the target converted to the yellow paper profile (green dots).

Looking specifically at figure 4, the chart for the yellow ink required in the yellow patches, and figure 7, the chart for the yellow ink required in the black patches, it is interesting to note that at the dot percentages of 80% and 95% respectively, a cross over occurs where the amount of yellow ink on the paper has become opaque as it is too thick, and therefore the profile no longer takes into account the paper white in regards to the dot percentage of ink needed. In both of these instances there is now more yellow ink being output on the Domtar Opaque sheet with the more yellow cast, than the white Williamsburg sheet. The L*a*b* values calculated from the measurement of spectral data from the IT8.7/4 target taken with the i1iSis were not a result of average readings. This could have influenced the results, as there may have been outliers in the measured data. In addition, a small fragment such as a hair or spec of dust could have been on the target as it was being measured, therefore preventing a thorough reading of the patches. As mentioned above, factors not measured in this test such as texture, humidity, and coating, could have also aided in the shift in colour between the two substrates however, attempts were made to keep these factors consistent between samples. The results of this test confirmed any preconceived assumptions, and are in agreement with published information. According to Adams, Sharma, and Suffoletto (2008), using an absolute rendering intent does not scale the white point of the paper. It allows the ICC profile to take into account the “paper white” of a substrate and renders the colour accordingly. This is reflected in the results, less yellow ink is needed on the Domtar Opaque than the Willamsberg substrate because of the yellow cast of the Domtar Opaque.

100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0

Converted to Yellow ICC Profile

5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

Figure 4: Dot percentage in and out of yellow ink on the Lin - i1 iSis IT8.7/4 target yellow patch. Original CMYK Magenta Patch Converted to White ICC Profile 100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0

Converted to Yellow ICC Profile

5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

Figure 5: Dot percentage in and out of yellow ink on the Lin - Eye-One iSis magenta patch.

Original CMYK Cyan Patch Converted to White ICC Profile 105 100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0

Converted to Yellow ICC Profile

RECOMMENDATIONS 0

5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

Figure 6: Dot percentage in and out of yellow ink on the Lin - Eye-One iSis cyan patch. Original CMYK Black Patch Converted to White ICC Profile 105 100 95 90 85 80 75 70 65 60 55 50 45 40 35 30 25 20 15 10 5 0

Converted to Yellow ICC Profile

5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

Figure 7: Dot percentage in and out of yellow ink on the Lin - Eye-One iSis black patch.

It can be concluded that colour management is necessary when printing an image on two different substrates for a correct colour match. Colour management and printer profiling allow consistency to flow throughout a printed piece when printing the same image across multiple substrates. Creating an ICC profile for the output device takes into account the paper white of the substrate being used, which in turn determines the amount of ink necessary to produce the correct colour intended for the image.

Printability Printability applications for this test concern whether a match exists between the colours of an image printed across multiple substrates. This is important in regards to branding, particularly in maintaining brand colour consistency of logos and accurate reproduction of spot colours in printed media. It is recommended that an ICC profile be created for an output device each time the substrate changes, or other factors such as, texture, humidity, and coating change, as these can cause a significant shift in colour. An absolute rendering intent should be used to simulate the paper, as well as its resulting colours when converting an image to the proofing space. An absolute rendering intent enables the ICC profile to account for the “paper white” of the substrate when setting the white point for colour conversion. For example, imagine two identical products side by side on a shelf. Both products were manufactured on the same date, but were packaged in boxes made

of different substrates because the company was in the process of changing over their package design. The appearance of the colours on box “A” are less vibrant than on box “B” because different substrates were used. Which box would the consumer be more likely to purchase, box “A” with the dull colours, or box “B” with the vibrant colours? Box “A” would give the perception of having an old product because the colour of the packaging appears to be faded, therefore the customer would gravitate towards box “B”. Proper colour management and paper profiling could have minimized the colour differences between boxes "A" and "B". Using an absolute rendering intent to create ICC profiles for each substrate would allow the press to adjust the dot percentage of ink required, in order to compensate for differences in the colour of each substrate. This will produce a closer colour match of the image between the two substrates. Runnability It is recommended when measuring the target on the i1iSis to keep the paper straight as it is feeding into the device to ensure a proper reading of the ink on paper. This way, a more consistent scan is done of the targets, resulting in the creation of a proper profile. Measure the target as soon as it is printing to avoid damage to the substrate. Also if possible, it is recommended to take a few readings and average the results to help eliminate outliers of the data. End Use End-use applications for this test concern that the printed colour matches as close as technically possible to the colours intended by the client in the original file in order to protect brand identity.

“Colour management begins with the standards set by the International Color Consortium (ICC) and spans across the entire design and print workflow” (Riot Creative Imaging). ICC profiles are used in file creation through to final printing, in order to control colour accurately to please the end user. This also includes “standardizing colour workspaces, calibrating monitors and printers, standardizing RIP settings and print operator workflows (Riot Creative Imaging)”. Attention to detail is of the utmost importance in maintaining consistent colour throughout a workflow. National brands must be assured that their corporate identity is consistent in look no matter where it is printed or what it is printed on, whether it is business cards, product, or advertising. In an article published in October 2012, by the New York Times regarding the top global brands, Coca-Cola ranked number 1 on the list for the 13th consecutive year with a brand value of an estimated $77.8 billion (Elliot, 2012). Worldwide, the printer must conform to colour management standards to protect Coke’s brand and make sure that exact red is output, as big money is at stake. Coke uses the offset lithography standard G7, however in regards to digital printing which is growing in prevalence as technology evolves, there isn’t a technician who mixes inks to an exact specification (Riot Creative Imaging). Digital printing depends on the RIP, the printer, and colour management to produce and verify colour output to ensure that colour remains the same over time. Through the proper use of ICC profiles and colourimetric rendering intents in a workflow, the end-user can be assured the integrity

of their brand identity is not at stake. Using an absolute colourimetric rendering intent will give you consistent values when measuring. However, no one measures his or her Coke container. A person usually uses their eyes, which is similar to the relative colourimetric workflow. With digital printing becoming more prevalent will absolute colour matches on digitally printed material be made irrelevant? As long as a digitally printed logo looks relatively similar is this acceptable?

ACKNOWLEDGEMENTS

REFERENCES

The author would like to thank Scott Millward and Dr. Abhay Sharma for their suggestions and guidance, as well as the Heidelberg School of Graphic Communications Management for providing the equipment and software necessary for this project.

Adams, R., Gilewicz, N., Habekost, M., Lisi, J., & Seto, A. (2009). Digital photography for print. Sewickley: Printing Industries Press. Adams, R., Sharma, A., & Suffoletto, J.J. (2008). Color management handbook: A practical guide. Pittsburgh: PIA/GATF Press. Anderson, R. (2007). Exploring digital prepress: The art & technology of preparing electronic files for printing. Canada: Thompson Delmar Learning. Cambridge in Colour. (2013). Color Space Conversion. Retrieved October 12, 2013 from http://www. cambridgeincolour.com/tutorials/color-spaceconversion.htm Cruse, P. (2013) Colour management â&#x20AC;&#x201C; how it works?. Retrieved October 12, 2013, from http://www. colourphil.co.uk/rendering_intents.shtml Daalder, J. (2011). Color management theory. The digital fine print (Chapter 4). Retrieved February 13, 2013, from Image Science: http://www.imagescience. com.au/pages/Color-Management-Theory. html#ColorManagement Field, G.G. (2004). Color and its reproduction: Fundamentals for the digital imaging and printing industry. Pittsburgh: GATF Press.

Global Graphics. (2013). Color accuracy. Retrieved October 12, 2013 from http://www.globalgraphics. com/technology/color-management/coloraccuracy/ Lamb, J. (2010, August). Sublimation color management. Retrieved March 30, 2013 from Printwear: http://printwearmag.com/article/ sublimation-color-management Riot Creative Imaging. (2013). Spotlight on color management. Retrieved March 30, 2013 from Riot Creative Imaging: http://www.riotcolor. com/article/spotlight-color-management#.UWf_ cqVGshx Stuart, E. (2012, October 1). List of global brands keeps coke on top, and apple jumps up. The New York Times. Retrieved March 30, 2013 from: http://www. nytimes.com/2012/10/02/business/media/bestglobal-brands-report-has-coca-cola-on-top-andapple-climbing.html?_r=1& Xerox Partners. (2010, September 10). Spot Colors and Fiery: Everything you need to know (Part 1). Retrieved March 30, 2013 from http:// digitalprinting.blogs.xerox.com/2010/09/spotcolors-and-fiery-everything-you-need-to-know/#. UWQbEqVGshx

M-Score of the Xerox DocuColor 7000 & The Epson 4880

ABSTRACT This paper examines the M-Score of the Xerox DocuColor 7000 and the Epson4880. M-Score is numerical representation of the uniformity of a press sheet. The Xerox DocuColor 7000 is a xerographic press. Xeropgrahic presses utilize a photo conductor drum and static electricity to transfer toner onto the substrate. The Toner is then fused to the substrate with heat. The Epson 4880 is an ink jet proofer with a Micro Piezo ink head. The Piezo ink head utilizes a metal form that physically deforms in order to create a droplet of ink, which is then ejected out onto the substrate to create a dot. For this test, three different press sheets are examined for each printer. The three press sheets will respectfully be at low, medium, and high densities.

M-SCORE OF THE XEROX & EPSON

Julia Mei

As mentioned in this report, two factors are predicted to have an effect on the M- Score of a printer; the density of the test form and the device being tested on.The low,

medium. and high density test forms show a decrease in M-Score as the density increases. This is relative to the L*a*b* values that were measured because they are calculated using the CIE standard observer. The standard observer is based on the human eye which is more sensitive to the mid-tone areas than the highlights and shadow areas. The medium and high density test forms are closer to the mid tones, and thus show a lower M-Score than the low density test forms. In addition, the ink jet shows a greater uniformity in its press sheets than the xerographic press. The press sheets for the Epson 4880 had a significantly higher M- score in comparison to the Xerox Docu Color 7000. Because the lowest M-Score of the Xerox press falls below the tolerance (An M-Score of 60), it does not pass the M-Score test. The Ink jet proofer on the other hand, does pass the test.

INTRODUCTION In recent, years, Xerographic printers and inkjet printers have brought upon many improvements to the digital printing processes. Xerography is a dry digital photocopying technique. Images are created with the use of a photo-conductor drum that utilizes static electricity to print the images. First the Photo-conductor drum is charged with a negative electrostatic charge. Afterwards a lamp will light up with the image. Areas where there are white will be lit up and will discharge the electrostatic charges. The areas where there are black will absorb this light and remain negatively charged. These negatively charged areas attract the toner and will then be transferred and fused onto the press sheet. This is made possible because each toner particle is coated with wax and when heated the wax fuses the toner particle to the paper, trapping these particles. Image transfer to plain paper is usually 80-90% efficient (Williams, 1984). Similarly, the invention of the ink jet printer has also made great improvements to the society as well as the printing industry. Although ink jet printing can also be considered as digital printing it is however, a very different process compared to Xerographic printing. Ink jet printing is done by propelling tiny droplets of ink onto the substrate. It is also currently the most commonly used type of printing(Piazza). There are two types of ink jets; drop-on-demand and continuous ink jet. A Drop on demand system is based on dropping the ink only when necessary while continuous ink jet continuously ejects a stream of ink that is deviated to create a pattern (lmpika). The Epson 4880 is a drop on demand ink jet printer that utilizes a Piezo Ink head.

The Micro Piezo ink head is developed by the Japanese imaging company known as Epson who's ink jet printer will be examined in this experiment. As shown on the above illustrations. The Piezo ink head utilizes a type of material that reacts physically to bending which causes a droplet of ink to be formed and ejected. The deformation of the metal as well as the size of the ink droplet can be controlled by the amount of electrical charge being applied to it. Often times digital presses are used for shorter press runs. This is because digital presses have relatively low costs regarding make ready as well as low make ready times as well compared to offset presses. Digital presses are able to print very high quality jobs, and are also becoming more accepted in the industry. Offset presses however still have a lower cost per sheet in comparison to digital presses. This is why they are still used for high quantity jobs. In regards to this test, the uniformity within a press sheet test addresses the issue of how evenly colors are printed across a press sheet. The three gray balance test forms that will be used for this experiment represent low, medium and high density levels. These press sheets are GRACoL gray-balanced tints at 20%, 40%, and 65%. (Sharma, 2012) These press sheets printed by the Xerox DocuColour 7000 and Epson 4880 will be measured by the X-Rite iSisXL. The Press sheet will be divided into 46 columns and 64 rows, to be measured and tabulated. These patches are 6mm in diameter which means that a single press sheet had 2944 patches to be scanned by the spectrophotometer. These measurements for each

row and column are compared with the neighboring values and compute the Delta E values for each difference. These Delta E values will finally be used to calculate the M-Score for the single printed press sheet. M-Score is also known as the homogeneity test. It tests the streakiness, graininess, mottle, and inking variation across a press sheet. (Fujifilm) It is a numerical value that represents the uniformity of the press sheet and can be a value between 0-100. 100 being the high possible score, and 0 being the lowest possible score. “This is a metric from the proposed ISO 15311 standard that measures the uniformity across the printed sheets.” (Sandstad, 2012) A Score of above 95 is considered perfect and shows no sign of inhomogeneities (Fogra). However, a score of below 50 is considered very poor. In order to pass the Uniformity within a press sheet test, the lowest M-Score of all three sheets must be over 60. (Sharma, 2011). M-Score ≥ 95 ≥ 85 ≥ 70

Meaning Perfect Very Good Good

≥ 60 ≥ 50 < 50

Satisfactory Adequate Poor Table 1: Varying values of M-Score

The purpose of this experiment is to evaluate the quality of the Xerox DocuColour 7000 and the Epson 4880 printer as well as the possible reasons behind

the uniformity or lack thereof, while printing on a press sheet. It is predicted that darker tints would show less uniformity than lighter tints. The calculation of M-score is based on L*a*b* values from a spectrophotometer. L*a*b* values are calculated using the CIE standard observer. The standard observer is calculated based on the human visual system, and how the human eye views different colours. The Human eye is able to differentiate smaller differences in the mid tones in comparison to highlights and shadows. As for the high lights and low lights, the human eye has trouble detecting minute differences in colour. This explains why the MScore is lower for the medium and high density forms. These two test forms’ densities are closer to the 50% mid-tone, and thus the human eye can detect smaller differences in colour which results in a lower M-Score. As for the low density test form, the human eye cannot detect the small differences in colour, and abnormalities in colour uniformity. This is because the lighter tint makes it difficult for the human eye to detect abnormalities in the uniformity of the press sheet. Thus, it predicted that the low density test form will have the highest M-Score out of the three for both devices. It is also predicted that the M-Score of the test forms may be affected by the type of device it is printed on. Offset printing usually gives good uniformity across the press sheet and is often affected by the changing of ink keys on the press itself. Thus, it is irrelevant to conduct this test

PROCEDURE in relation to an offset processes. Digital presses however are suitable for the test. Xerographic processes and ink jet processes are significantly different, and thus it can be predicted that there may be a significant difference in M-Score results as well in relation to the type of process the test forms are printed on. For example, the quality of xerographic press sheets depend greatly on quality and conductivity of the conductor drum, while the quality of the ink jet press sheets depend on the quality of the ink heads. One problem ink jet printing may face is the fact that it prints across the press sheet in rows. This may leave bars on the press sheet if the ink heads are of lower quality.

01. Print out the Digital Press test forms on both the Xerox and Epson printers. Print the Digital Press test forms out on the Xerox DocuColor 7000 using the Fiery RIP and print out the Digital Press test forms out on the Epson 4880 using the ColorBurst RIP.All test forms are printed out on 12x18” paper. The Epson 4880 prints on 17” paper, and thus needs to be carefully trimmed to the correct size after being printed. There should be a set of low, medium and high density test form sheets for each printer which is a total of 6 press sheets that need to be measured.

03. Save each measurement as a separate CVS File. 04. Import the CVS Files into separate Excel files and begin creating the spread sheet to calculate the M-Score. 05. For each CVS/Excel file, average the L*a*b* values along the columns and rows. There should be 46 Columns and 64 Rows. 06. Calculate the Delta E L*a*b* colour differences of each Column with its neighboring column and each row with its neighboring row. There should be 45 Delta E values for the columns, and 63 Delta E values for the rows. 07. Sum all the Delta E values for the columns and multiply by 10 and Sum all the Delta E values for the rows and multiply by 10 using the below equation.

SOFTWARE USED Microsoft Excel ColorBurst RIP 7.3 EFI Fiery EXP8000

EQUIPMENT USED Epson Stylus Pro 4880 Xerox DocuColor 7000 X-Rite iSisXL

MATERIALS TESTED 12 x 18” Press Sheet Digital Press Forms (Low, Medium and High Density)

RESULTS By visual examination, the uniformity of the press sheets printed by the Epson 4880 are significantly higher in comparison to the press sheets printed by the Xerox Press. The press sheets printed by the Epson do not show any visible abnormalities or any bars from the ink heads printing across the press sheets. For the High density test form printed on the Epson 4880, upon visual examination, the press sheet has very uniform color. However the press sheet may have been slightly damaged due to handling of the press sheets prior to measuring with the iSis. And thus, the M-Score result of this press sheet is slightly lower than expected.

(Sharma, 2011)

08. Using the below equation, calculate the Delta E Long & Short. (Sharma, 2012) 02. Measure the L*a*b* values of each test form using the X-Rite iSisXL. There should be 2944 Patches measured. (46 x 64 patches) When measuring the test forms using the iSis, use ColorPort to measure the targets. Select the correct target in correspondence to each test form being measured.

(Sharma, 2011)

09. Using the following equation, calculate the M-Score for the test form. In order to pass the test, the lowest M-Score of all three sheets must be above 60.

However, for the test forms printed on the Xerox DocuColour changes in the color on the press sheets can be easily seen. There are visible differences in colour as well as gradients of the tint. This is especially visible in the high density press sheet for the Xerox. The M-Score for the Epson 4880 shows significantly higher uniformity across the press sheets in all three densities. According to the Fogra chart that defines

REFERENCES the meaning of M-Score results, the Epson 4880 shows very good quality with no visible stripes which is true when examining the press sheets. As a result the Xerox DocuColor does not pass the M-Score test due to the fact that the lowest M-Score out of the 3 press sheets is 54.5 which is below 60. The Epson 4880 however, does pass the M-Score test because the lowest M-Score out of the 3 is 77.6, which is above 60.

the M-Score is calculated, which requires the use of the L*a*b* measurements of 2944 patches on each press sheet. L*a*b* are values based on the CIE standard observer. The standard observer is able to identify smaller differences in colour near the midtone ranges. This is the reason why the medium and high density test forms show the lowest M-Scores in comparison to the low density test form, reason being that the medium and high density test forms are respectfully 40% and 65% GRACoL tints, which are relatively closer to the 50% mid-tone areas than the 20% low density press sheet. The test forms for the Epson 4880 were expected to show some visible bars or striping caused by the ink head printing across the press sheet. However, the qualities of the printed test forms exceed expectations, and did not show any visible stripes across the press sheets. This resulted in very uniform and high quality press sheets with higher than expected M-Scores.

Figure 1: Comparison of recorded M-Score Samples

CONCLUSION Both Hypothesis made prior to the experiment are accepted. The M-Score declines as the percentage tint increases. In addition, medium and high density test forms show lower M-Scores than the low density test form. This can be explained by the method in which

Fogra. (n.d.). Evaluation of the in-homogeneity of toner based printing systems .Retrieved from http://www.fogra.org/en/fogra-research/wcdigital-printing/digitalprinting-current-projects/ imagequality-35003/research- topics/homogeneity/ evaluation-of-the-in-homogeneity-of-toner-basedprinting- systems.html

Piazza, P. (n.d.). Types of printers and scanners. Retrieved from http://printscan.about.com/od/ printerscannertypes/a/inkjets.htm

Fujifilm. (n.d.). Development of a standard for digital print (iso 15311).

Sharma, A. (2011). Idealliance digital press certification program. (Version 2.2 ed., pp.14-15).

Ientilucci, E. (1994, February 22, 1994 22). Fundamentals of xerography. Retrieved from http://www.cis.rit.edu/~ejipci/Reports/Xerography. pdf

Sharma, A. (2012). Digital presses in 2012: Capability review. TechConference IDEAlliance Educational Seminars,

Impika. (n.d.). Technology. Retrieved from http://www. impika.com/index.php?id=lejetdencredod

Sandstad, B. B. (2012, March 19). Digital press certifications. Retrieved from http://optirep.net/ digital-press-certifications/

Williams, Edgar M.(1984), The Physics and Technology of Xerographic Processes, New York: John Wiley and Sons

The test forms printed by the ink jet printer shows a significantly higher uniformity than the test forms printed from the Xerox. After calculation, the M-Score value for the ink jet printer was significantly higher as well. And thus, the second hypothesis is accepted. The device on which the test forms are printed on does affect the uniformity of the press sheet. In Conclusion, both hypothesis are acceptable as Epson 4880 passes the M-Score test, while the Xerox DocuColor 7000 does not.

4 Methods of Matching PANTONE® Spot Colours on Xerox DocuColor 7000

ABSTRACT

METHODS OF MATCHING PANTONE SPOT COLOURS

Florence Kong Elaine Leung

Many graphic designers have run into colour reproduction problems when printing off PANTONE® colours from a digital printer. A digital printer usually has a CMYK workflow; all spot colours, such as PANTONE®, must be converted to CMYK equivalents. Inaccuracies will definitely be present due to the different properties of the output device. The difference in digital production methods of PANTONE® colours will be explored. This laboratory experiment is designed to test PANTONE® colour reproduction according to accuracy defined by the lowest E using Fiery Command Workstation, Fiery RIP, and

the DocuColor 7000. Seven uncoated PANTONE® colours, 282, 5125, 456, 3125, 699, 7506, and Hexachrome Green were tested with four methods of reproducing PANTONE® colours, Spot Off, Spot On, Fine Tuning, and with a Spectrophotometer, on a digital press. In gamut PANTONE® colours had good matches when using any method aside from Spot Off. Out of gamut PANTONE® colours generally have larger E values from the high visual differences. No single method can effectively overcome this issue and it is found to be best dealt by Fine Tuning the colour manually.

INTRODUCTION In an industry where the efficiency of an existing workflow is precariously balanced with complexity of the job, printers must find ways to constantly cut costs and time from their work. While offset lithography and flexography have their own markets and end uses, digital printing continues to push printers towards the future due to their versatility in on-demand printing, less makeready/ wash up times, and variable data printing. The biggest difference from traditional printing methods and digital printing methods is the need for a manually created printing plate. While the technology in a digital press may rely on a more expensive heat processed toner or UV curing process (and similarly, drop-on-demand in inkjet printers), the overall reduction of makeready and wash up significantly reduce costs. This enables the digital press to become more than just a faster alternative to offset. The flexibility in applications that a digital press is capable of can overtake a traditional offset press in uses. Some applications include variable data printing, which has effectively changed the way marketers can target specific demographics with printed media, and printon-demand, which allows for short runs to be printed and bound at a low cost. Use of digital presses largely depend on the shifting market of what print media is being bought and what is effective for consumers in this day and age. Ongoing trends in printing include less turnaround time due to reduced run lengths and increased complexity in design. These two trends are particularly important to note in the digital press industry because it encompasses what digital presses are known for: being able to produce cost effective, short runs, for lower inventory holding and small jobs, and variable data printing. The rigidity

of offset printing, as well as the large maintenance costs associated with such a large press, can be daunting for new companies, and as such more and more small to mid-sized companies opt for using digital presses as their main means of providing print solutions. As digital presses move into the spotlight as the goto solution for many companies, there will be a higher importance on colour matching, and in particular, spot colour matching. Marketing campaigns, branding and variable data printing all go hand in hand, and without confident accuracy in spot colour matching, companies will be unable to fully make the transition from using offset to digital for marketing services. PANTONE® swatch colours, as made prevalent in offset lithography and flexography jobs, play an important part of branding in a company’s appearance to all end users. PANTONE® colours are used to make printing efficient and effective-they help to boost the range past a regular CMYK gamut, as well as ensure consistency from a page to page document, especially in the smooth coverage of large areas (“PANTONE® Color”, n.d.). However, these PANTONE® inks cannot be mixed in the traditional sense for digital presses, since no actual ink is used. Instead, digital presses can only, through their RIP, make the closest match it can in CMYK, and unfortunately there are drawbacks to this solution. In this experiment, the Xerox DocuColor 7000 Digital Press and EFI Fiery RIP were used in tandem to simulate the situation of printing spot colours on a digital press. While this experiment is limited to one press and one RIP workflow, the data accrued can be applied to other presses and can correctly colour manage the use of PANTONE® colours when they must be simulated through 4-colour heat processed toners.

The EFI EXP8000 Color Server, more simply known as the Fiery RIP used with the DocuColor, was announced for use in September 2004. Despite the 9 year gap in technology, the RIP is sufficient for high- end production printing, as well as a way for spot colour management. Simulating spot colours on the Fiery RIP is possible and the colour is matched to what the RIP assumes to be the correct colour, but some colours are still difficult, or even impossible, to replicated on a digital press. This is due to the limited CMYK toner that must be used in the DocuColor and the range of colours that the DocuColor can print, i.e. the colour gamut. Some colour challenges encountered by the Fiery RIP include colour accuracy, consistency across print engines/documents, usability, and meeting the needs of operators of different skill levels (“Fiery Servers”, n.d.). However, this does not imply that the RIP is not without its own methods of handling spot colour channels, and that is what will be explored in the experiment.

SCOPE OF THE EXPERIMENT This experiment aims to replicate a real-life printing situation of printing PANTONE® spot colours using various methods provided by the Fiery RIP. The procedures assume that a Xerox Docucolor 7000 and EFI EXP8000 Fiery Color Server are used together, and if this experiment is used with different equipment, varying data may occur. While digital presses are becoming a popular solution for short-run, specialty printing, the limitations of using 4-colour toner must be addressed. This experiment’s procedures and results will provide the basis of explaining the discrepancy between offset press PANTONE® matching and digital PANTONE® matching, as well as proving and addressing the fundamental reasons why PANTONE®

spot colour reproduction on a digital press is inaccurate. The Fiery RIP’s understanding of spot colour is ultimately dependant on offset CMYK equivalent values of spot colours. Applying the device counts from an offset press, which uses CMYK ink to reproduce spot colours, has extreme differences from using CMYK toner in a digital press to reproduce spot colours. However, the system built into the Fiery RIP still utilizes offset values. This experiment, in addition to the aims stated above, will also aim to match colours on the digital press based on offset CMYK equivalent values and see if a E of </= 1 can be achieved. An acceptable E of 1 or less ensures that the colour difference of a spot colour to its offset counterpart is nearly unnoticeable, and would be a great achievement. Colour accuracy and comparisons will be defined in this way, through measured values with an X-Rite550 Spectrophotometer. The experiment will approach the printing of spot colours through 4 distinct methods. The first method is achieved through purposely turning off the default Spot On feature through the ColorWise Pro window in the Fiery RIP. By turning the feature off, all colour management of spot colours go straight to their CMYK equivalents according to the spot colour manufacturer. In our experiment, this would be the CMYK counts as defined by PANTONE®, available from any generic device or program that uses the PANTONE® libraries, such as Adobe Photoshop. This is different from the Spot On tool where each PANTONE® colour can be edited and fine-tuned to user specifications--turning Spot On off completely disables any spot colour usage, useful for situations where the printer does not require colour accuracy.

SETUP AND PROCEDURE The second method built into the Fiery RIP, the Spot On feature, is based on International Color Consortium standards, and is the default method of printing spot colours. All Fiery servers are PANTONE® calibrated, and is a unique feature that provides consistency across Fiery-driven devices. By enabling Spot On in the Fiery RIP, spot colour channels are converted to the CMYK equivalents according to the spot colour library found within the RIP itself, and not the external database from spot colour providers. Users can manually adjust the reproduced spot colour. This is most useful if the stock required for jobs are tinted and not plain white. The software is ideally streamlined for the user and all aspects of colour matching are integrated into the software, without need to look up CMYK values for matching PANTONE® colours. In addition, the Fiery RIP saves the values from the conversion across the server, and streamlines for consistency, repeatability and efficiency. The third method, related to method two with Spot On, is a more precise approach to printing spot colours on a digital press. The ring-around “Fine Tuning method”, as we will refer to it in this experiment, is a way of visually inspecting a printed spot colour and choosing one that is close to what the user believes is a close match. This method is done by printing a ring-around made by the Fiery RIP itself, and manually matching what is deemed close by the user. As this is a very subjective and physical method, there are specific pros and cons associated with doing this. The fourth method that was investigated was by using the EFI ES-1000 Spectrophotometer to scan and input the L*a*b* values of whatever physical sample was available and creating a custom colour library for our chosen PANTONE®

colours. This method is theoretically the method that allows for the closest match, as the Spectrophotometer will be able to make a reading of a physical, existing sample. This method can also be fine-tuned as done in method three. As such, methods two, three and four can all be done in tandem if chosen, creating a wide array of useful tools for the user to colour match a spot colour. Ultimately, this experiment will attempt to prove that the built in Fiery RIP software is able to reproduce and match spot colours to a E of 1 or less, using several methods outlined in Section 3. When the Fiery RIP is used to match PANTONE® spot colours, the accuracy of the colour to the offset equivalent will be higher than if the match was done manually through user inputted values.

EQUIPMENT & MATERIALS

EFI EXP8000 Colour Server (EFI Fiery RIP) Fiery Command Workstation 4.5.1.9 Xerox DocuColor 7000 X-Rite 550 Spectrodensitometer (Serial no.: 011828) EFI ES-1000 no. 3.278-877623-5 Light Booth: Graphiclite, no. 59003 PANTONE® solid uncoated swatch book Report MultiPurpose (Suzano Pulp And Paper; 8.5” x 11”, 20lb, 75 g/m2)

General Setup: 01. 7 PANTONE® colours were chosen according to PANTONE® swatch book (For this experiment, PANTONE® 282 U, PANTONE® 3125 U, PANTONE® 5125 U, PANTONE® 699 U, PANTONE® 7506 U, PANTONE® Hexachrome Green U, PANTONE® 456 U was chosen to be tested). 02. In a design program, the 7 PANTONE® colours were organized and outputted to a single page PDF document as a test form for method 1 and Ensured proper PANTONE® channels were used, instead of CMYK equivalents. 03. Fiery RIP software was used in accordance to procedures outlined in method one (detailed procedures below). 04. Varying numbers of sheets of uncoated stock paper were printed using DocuColor 7000. 05. Steps 4 and 5 were repeated with procedures outlined in all subsequent methods (two, three and four). 06. 3 L*a*b* readings were made and recorded from each printed PANTONE® patch. 07. 3 L*a*b* readings were made and recorded from each source PANTONE® patch from the swatch book. To carry out this experiment exercising in industryused methods of replicating spot colours on a digital press, four methods were outlined and followed after preliminary setup of the working digital file. The goal was to obtain E between reproduced PANTONE® patches and respective source patches with equivalence to 1 or less.

Calibration of press is assumed. Instructions based off of Fiery Command Workstation 4.5.1.9. For newer software version instructions, refer to official PDF distributed on EFI Fiery’s website. Method 1- Spot Off 01. Created test form was sent to the Fiery server as a new job. 02. New active job was right clicked to show options, and Properties was clicked. 03. Within the Job Properties window, the ColorWise tab was clicked, revealing various colour related settings. The Expert Settings button was clicked to open a window detailing current colour workflow settings. 04. Spot Color Matching was unchecked, changing the workflow’s path for spot colours to divert upwards to the CMYK Simulation profile. 05. Test form was printed and each colour’s L*a*b* values were measured five (5) times and averaged. These values were then calculated with the source L*a*b* values for that colour for E. Method 2- EFI Fiery RIP’s Spot-On Default Feature 01. Within Fiery’s Command WorkStation, Server was clicked to produce a dropdown list of options. From that list, Manage Color... was clicked to open the ColorWise Pro Tools window. 02. The Spot On utility was clicked to open a new window detailing existing colour libraries. The Fiery DC8000 Uncoated 60-80 gsm V2F output profile was selected. 03. Spot Color Matching was ensured to be checked in the workflow by going to Edit > Preferences. 04. PANTONE® Uncoated was expanded and the

7 tested colours were chosen and highlighted together by holding down CTRL on the keyboard. 05. Without deselecting the chosen spot colours, the printer icon on the Spot On window was clicked and the chosen colours were printed via Fiery’s Spot On test form. 06. Test form was printed and each colour’s L*a*b*values were measured five (5) times and averaged. These values were then calculated with the source L*a*b* values for that colour for E. Method 3- EFI Fiery RIP’s PANTONE® Fine Tuning 01. Steps 1 to 3 from method 2 were repeated. 02. From the PANTONE® Uncoated library, each colour was found and the circle icon to the left of the spot colour’s name was double clicked to open the Spot On Colour Search dialog box. 03. The Spot On Color Search window was used to adjust and print each ring-around of a single spot colour. Color Spacing generates a varying level of colour difference for a more obvious visual difference, and Lightness/Saturation choices which affect how the colours will change in the ring-around. Monitor Compensation was checked to help visually differentiate each colour patch, but was not for soft proofing as the monitor was not determined to be colour accurate. 04. An ring-around of a spot colour was printed and matched to the source, the PANTONE® solid uncoated swatch book, and the closest match was chosen and clicked on the Spot On Color Search window. 05. The center patch of 2nd ring-around was used as the current colour output as the chosen patch from step 4 has moved to the center. Step 4 was

repeated up to a maximum of three times (to simulate time constraints and other possible production restrictions) after visual inspection and matching of the colour before accepting a colour as the new, fine-tuned spot colour. 06. The accepted colour was measured for L*a*b* values five (5) times and averaged. These values were then calculated with the source L*a*b* values for that colour for E. 07. Repeat steps 2-6 for the other spot colours from the chosen seven. Method 4- Colour Matching with Spectrophotometer 01. Steps 1 to 3 from method 2 were repeated. 02. From the Spot-On dialog window, go to Edit > New Group, and a new group library for new colours will be added to the list. The new group was renamed to reflect the experiment title. 03. The output profile “Fiery DC8000 Uncoated 6080 g/m2 V2F” was chosen. 04. The EFI ES-1000 spectrophotometer was calibrated by clicking Instrument > Start > ‘Choose Instrument’. From the dialog that appears, “EFI Spectrophotometer ES-1000” was selected. The spectrophotometer was ensured to be sitting flat in the cradle, and the calibrate button was clicked. 05. New custom colour in library were set up by clicking Edit > New Color. Under the group created earlier, a new colour will appear, and it was named appropriately for the colour about to be inputted. 06. The circle icon beside the new colour was double clicked, opening the Spot On Color Search window. 07. Using the spectrophotometer, one of the chosen seven PANTONE® test colours were scanned

from the source PANTONE® solid uncoated swatch book, by pressing the side button on the spectrophotometer. The scanned colour now appeared within the patches. 08. The colour was printed, and the center patches were measured for L*a*b* values five (5) times and averaged. These values were then calculated with the source L*a*b* values for that colour for -E. 09. Repeat steps 5-8 for the other spot colours from the chosen seven.

RESULTS The basis of the results are formed off the initial goal of aiming for a E of 1 or less, which is the industry standard of little to no colour difference unless to a trained eye. This goal, while challenging, simulates the importance of matching a PANTONE® spot colour on a digital press, since branding and brand awareness are closely linked with how a colour is perceived by an observer.

Figure 1 Fiery DC8000 Uncoated 60-80 gsm V2F Output profile and 7 spot colours, out of gamut

The out of gamut spot colours, PANTONE® Hexachrome Green U, 7506 U, 3125 U, and 699 U (Figure 1) will be expected to skew results as they will not be able to be reproduced accurately on the Xerox DocuColor using this specific output profile. They will be addressed with special considerations in section 5 of this report. In terms of the three in gamut PANTONE® spot colours chosen, PANTONE® 456 U, 5125 U and 282 U (Figure 2), these colours showed relatively good results in comparison to the experiment’s goals.

Figure 2 Fiery DC8000 Uncoated 60-80 gsm V2F Output profile and 7 spot colours, in gamut

PANTONE® 456 U (Figure 3), a shade of olive-like dark yellow, obtained very typical results, having a very high -E when printed with Spot Color Matching turned off, with a -E of 25. Visually, the colour printed with Spot Color Matching off was not close to the source (PANTONE® solid uncoated swatch book) at all. However, when using other methods, the E shows significant improvements, with the best method being method 3, Fine Tuning the colour through visual inspection. This method yielded a E of 6, and while this was a calculated value according to what was measured, a visual inspection showed little variance in the actual hue. PANTONE® 5125 U (Figure 4), a shade of dark purple, also has very similar results, but an even closer E than PANTONE® 456 U. With method 3 again, Fine Tuning of this purple to match the source resulted in a E of 2.5, an incredibly small visual difference under D50 lighting. It is important to note that the procedures of this experiment has limited Fine Tuning steps to three, i.e. three chances to get as close a visual match as possible. Even within three tries, PANTONE® 5125 U can already be matched to a E of 2.5 – with additional tries, a closer E is believed to be achievable. Like the previous colour, PANTONE® 5125 U has the largest E when Spot Color Matching is turned off from the workflow, signaling that using L*a*b* values from the manufacturer (in this case, PANTONE®) is not always the best method in every CMYK workflow. PANTONE® 282 U (Figure 5) was a colour chosen because of its hard to reproduce nature on offset presses. The colour is visually a very dark blue, and it was questioned as to whether the same difficulty of reproduction would

Pantone 456 U Spot On

Fine Tune

Spectrophotometer

25 20 15 10 5

Trials Figure 3 PANTONE® 456 U, in gamut Pantone 5125 U Spot On

Fine Tune

Spectrophotometer

15 10 5 0

Trials

Figure 4 PANTONE® 5125 U, in gamut Pantone 282 U Spot On

Fine Tune

Spectrophotometer

15 10 5 0

Trials Figure 5 PANTONE® 282 U, in gamut

be on a digital press. The results show that the 282 U has the lowest E calculated of the three in-gamut colours chosen, with a E of 1.5 using method 4 (spectrophotometer use). Method 3 resulted in a E of 3.4. It is interesting to note that this very dark colour is the only one that breaks the trend of visual inspection being the best way to get the closest PANTONE® match, and it may be that blues are harder to discern under bright illuminants (Kaiser, n.d.). PANTONE® Hexachrome Green U (Figure 6) is a bright green that is part of PANTONE® Inc.’s six-colour printing process. Being a colour that is hard to replicate on a normal offset press due to the addition of two extra colour channels for increased gamut volume printed, printing Hexachrome Green on a digital press was an experiment in itself. The results are bad, as expected of an out of gamut colour, with the lowest E achieved being 16 with Spot Color Matching turned off. The results were less desirable when trying to get the colour closer with methods 2, 3 and 4. These high E's are typical of colours that are out of gamut, as mentioned earlier, and must be dealt with differently than trying to aim for an exact match that cannot occur. In general, for the out of gamut colours, it appears that the E is relatively high if left to print straight from Spot Colour Matching on in the workflow. Though the L*a*b* values are taken from what the Fiery RIP believes to be the correct values for colour matching PANTONE® colours on a digital press, these values, based purely on E, are not visually as colour accurate as matching it through Fine Tuning, which ultimately gives a lower E on a whole, whether the colour is out of gamut or in.

Pantone Hexachrome Green Spot On

Fine Tune

Spectrophotometer

25 20 15 10 5 0

Trials Figure 6 PANTONE® Hexachrome Green U out of gamut

Figure 7 Proximity of Source spot colours and finetuned for a closer visual match

Using the ES-1000 spectrophotometer, method 4, did not give results that matched or surpassed Fine Tuning in terms of colour accuracy. Initially, it was assumed that using a device to measure colour would be more accurate than visually inspecting and matching colour, but in the case of the 7 chosen PANTONE® colours, looking at what was printed and visually matching to the sourced proved much more accurate than using a device. Overall, using the ring-around to fine tune the colours via visual inspection under a D50 illuminant made for the best and closest match, as seen in the gamut shown in Figure 7. Using Spot Color Matching on by itself also yielded similar results. Colours that were out of gamut were ‘remapped’ to within the gamut, and depending on which method was used, a different reassignment of L*a*b* values were used. Ultimately, these differences are up to user discretion, depending on what the end user’s preference for the spot colour is.

DISCUSSION & RECOMMENDATIONS Evaluation of Methods The method with evident inaccuracy of colour matching is turning Spot Colour Matching off. The E for this method averaged 5x - 8x higher the E of the best method. Through the printed samples, it is clear that the reproduced colours are more saturated than the referenced source colour. If colour accuracy is not a concern for the print job and the client prefers saturated colours, this method can be used for limited colour prints such as two colour jobs. Spot Off would be used

for evaluating the printer’s native output without colour management to print test targets (“Booksmart Studio”, n.d.). The CMYK values of Spot Off are provided by the spot colour manufacturer and are widely available on other commercial products i.e. Photoshop. (“Fiery Color Reference”, 2002).

black ink to make colour increasingly lighter or darker, respectively. Another limitation of Fine Tuning is the sole option of choosing colours from the ring around. It is very restrictive in cases where a specific ink level is determined to be adjusted a certain way but the ringaround does not provide this option.

The method a printer should use best depends on individual situations. Spot On’s CMYK equivalent values are assigned by EFI, the Command Station software’s manufacturer. These values were specifically deduced for users of the Fiery RIP for digital presses (“Fiery Color Reference”, 2002). The output of these values would be more accurate and relevant than the broader, more general values used in Spot Off. Spot On is the quickest color managed method to use. This method requires no external definition or alteration. As each press is different, adjustments may be required to produce a more desired colour. The next two methods using Fine Tuning and the spectrophotometer will allow for adjustments.

Using a spectrophotometer provides a quick alternative to getting a good colour match requiring only seconds after the initial setup. This method requires the presence of a physical sample with the desired colour to scan L*a*b* values. Custom colours made from Fine Tuning and using the Spectrophotometer can be saved in custom spot colour libraries to be accessed at a later time. Well maintained libraries help achieve colour consistency across documents and jobs for clients. In general, using Spot On, Fine Tuning, or the spectrophotometer, increases colour accuracy. The process of matching colours manually would be greatly improved through colour managed monitors.

If colour matching is of high importance and the accuracy is worth more than the excess time spent on manual labour, the Fine Tuning method would be the best choice. This should also be the method of choice if the customer requires a specific shade of colour for branding that needs a close visual match. This method is highly time consuming and requires an operator with excellent colour acuity. Fine Tuning has its own limitations. By using the lightness option to generate options, there is a certain point where no further alternate option is generated. This occurs when a colour with black ink levels at either 0 or 100, prevent the system from further decreasing or increasing the

Workflow In a Spot Off workflow (Figure 8), a spot colour immediately gets converted into CMYK device counts by solely using general and universal values. These values do not include customized factors such as paper and output device.

Figure 8 Spot Color Matching off, workflow

Figure 9 Spot Color Matching on, workflow

In a Spot On workflow (Figure 9), the spot colour goes straight into EFI Fiery’s Spot Color Matching before output. By ignoring manufacturer CMYK recommendations, such as using PANTONE® Inc.’s recommended CMYK breakdown, the built-in Spot Color Matching that the EFI Fiery RIP has will automatically use the DocuColor’s recommended

Pantone 7506 U

7 Pantone Colours with 4 Methods CMYK equivalents for certain spot colour libraries. This has the major advantage of using CMYK values that are not for a generic press. CMYK values are often for the capabilities of generic offset press profiles, such as General Requirements for Applications in Commercial Offset Lithography (GRACoL) or Specifications for Web Offset Publications (SWOP), and ultimately these CMYK values do not translate to digital press toners. As such, using Spot Color Matching in a digital press’s workflow can automate the calculations of spot colours to what Fiery deems as the best match. Referring to the graph, Figure 10, a major trend is the ineffectiveness of the Spot Off method for the in gamut colours, PANTONE® 282 U, 456 U, and 5125 U. Effectiveness of other three methods are varied, but is significantly better than Spot Off.

Spot On

Fine Tune

Spectrophotometer

30 25 20 15 10 5 0

5125 U

282 U

3125 U

Trials Figure 10 Four methods in seven PANTONE® colours Pantone Hexachrome Green Spot On

Fine Tune

Spectrophotometer

25 20 15

PANTONE® 7506 U, 3125 U, 699 U and Hexachrome Green U are out of gamut. Results of methods that require little to no operator intervention (Spot Off, Spot On, and spectrophotometer) should be avoided and used in moderation, due to a large variance in resulting E values. Only the manual method, Fine Tuning maintained accurate colours. Although this method maintained the lowest -E of all methods, it does not necessarily mean an accurate colour. A better r epresentation of the colour may have a larger -E. There are many possibilities when an out of gamut spot colour gets remapped within the gamut to become printable (Levoy, 2011). The out of gamut spot colours closest to the gamut’s edge should have the lowest E due to the short distance to possible remapping options. The resulting colour

Spot On

Colours that are out of gamut, such as the graphs shown from Figure 11 - 14, generally do not reproduce well using Spot Color Matching on by itself, and are printed best after Fine Tuning from a press operator. The E values of these colours will be high, due to the fact that the L*a*b* values from the source (the PANTONE® solid uncoated swatch book) have values that cannot be reproduced with the colour capabilities of the Xerox DocuColor 7000. In situations where the printer must print spot colours that are out of gamut of the digital presses capabilities, they will have to make the decision with the client to decide how to treat these out of gamut colours. Similar to how conversion of ICC profiles include rendering intents that shift colour depending on different behaviours, each method had a different way of approaching the shift of out-of-gamut colours. A high E, though visually different from the source, may be a colour that is visually pleasing and acceptable to a client, so proper deliberation of each method and the outputted spot colour match should be done.

Trials

Figure 11 PANTONE® Hexachrome Green U, out of gamut may not be the best colour as lower E does not indicate the best representation. In these scenarios, it is highly dependent on the client’s requirements. For example, the resulting colour may be remapped with lower lightness to maintain hue, but client may prefer a shift that maintains saturation.

There are many factors contributing to the discrepancies between offset lithography press and digital press PANTONE® matching. Digital presses have larger gamut than offset CMYK workflows (Rys, 2012). (Figure 11) shows an overlapping comparison between a digital gamut and an offset lithography gamut. Although offset gamut has some areas of colours (lighter shades of red and cyan) that surpasses the digital gamut, digital has much more reproduction capabilities in the dark colors. The gamut volume of the DocuColor (467,453) far exceeds that of an offset press with CMYK workflow (403,137) by almost 14%. Additionally, digital presses use dry toner instead of wet ink offset presses use. The significance is how the dry toner remains at the surface of the substrate rather than

Spectrophotometer

15 10 5 0

Trials

Figure 12 PANTONE® 7506 U, out of gamut

Pantone 699 U Spot On

Fine Tune

Spectrophotometer

20 15 10 5 0

Trials Figure 13 PANTONE® 699 U, out of gamut

5 0

Fine Tune

Pantone 3125 U Spot On

Fine Tune

Spectrophotometer

20 15 10 5 0

Trials Figure 14 PANTONE® 3125 U, out of gamut

ACKNOWLEDGEMENTS the wet ink soaking into the substrate. The heated toner produces colour that is observably brighter and glossier, subsequently affects the way colour is perceived. A survey has shown statistics that digital presses are found to have more colour variation than offset by almost 1/3 (Chung, 2007), further creating inconsistencies in spot colour reproduction.

CONCLUSION & PERSPECTIVES For this experiment, the goal of matching the chosen PANTONE® colours to a E of 1 or less was established at the beginning, as a baseline for the matching methods devised. By using this baseline, the effectiveness of each method could then be determined from this viewpoint. The hypothesis made at the beginning of the experiment states that of the four methods, using a spectrophotometer would be the most accurate at measuring the L*a*b* values of an existing sample and replicating it through Spot Color Matching, but it was confirmed that through visual inspection and manual colour matching through Fine Tuning (method 3) a match with the lowest E would be created. This experiment, while it uses E as a baseline for what is determined to be a good or bad match, furthers this conclusion with the fact that while some methods of matching may yield a low E, the end user may purposely chose to find a match that simply looks more visually appealing than a match with the lowest colour difference. While that may defeat the purpose of a spot colour, many colours that can be picked from a PANTONE® solid uncoated swatch book, like the ones picked for the purpose of this experiment, are out of gamut

of what the digital presses’ capabilities are. Colours that are out of gamut must be remapped, similar to rendering intents of normal CMYK workflows, to a colour that is to the client’s discretion. Each method explored in this experiment gives a different, remapped CMYK equivalent match of the initial PANTONE® spot colour, so depending on what the client chooses, a different method can be recommended on the basis of their needs.

We would like to give special thanks to the school of Graphic Communications Management at Ryerson University for allowing us to use the Xerox DocuColor 7000 and associated EFI Command Workstation/Fiery RIP, and all materials, for the experiment.

Ultimately, the main reason for the difficulty in reproducing spot colours on a digital press lies in the colour technology that the Xerox DocuColor 7000 prints with. In a digital press, the way the toner fuses and sits on top of paper has a dramatic difference in the way colour is then perceived, in stark contrast to the way ink is absorbed into the substrate and paper colour shifts colour in offset printing. The difference in the way ink is applied to the substrate in both technologies should be addressed when relating the PANTONE® libraries to digital printing. PANTONE® libraries are ultimately made for the purpose of mixable inks, and therefore translation to CMYK equivalents still require a manual touch for the closest colour match.

Booksmart Studio. (n.d.). Color Managment Tutorial, Printer Profiling. Retrieved April 11, 2013, from Booksmart Studio: http://www.booksmartstudio. com/color_tutorial/printers.html

All in all, getting a PANTONE® spot colour match on a digital press is not impossible--it is the process that needs to be defined and methodology that printers need to understand to get the closest match for their clients, in the situation where spot colours must be printed on a digital press. Given time, as digital print technology continues to expand, better spot colour matching will become more prevalent and more intuitive, for the widest range of colour in this growing sector of print.

REFERENCES

EFI. (2011, February 23). Fiery Servers: The Easiest Way to Get the Right Color Every Time. Retrieved February 12, 2013, from EFI: http://w3.efi.com/~/media/Files/ EFI/Fiery/dm/EFI_Fiery_Spot_On_WP Fiery Color Reference. (2002). Retrieved April 10, 2013, from Oce Printers, Copiers, and Plotters: http:// www.oceusa.com/main/view_media.jsp?WebLo gicSession=pLghLZHZFmNBGQZSnhq6VFvj1NZ GKn 6YHv1PnNJJ2QQs2KHNKbhL!680376898 Howard, A. (2012, July). Accurately Reproducting PANTONE® Colors on Digital Presses. Retrieved March 5, 2013, from California Polytechnic State University: http://digitalcommons.calpoly.edu/ cgi/viewcontent.cgi?article=1084&context=grcsp Kaiser, P. (n.d.). Lambdas. Retrieved April 11, 2013, from The Joy of Visual Perception: http://www. yorku.ca/eye/lambdas.htm

Levoy, M. (2011). Gamut Mapping. Retrieved April 11, 2013, from CSE131: The Science of Digital Photograph: http://www.cs.washingtonedu/educ ation/courses/cse131/12sp/ appletsgamutmapping.html PFL. (n.d.). Printing with PANTONE® Colors and Spot Colors. Retrieved February 12, 2013, from Printing for Less: http://www.printingforless.com/PANTONE®Colors-in-Printing.html Robert Chung, M. R. (2007, July). A Survey of Digital and Offset Print Quality Issues. Retrieved April 11, 2013, from RIT Printing Industry Center: http://print.rit. edu/pubs/picrm200604.pdf Rys, R. (2012). When to Print Digital Versus Offset Printing - Think Outside the Gamut. Retrieved April 11, 2013, from HiDef Color: http://www.hidefcolor. com/marketing/print-digital-offset-printing/ Xerox. (2010, September 10). Spot Colors and Fiery: Everything You Need to Know (Part 1). Retrieved April 10, 2013, from Digital Printing Hot Spot: http:// digitalprinting.blogs.xerox.com/2010/09/spotcolors-and- fiery-everything-you-need-to-know/

Understanding the New ISO 13655 Measurement Standard in Press and Proofing Applications

ABSTRACT There is a revised ISO standard that specifies the illuminant characteristics when using a measuring instrument to measure printed samples. The standard is called ISO 13655:2009 - Graphic technology — Spectral measurement and colorimetric computation for graphic arts images. In this research the instrument measurement modes described by ISO 13655are examined: M0 – legacy mode, M1 – UV included, M2 – UV excluded, M3 – polarizing mode (for measurement of wet offset press sheets).

ISO 13655 IN PRESS AND PROOFING APPLICATIONS

Erin Luu

In this research the different modes were used to measure papers with and without optical brightening agents (OBAs). Four samples of OBA papers were measured and these show a peak in the blue spectrum due to OBA induced fluorescence. Two proofing papers with no OBAs were also measured, in the case of papers containing no OBAs, M1 and M2 measurements are identical, as expected, as the presence of ultraviolet (UV) light in the illuminant only has an effect on the measurement when the samples contain OBAs.

Many of the metrics in offset printing rely on process control using wet press sheets. It is understood and accepted that these sheets will “dry back” resulting in a lower final density measurement. One of the measurable differences between wet and dry sheets is gloss. Measurement mode M3 using a polarizing filter was tested to see if it can predict the density of a dry sheet from a measurement of a wet sheet as claimed by the instrument manufacturers. In this research paper the Konica Minolta FD-7 spectrophotometerwas used and the X-Rite i1iSisXL and 530 instruments werealso used in this testing. A press run was done using coated and uncoated paper on Ryerson University’s Heidelberg PM74,4-color,offset press. We conclude that the ISO 13655 standard needs to be widely implemented by instrument manufacturers in order to improve inter-instrument agreement between different makes and models.

INTRODUCTION In the printing industry, one of the major considerations is the ability to deliver accurate and consistent colour to the customer. Colour matching is now often done using instrumentation via a process known as “printing to the numbers”. The numbers in this context are usually CIELAB characterization data values that are measured and monitored via use of a measuring instrument. One challenge with instrumentation has been that the ultraviolet (UV) component in the measuring illuminant of different instruments can be different, which causes different instruments to give different readings for the same sample. If the paper or ink exhibit fluorescent behaviour then there are variations between measurement devices when measuring the same press sheet. The use of optical brightening agents (OBAs) has become very common in paper manufacturing. The OBAs within the paper coating reacts to the UV content in the illuminant, the UV light is absorbed by the OBAs and emitted in the blue part of the spectrum, counteracting the yellow hue naturally found in wood fibre paper, thus making papers appear bluer and brighter. The UV component in a measuring instrument lamp may be small or may not be well defined. When the UV component in a measuring instrument is not well defined, this causes the following problems when measuring samples that exhibit fluorescence: there are inter-instrument differences as each measuring device measures the same sample differently there are problems with determining a “real” or “absolute” measurement in order that the measured colour will correlate to the colour created when the same sample is viewed in a viewing booth.

X-RITE NETPROFILER 3 Both the above issues can be solved if the amount of UV light in the illumination (measuring instrument and/or viewing booth) is better defined.

MEASURING REFLECTANCE It is not immediately obvious why the UV light should be known (and the same) for all spectrophotometers. Normally there is no requirement for all instruments to have the same spectral emission characteristics. Instruments are designed to have different light sources yet they can still measure the same, this is because when measuring the sample spectrum we are only considering the % reflectance, so there is no requirement for different instruments to have the same illuminant properties. A spectrophotometer measures the sample’s spectrum wavelength by wavelength. The instrument records what percentage of each wavelength is returned by the sample. A light internal to the instrument is used to make this measurement. Generally the actual source used in the instrument is irrelevant and not considered in the calculation of the object spectrum (Sharma, 2004). All that happens during a spectral measurement is that we compute the percentage of the light reflected back at each wavelength. Thus, if we illuminate the sample with 40 units of light and we get back 20 units of light, the reflectance at this wavelength is 50%. An important feature of the spectrum is that it forms a colour description of the object independent of the illuminant. In other words the instrument illuminant can be very different in different devices yet they can all report

the same spectral reflectance or “colour footprint” for a given sample. (There are limiting conditions due, for example, to low light levels and signal to noise ratio.) The above description is correct and works well for most situations, however in the case of UV light and OBAs the above theoretical explanation breaks down and we see that the measured spectrum instead of being independent of the illuminant actually starts to be dependent on the instrument’s illuminant. The UV in a measuring instrument represents the classic case of the observer causing an effect in the experiment that they are trying to observe because the light used to determine the % reflectance is itself influencing the measurement. In the case of samples containing OBAs, UV from the measuring light is absorbed and emitted in the blue part of the spectrum, so we are no longer simply measuring the % reflectance because the light used to probe the characteristics of the sample is changing the sample’s characteristics. When OBAs are present, then the % reflectance will change with the amount of UV in the measuring instrument and different instruments from different manufacturers will compute different spectral data for the same sample.Of course if the sample contains no OBAs then everything works according to theory and we can mathematically determine the ratio of incident to reflected light and correctly compute the % reflectance at all wavelengths. In other words it is easy to see that the reported colour for a sample will be different based on the UV content emitted by the measuring instrument. So to deal with the issue of UV induced fluorescence we must specify the illuminant characteristics, and when that is done we see closer inter-instrument agreement.

There is the need for better inter-instrument agreement because a range of physical devices are used in printing. A device may be inline (Xerox iGen4), or in an offset press-side scanning system (X-Rite IntelliTrax), press operators may use handheld spot instruments, or the colour management professional may employ a chart reading device (Barbieri Spectro Swing or X-Rite i1i0). When there are so many different devices used, it is imperative that different devices provide similar readings for the same sample. In order to improve inter-instrument agreement we need to clarify the description of UV in measuring instruments.One way to deal with instrumental differences is to construct a device mapping solution; a commercial example may be X-Rite NetProfiler 3, which seeks to align instruments via a device “profile”.

ISO 13655- M0, M1, M2, M3 The problem to date has been that the UV component in the measuring instrument was not specified and while the illuminant should theoretically not affect the measurement, we have seen that in the case of OBAs it does.An ISO standard – ISO 13655 - has recently been revised and the new revision provides much more clarity for the illuminant and measurement modes. ISO 13655 was originally published in 1996 and revised in 2009. The new version is called ISO 13655:2009 - Graphic technology — Spectral measurement and colorimetric computation for graphic arts images (ISO, 2009). The new version of the standard now defines four measurement modes - M0, M1, M2 and M3. Suppliers have produced documents that seek to educate their customers on

implementation and relevance of this revised ISO standard (Cheydleur & O’Connor, 2011, Konica Minolta, n.d.) and a respected blog site has reviewed the KonicaMinolta FD-7 instrument that offers all modes M0-M3 (ColorWiki, n.d.). The new ISO stipulated instrument measurement modes are:

M0 – legacy mode (based on Illuminant A - tungsten bulb found in older devices) M1 – D50 mode M2 – UV-cut mode M3 –polarizing mode (for measurement of wet offset press sheets).

measurements. It can be used in situations where it is not necessary to know the “absolute” measurement value, and there is no exchange of information or correlation with other measurement scenarios. An M0 instrument may not read the same as another instrument that is measuring the same sample, but an M0 instrument is expected to read the same day after day. We would say that a M0 instrument is repeatable or consistent, but not necessarily accurate. (There is the special case of no OBAs. In this instance, it is true that measurements performed under M0 should be completely accurate.)

M0 is known as the legacy mode and is a standard that expresses the majority of measuring instruments used in the field today. It is directed to instruments that use an unfiltered gas-filled tungsten lamp to illuminate the sample being measured. Prior to LED based devices (e.g. X-Rite i1iSis), the tungsten bulb based device was the primary type of device in the market. We remind the reader that a CIE Illuminant such as Illuminant A is simply an approved spectral graph and not a real physical bulb of any sort, so in practice we say that we would expect that “the light contained within the instrument should have a correlated colour temperature of 2856K”. It should be noted that in this mode, the light is neither UV-filtered nor polarized. The UV component can be very weak, Figure 1.

M1 is also known as the “D50 mode” or “UV included mode”. A major difference (and improvement) over earlier configurations is that the amount of energy in the UV and visible wavelengths is now specified. The light source in the instrument must match CIE Illuminant D50. Again it is well known that D50 is simply a spectral power distribution curve and there may be different ways to elicit a D50 response. ISO 13655 allows for different methods to achieve conformance to the M1 illumination condition and whichever method a supplier chooses, “…the instrument manufacturer should supply a representative spectral power distribution of the measurement source….” D50 is one of the standard viewing booth modes, the basis for the Profile Connection Space in the ICC architecture, all of which makes M1 the most desirable mode for today’s colour measurement and colour management systems. We hope that all supplier systems are soon updated to meet this revised international standard.

An M0 instrument can safely be used for process control applications where it is adequate to make repeatable

M2 is defined as a“UV-cut” mode. ISO 13655 states that “….to exclude variations in measurement results

between instruments due to fluorescence of optical brightening agents….. the spectral power distribution of the measurement source ….shall only contain substantial radiation power in the wavelength range above 400 nm….” How is this mode used in practice? There will be times when a customer will request a print to be measured using M2 because the lighting used to display the job is expected to be free of UV content. A museum is an example of one of the major places that uses UV-free lighting. In colour management circles OBA induced color shifts were often dealt with by removing UV light from both the measuring system and the viewing conditions. Now with the new standard we have a specific definition for “UV-cut” and the wavelength at which it happens. Note that the rest of the illuminant spectral power distribution for M2 is not specified – it does not have to be, as in this spectral range we are in a situation where the instrument illuminant does 250

not interact with the specimen or change the spectral response in these wavelengths, so it is not necessary to define the spectral power distribution of the source from 400 to 700 nm, and a measuring instrument in this range can simply compute the % reflection via the process described earlier. M3 is a polarizing mode and consists of UV-cut up until 400 nm and then a polarizing filter is also applied to the remaining wavelengths. As above, the illuminant spectral power distribution from 400-700 nm for M3 is not specified – it does not have to be, as in this spectral range we are in a situation again where the instrument illuminant does not interact with the specimen. The main use of M3 is to limit or completely remove surface reflections. In the offset printing industry, the customer pays for the final dry product. One of the main concerns is that the press sheets come off the press wet and as they dry, the density of the ink drops. The M3 mode can aid printers in cutting 150

Illuminant A

100

200

D50

150

60 100

50 0 200

20 300

150 100

UV Cut

400 500 600 Wavelength (nm)

700

800

0 200 150

100

UV Cut

400 500 600 Wavelength (nm)

700

800

Polarizer

80 60

0 200

300

400 500 600 Wavelength (nm)

700

800

0 200

300

400 500 600 Wavelength (nm)

700

800

Figure 1: The 4 measurement modes (M0, M1, M2, M3) stipulated in ISO 13655:2009.

the surface gloss from wet inks, and if drying is primarily represented by a change in surface gloss then by removing the gloss we may have a better prediction of the final expected dry density. Because of the polarizing filter the measured density value using M3, may be different to the density achieved from the other modes. In fact in the data we see that each mode (M0-M3) can produce a very different spectral response and thus any computed metrics (CIELAB, CIEYxy, density) can be different. In order to report measurement data in an unambiguous way, ISO 13655 suggest the following nomenclature “ISO Standard/measurement mode/white backing (wb) or black backing (bb)/make and model”. In our data for example we would therefore quote the following:

ISO 13655 (M0, wb, X-Rite i1iSis) ISO 13655 (M3, wb, Konica-Minolta FD-7).

VIEWING BOOTHS The clarification for illuminant in measuring instruments (ISO 13655:2009) is accompanied by a similar clarification in the standard for viewing booths ISO 3664:2009 (ISO, 2009). Via updated standard ISO 3664:2009, emphasis has turned to requiring a closer simulation of Illuminant D50 thus clarifying the amount of UV illumination in the viewing booth. The new viewing booth standard refers to issues such as excluding stray light and that the walls of the booth should be a type of neutral gray, but in the current context, ISO 3664 has called for tighter tolerances on the quality of the light source to ensure that it closely matches the D50 (M1) curve especially in the UV spectrum (Dalton, 2010).

We are at a truly exciting juncture in colour management technology – we have a clear specification for the UV component in both the measuring instrument and the viewing booth, together these are able to deal with the challenges of OBA-induced colour changes.

POLARIZING FILTERS AND DENSITY DRY BACK It is generally agreed that a polarizationfilter can give less difference in densityreadings between a wet and a dried-back printed sheet (Tobias Associates, 2010). This effect, however, is not always consistent, since ink ‘‘soak-in’’ depends upon the constitution of the ink as well as the porosity of the paper. When a wet film of ink is applied to paper, the surface of the ink is fairly smooth. The instrument illuminates the ink surface vertically and views the reflected light at 45° (or the converse geometry). Thus, the density measured, approaches the true diffuse density of the body of the ink. As the ink dries, the surface becomes rougher and, under normal conditions, the density is lowered due to an increase in surface reflections. The effect of these surface reflections can be substantially reduced by the use of a polarizing filter to give us a better predictor of dry density from wet density readings. There is considerable debate around the use of polarization filters for density dry back measurements. The use of polarization filters is somewhat controversial, since the effect is not controllable and each situation will produce different results, until now there have been no published standards for the use of polarization filters. The situation was akin

to the use of UV light in the instrument, it was not stipulated or clearly defined. ISO 13655 now clarifies the situation for the response of the polarizing filter (and clearly defines the UV component). In this experiment offset press dry back was measured with a polarizing instrument (Konica-Minolta FD-7) and compared to a non-polarizing instrument (X-Rite 530).

(uncoated). Hostmann Steinberg Perfexia (PXV) CMYK process inks were used.The KonicaMinolta FD-7 was utilized in M3 mode to study the effect of using a polarizing filter in limiting surface gloss and measurement of ink dry-back. Comparator measurements were also made with the X-Rite 530 spectrophotometer.

EXPERIMENTAL DETAILS In order to investigate the use of the new ISO 13655 measurement modes a Konica-Minolta FD-7 spectrophotometer was used to measure papers containing OBAs and proofing papers without OBAs.The experiment compared the effects of measuring samples with modes M0, M1, M2 and M3. The spectral reflectances of six different types of paper were measured, four samples contained OBAs and two samples were proofing papers with no OBAs. The same samples were also measured using an X-Rite i1iSis spectrophotometer that provides M0 and M2 modes only.The data between the two were compared for the modes where commonality exists. In the second part of the experiment, an offset press run was conducted to measure dry-back of ink density for offset inks. In this part of the experiment, a job was run on coated and uncoated paper using Ryerson University’s 4-colour Heidelberg PM74 press.The paper used was 100M Condat Supreme Gloss Text (coated) and 70M Willamburg Offset Smooth

FD-7 - Epson Gloss (OBAs) M2

M1 1.9

1.7

1.5

1.3 1.1

Reflectance

1.9

1.3 1.1

0.9

0.7

450

500

550

600

Wavelength (nm)

650

700

0.5 350

750

400

FD-7 - Colonial Bond (OBAs) M1

450

500

550

600

Wavelength (nm)

650

700

0.5 350

750

1.7

1.5

Reflectance

1.7

Reflectance

1.9

1.3 1.1

0.9

0.7

450

500

550

600

Wavelength (nm)

650

700

750

0.5 350

400

450

500

550

600

Wavelength (nm)

650

550

600

650

700

750

700

750

1.1

0.9

400

500

1.3

0.9

0.5 350

450

1.9

1.1

400

In Figure 4, we see that the 4 samples measured using the Konica-Minolta FD-7 with OBAs provide different spectral responses for each of the 4 measuring modes (M0, M1, M2, M3) as expected. The M1 mode which includes UV light, shows a surge in the blue spectra from 400-450 nm, as expected. In the lower graphs of Figure 4, we see measurement of the white point of two papers with no OBAs. In these graphs we see overlap of M0, M1, M2 graphs, but that M3 with a polarizing filter creates a â&#x20AC;&#x153;lowerâ&#x20AC;? response curve and appears to be a darker colour. This is the expected response as a polarizing filter removes some of the measured light causing it to appear darker (M3).

FD-7 - GMG Proof Paper (No OBAs)

1.9

1.3

RESULTS

Wavelength (nm)

FD-7 - Office Paper (OBAs) M4

1.1

0.9

400

1.3

0.9

0.5 350

Reflectance

FD-7 - CGS Oris Proof (No OBAs)

1.9

Reflectance

FD-7 - Epson Matte (OBAs)

0.5 350

400

450

500

550

600

Wavelength (nm)

650

700

750

Figure 4: The Konica-Minolta FD-7 spectrophotometer was used to measure the paper white of 4 samples with OBAs (top), and 2 samples with no OBAs (lower).

iSis - Epson Gloss (OBAs)

Espon Matte (UVCUT)

Epson Glossy (UVINCLUDED)

Epson Glossy (UVCUT)

iSis - CGS Oris Proof (No OBAs)

Epson Matte (UVINCLUDED)

CGS Oris Proof (UVCUT)

1.9

1.7

1.5

1.3 1.1

Reflectance

1.9

Reflectance

iSis - Epson Matte (OBAs)

1.3 1.1

1.1

0.9

0.7

400

450

500

550

600

650

700

0.5 350

750

400

450

500

550

600

650

Wavelength (nm)

iSis - Colonial Bond (OBAs)

iSis - Office Paper (OBAs)

Colonial Bond (UVCUT)

700

0.5 350

750

450

500

550

600

650

700

750

iSis - GMG Proof Paper (No OBAs)

Colonial Bond (UVINCLUDED)

GMG Proof Paper (UVCUT)

In Figure 5 we see that the 4 samples with OBAs (top graphs) measured using the X-Rite i1iSisXL spectrophotometer provided different spectral response for each of the two measuring modes (M0 and M2) as expected. The implication from the top graphs is that the M0 mode implementation in the iSis includes a UV component, which is seen as a boost in the 400-450 nm (blue region) of the measured spectra. In the lower graphs of Figure 5 we see measurement of the white point of two papers with no OBAs. In these graphs we see more similarity between graphs for M0 and M2, as the papers have no OBAs thus both measurement modes produce similar results. Figure 5 helps us understand and “reverse engineer” the X-Rite implementation of M0 measurement mode in the i1iSis (LED) based instrument.

GMG Proof Paper (UVINCLUDED)

1.9 1.7

1.5

1.3 1.1

Reflectance

1.7

Reflectance

1.7

1.3 1.1

0.9

0.7

0.5 350

400

Wavelength (nm)

1.9

Reflectance

1.3

0.9

0.5 350

CGS Oris Proof (UVINCLUDED)

400

450

500

550

600

Wavelength (nm)

650

700

750

0.5 350

400

450

500

550

600

650

700

750

Wavelength (nm)

0.5 350

400

450

500

550

600

650

700

750

Wavelength (nm)

Figure 5: The X-Rite i1iSisXL spectrophotometer was used to measure the paper white of 4 samples with OBAs (top), and 2 samples with no OBAs (lower). The iSis only offers M0 and M2 measurement modes.

iSis - Epson Gloss (OBAs) M2: iSis

M0: FD-7

iSis - Epson Matte (OBAs) M2: FD-7

1.2

M2: iSis

M0: FD-7

iSis - CGS Oris Proof (No OBAs) M2: FD-7

1.2 1.0

0.8

0.6 0.4

0.0 350

0.6 0.4

400

450

500

550

600

Wavelength (nm)

650

700 750

0.0 350

M0: iSis

M2: iSis

M0: FD-7

1.2

400

450

500

550

600

Wavelength (nm)

650

0.0 350

700 750

M0: iSis

M2: iSis

M0: FD-7

M2: FD-7

1.2

0.8

0.0 350

Reflectance

0.8

Reflectance

1.0

0.6 0.4 0.2

0.2

400

450

500

550

600

Wavelength (nm)

650

700 750

0.0 350

M2: FD-7

The data shown in Figure 6 has been presented earlier. This is data already shown in Figures 4 and 5, but extracted here for easier comparison. The data shows that there is still some difference between the two instruments despite both devices claiming to measure using modes M0 and M2. In order to comment more fully, it would be necessary to compute the difference between these graphs either as a CRI or by converting the spectral data to L*a*b* and then computing a Delta E colour difference.The difference in the spectral graphs contributes to inter-instrument disagreement. 400

450

500

550

600

Wavelength (nm)

650

700 750

iSis - GMG Proof Paper (No OBAs)

1.0

0.4

M0: FD-7

0.4

1.0

0.6

M2: iSis

0.6

iSis - Office Paper (OBAs) M2: FD-7

M0: iSis

0.2

iSis - Colonial Bond (OBAs) 1.2

Reflectance

1.0

0.2

Reflectance

M0: iSis

1.0

Reflectance

1.2

M0: iSis

M2: iSis

M0: FD-7

M2: FD-7

0.6 0.4 0.2

400

450

500

550

600

650

700 750

Wavelength (nm)

0.0 350

400

450

500

550

600

650

700 750

Wavelength (nm)

Figure 6: The Konica-Minolta FD-7 and X-Rite i1iSisXL spectrophotometer data is compared for modes common to both instruments â&#x20AC;&#x201C; M0 and M2.

FD-7 - M3 - Offset- Coated Magenta

Black

Yellow

Cyan

1.7

1.5

Reflectance

Cyan

X-Rite 530 - Status T - Offset- Coated

1.3 1.1

0.7

0.5

Wavelength (nm) FD-7 M3 - Offset- Uncoated Magenta

Black

Yellow

1.7

1.5

1.3 1.1

0.7

Wavelength (nm)

Magenta

Black

0.5

Yellow

Wavelength (nm)

Figure 7: In an offset press run on Heidelberg 4-colour PM74 CMYK ink densities were measured on coated and uncoated stock to evaluate dry-back using the new M3 mode compared to the traditional status T density.

Figure 7 shows results from the dry-back test. The press form was printed on coated stock (100M Condat Supreme Gloss Text) and then the press was stopped and uncoated paper (70M Willamburg Offset Smooth) was loaded. In both instances, during make ready, the press was set to achieve wet target house densities. A press sheet was pulled from the press and measurements were done every 5 minutes for a total elapsed time of 75 minutes. Solid 100% C-MY-K patches were measured from a control bar in a single location on the edge of the press sheet. Measurements were taken using the Konica-Minolta FD-7 and then the same patches were used and re-measured using the X-Rite 530 device. Figure 7 shows that for this paper-ink combination there was in fact not much dry back and the measured density did not drop significantly. The amount of dry-back from this test is not significant, with inks changing by a density of perhaps 0.01 every five-minutes from when the sheet was first pulled from the press. The density continues to drop slightly and the rate of the change decreases as the ink begins to dry. From start to finish there was only around a total of 0.1 density difference. The difference between a wet and a completely dry ink film ranged from 0 to 0.16.The black density for both the FD-7 and 530 devices showed the most dry back.

1.1 0.9

1.3

0.9

Cyan

1.7

0.5

Wavelength (nm) X-Rite 530 - Status T - Offset- Uncoated

Reflectance

Cyan

Yellow

1.1 0.9

Black

1.3

0.9

0.5

Magenta

The expectation for the Konica-Minolta FD-7 in M3 – polarization mode, was that the initial and final density measurements would be similar. The behaviour of the paper-ink combination in this instance unfortunately did not demonstrate huge dry back, and so it is difficult to draw any sensible conclusions regarding the ability of M3 mode to measure and predict density change due to ink dry back. In general we had hoped to see that when the first reading is made, the polarization filter reduces the surface gloss and makes a reading of a smooth ink film. As the ink dries, the substrate absorbs the ink and there is no longer that smooth surface to measure from, resulting in a different “true” density when the sheet fully dries (Tobias Associates, 2010). An important result to note is that M3 mode employs a polarization filter, so density measurements in M3 mode are darker than “direct” measurement of density. Some observers suggest that M3 should not be used if trying to obtain perfect colour match because polarization filters will “darken” the measured colours. There is also the issue of the signal to noise ratio, where the reduced light levels require increased integration time. Some observers suggest that M3 is only effective if a high density of ink is used during printing (Meffre, 2013).

A similar academic project was conducted at Rochester Institute of Technology (Techavichien, 2004). In that study it was suggested that black and yellow has the least dryback and magenta was found to have the most. However, the results from this paper-ink combination suggest that black had the most dry-back for both uncoated and coated papers.

CONCLUSIONS The addition of OBAs in substrates introduces many challenges for accurate and consistent colour measurements because different instruments can report different measurements of the same sample. ISO13655 has helped clarify the UV composition in the instrument configuration, and has defined four measurement modes for the industry – M0, M1, M2, M3. We note that M0 type instruments are expected to be very repeatable and can therefore continue to be used for process control applications where we need only to measure consistently from day to day and are used where is there is no need to correlate with other external measurements. We saw that M1 is UV included and akin to D50, and M2 is UV excluded. In this experiment the effect of OBAs in printing paper was clearly demonstrated. If a sample contained OBAs then the spectral response clearly showed a peak in the blue part of the measured spectrum. If a sample contained no OBAs then measurement in M1 (UV included) and M2 (UV excluded) created the same response, as expected. M3 is a polarization mode and we see that the reported density will often be “darker” when compared to the other modes as some light is removed by the polarization filter. The M3 mode may be used to measure and predict dry back of printing inks. The amount of dry back in the chosen press run was not large, and it was therefore not possible to make convincing conclusions regarding the efficacy of the M3 mode in predicting dry back.

ACKNOWLEDGEMENTS In future work, we would consider instruments from more suppliers, e.g. Techkon and Barbieri. New analysis should also include a calculation of CIELAB values from the spectral data as this expresses any differences can be weighted by the Standard Observer colour matching functions which would “weight” the differences in the graphs according to human perceptual sensitivity. Also if a newer Delta E equation, such as Delta E (2000) was used to compute the colour difference this would continue to express the data in a perceptually relevant manner. Measuring instruments are supposed to provide a reliable and robust method for colour measurement, unfortunately in the case of UV and OBAs there has been considerable confusion and lack of inter-instrument agreement. The new ISO 13655 standard for instruments and ISO 3664 standard for viewing booths will greatly reduce the colour matching problems currently faced in the field. It is surprising to see the slow uptake and adoption of ISO 13655 by some instrument manufacturers. There are a number of devices and models in widespread use in North America, these devices are used every day by prepress, press and brand managers, yet many of these big-name products do not conform to the new standard launched in 2009 –it is now 2013! We appeal to the instrument manufacturers to upgrade or update their instrument portfolio to align with ISO 13655.

We are grateful to Martin Habekost and Peter Roehrig, Ryerson University for producing plates for the press run and running our jobs on press as part of GRA 634 Printing Processes course module. We thank Jim Luttrell, X-Rite for providing the iSisXL and Russell Doucette, KonicaMinolta for providing the FD-7 spectrophotometers for our research and testing. We thank Danny Rich, Sun Chemical; Ray Cheydleur, X-Rite; Bob Chung, RIT; and Mike Rodriguez for providing constructive review comments on this paper, these colleagues certainly helped improve the content of this paper. Abhay Sharma was partially supported in this project by NSERC Discovery Grant 341606-2008.

REFERENCES Cheydleur, R. (2011). The M Factor... What Does It mean? Successful Color Management of Papers with Optical Brightners. X-Rite, White Paper, L7510_M_Factor ColorWiki (n.d.). Konica Minolta FD-7. Retrieved May 3, from http://www.colorwiki.com/wiki/Konica_ Minolta_FD-7 Dalton, E. (Jan/Feb, 2010). ISO 3664:2009 - Why 5000K is not D50. IPA Bulletin, pp 22-25. ISO (2009). ISO 13655:2009 - Graphic technology —Spectral measurement and colorimetric computation for graphic arts images, ISO (2009). ISO 3664:2009 Graphic technology and photography - Viewing conditions.

Konika Minolta (n.d.). ISO 13655:2009 demystified: Why do we have M0, M1, M2 and M3? Retrieved April 11, 2013, from Konica Minolta Sensing Europe B.V.: http://www.konicaminolta.eu/en/measuringinstruments/learning-centre/colour-measurement/ colour/iso13655-demystified.html Meffre, W. (2013). The point about 2013 ISO 12647x Standards for CMYK Print and Proof Works. COLORSOURCE. Sharma, A. (2004). Understanding Color Management. Delmar Thomson, New York. Rochester Institute of Technology, http://cias.rit.edu/~gravure/tt/pdf/pc/ TT4_Natti01.pdf Techavichien, N. (2004). Analysis of Ink Dry Down For Hexachrome Inks For Sheetfed Offset Printing. Tobias Associates (2010). Reflection Densitometry. Tobias Associates, Inc.

Inspecting a set of Custom M&Ms Under a Microscope to Determine the Print Quality and Consistency

ABSTRACT This experiment involves inspecting a set of custom M&Ms under a microscope to determine the print quality and consistency of the candies. We hypothesized that approximately 20% of the tested candies will have print defects caused by various factors. The results showed that the opposite was true; we found that only 20% of the M&Ms had good quality prints, while the remaining 80% of the candies had at least one defect. This information would be useful when considering the use of custom printed M&Ms for advertising, as it would be more ideal for companies to utilize a promotional method that is guaranteed to produce a high quality product.

THE PRINT QUALITY OF M&MS Kaitlin Huynh Rebecca Yu

M&Ms were first introduced in 1941 to “American GIs serving in WWII” (Mars Inc., 2013). Since then, the product line has grown to include various flavours, but M&Ms are still best known for the printed M on each of the chocolates. The M&M printing process is described in one article of an online magazine:

1 bag of custom order M&Ms* Celestron 10X-‐150X digital microscope Peak Wide Stand Micro (60X)

“The M’s are applied to M&Ms in a process that Mars Inc. describes as “akin to offset printing.” Blank M&M's sit on a special conveyor belt that has a dimple for each candy to sit in, and roll through a machine where vegetable dye is transferred from a press to a rubber etch roller that gently prints the M on each piece” (Mental_floss, 2012). With that said, are these pieces of chocolate actually printed well, and is the print quality reliable? In this experiment, we analyzed and inspected a set of customized M&M’s to examine the consistency of the print quality. The chocolates were viewed under a microscope, and the candies were checked for any print defects. We also examined the legibility and cleanliness of the printed text, and finally analyzed the frequency of certain print defects found on the candies. Assuming that the colour of an M&M does not have any effect on its printing surface and the resulting print quality, this experiment should demonstrate the print quality and consistency of custom order M&Ms. We hypothesize that approximately 20% of the tested M&Ms will have print defects, with the remaining 80% having a good quality print.

Note: Candy customization is as follows: 3 colours (Aqua, Shimmer pearl, Light purple), 2 images (black square, clipart teddy bear), Arial font (text is variable).

20 18

Number of Occurrences

EQUIPMENT & MATERIALS Number of M&M’s

INTRODUCTION

14 12 10 8 6 4 2

PROCEDURE 01. Gather the equipment and materials needed. 02. Collect 10 M&Ms in various colours for each of the different designs, for a total of 40 M&Ms. 03. Plug in the Celestron microscope to a computer, and capture a picture of the printed area on one of the M&Ms. Repeat for each of the M&Ms; you should end up with 40 pictures. 04. Examine each of the M&Ms once more using the Peak microscope. Note any significant observations with the prints; pay special attention to text and the black square for any printing defects. 05. Using the pictures from Steps 3 and 4, and the observations from Step 6, produce a list of print defects found on the M&Ms. 06. Referring back to the pictures, create a tally chart for the occurrences of each print defect on the M&Ms. Make note of the most-‐ and least-‐ occurring defects. Note: The test specimens were stored and examined at room temperature.

None

One

Two

Three

Four

Number of Defects Found

Five

10 8 6 4 2 0

No Defects Smudging

Clarity Misalignment Spotting

Type of Defects

Cracking

Figure 1 – Graph of number of defects found on each M&M

Figure 2 – Graph of number of each type of defect found on M&Ms

Figure 1 plots the number of defects found on each of the tested M&Ms. Based on a total of 40 M&Ms, eight of them had no defects, 19 had only one type of defect, 12 had two defects, and one of the 40 had three defects on the same piece of candy.

Meanwhile, Figure 2 outlines the number of occurrences for each type of defect found on the M&Ms, including smudging, clarity, misalignment, spotting, and cracking. Of the 46 total defects found on the M&Ms, 10 of them had smudging, four had issues with clarity, 12 had misaligned images, another 12 had spotting, and the remaining eight defects were attributed to cracking. Refer to Appendix C for the classification of each defect.

ANALYSIS The most commonly occurring defects found on the M&Ms were spotting and misalignment, both with 12 occurrences, while the least frequent defect found was clarity, visible on only four of the tested M&Ms. The experiment yielded completely opposite results from the hypothesis; we predicted that 20% of the M&Mâ&#x20AC;&#x2122;s would have defective prints, but the results showed that only 20% of the specimens were free of defects. In manufacturing, the worst tolerable limit or quality is known as the acceptable quality limit of consumer goods, or AQL (Asia Inspection, n.d). In other words, AQL determines the tolerance level of defects. If the average number of defects on a product is within or below a certain threshold of the companyâ&#x20AC;&#x2122;s defective tolerance level, it is considered acceptable. With that said, although this is a tolerance level of defective products that are allowed to be shipped out, this does not mean that customers and end users who purchase the product will accept the number of defects on the product to be shipped to them. With regards to AQL, the values are listed in tables such as those in Appendix D. These AQL tables are a statistical tool to analyze the number of samples as well as the tolerance level for acceptability and the number of defects at which the entire batch of products will be rejected. This allows companies to hold a certain standard towards what is acceptable as the number of defects. It will also allow the companies to determine whether the batch of products is approved for shipping, or if they should be rejected. Different companies and industries have different standards towards AQL, and the charts would vary according to the company standards.

control for printed fabric or printed shirts. AQL has different inspection levels, and there are three inspection levels commonly used for general inspections (InTouch, n.d). There are known as GI, GII, and GIII, depending on the quantity that is randomly selected to be inspected. Knowing the quantity would mean knowing the category of the inspection level. For this experiment, based on the quantity inspected, it is considered to be at GI inspection level. This is because the sample size was small, and it was not a large percentage of the total M&Ms in one package. Looking at our results as compared to the AQL chart in Appendix C, the batchof 40 M&Ms that was sampled had a high amount of defects, and therefore should be rejected for shipment to customers. As seen in the chart, through 0.40% to 6.5% for a sample size of 50, the acceptable number of samples with defects is 1 to 8. The results of our experiment revealed that 32 out of 40 M&Ms had defects, which shows that the amount of quality control used was not very high in terms of the custom M&Ms. Depending on the standards of the industry, some companies may have a higher AQL for the specified quantity of production. While the experiment was conducted with as little bias as possible, there are several possible sources of error in the experiment. This includes human error when consulting the results, as the results are dependent on human perception of the M&Ms. When looking for defects on the M&Ms, the limitations of the test can affect the end results. As human perception is subjective, the results may change if another observer looked at the given data, and the results may have been recorded

differently. When proceeding with the observation of the M&Ms, the results can also vary because some people may perceive objects differently than others. Additionally, the classification of the types of defects is another important source of error. This is because the manufacturers of the M&M candies may have a different definition for each type of defect, and this could be one of the reasons that this batch of M&Ms was accepted in the quality control process. One final source of error is in the delivery and transportation of the candies. Because the M&Ms were delivered through postal service, the print quality of the candies may have been compromised throughout the delivery process. For this reason and for the purposes of the experiment, it is assumed that the candies were examined directly after they were printed.

CONCLUSION

Another more specific example is when a company is looking at custom order candies as a means of promoting their services and increasing brand recognition. If the company were to use their logo as the image on the candies, they would want to ensure that the logo is not distorted in any way. Even small differences from the original logo would affect the overall image of the company. Therefore, this experiment is useful for determining the consistency of the prints and the plausibility of advertising through custom printed candies, such as M&Ms.

ACKNOWLEDGEMENTS We would like to thank Abhay Sharma for providing the equipment to make this experiment possible, and the Graphic Communications Management program for funding the purchase of the test specimens.

The results of this experiment showed that only 20% of the tested specimens were free of defects. This was inconsistent with the original hypothesis of the experiment. Comparison with the industry Acceptable Quality Limit standards revealed that the print quality of the M&Ms were below average and thus, technically unacceptable. This type of information about standards is relevant in many industries. Setting and meeting standards is very important in manufacturing as a way of implementing quality control. With that said, AQL can be used in the printing industry as a means of quality control when producing a large quantity of print jobs. Other industries such as fashion can also use this in maintaining quality

APPENDIX APPENDIX A -MICROGRAPHS OF TEST SPECIMENS

4%%>:<9D$4$EE‐$$ 1987;C74%F6$;G$!>6!$6%>891>:6(

APPENDIX B – RAW DATA TABLES USED FOR GRAPHS Number of defects found per M&M:

Types of defects found on M&Ms

No. of Defects

No. of M&Ms

Types of Defects

None

No Defects

One

Smudging

Two

Clarity

Three

Misalignment

Four

Spotting

Five

Cracking

No. of Occurrences

APPENDIX C - CLASSIFICATION OF DEFECTS

Type of Defect Smudging

Clarity

Misalignment

This refers to any distortions of the image, resulting in an inconsistent image.

Spotting

Spotting is when patches of ink are missing from the image area, creating white dots.

Cracking

This refers to one or more cracks in the candy coating. Although this is not specifically a printing defect, it may affect the overall print quality of an M&M. These M&Ms are classified as being free of noticeable defects.

No defects

Definition Smudging refers to when the outer edges of the image areas bleed noticeably outwards onto non-‐ image areas. This is when there is a brown print of ink-‐like substance that was not part of the original image.

Example(s)

REFERENCES Asia Inspection. (2013). Quality Control Standards -‐ Acceptable Quality Limit (AQL). Retrieved April 2, 2013, from http://www.asiainspection.com/aql-‐ acceptable-‐ quality-‐limit InTouch. (n.d). Aql inspection | selecting an aql for product inspection. Retrieved April 11, 2013, from http://www.intouch-‐quality.com/ about/inspection-‐level-‐and-‐ selecting-‐an-‐ aql-‐acceptable-‐quality-‐level-‐for-‐product-‐ inspection/ Mars Inc. (2013). About M&M’s History. Retrieved February 14, 2013, from http://www.m-‐ms.com/us/ about/mmshistory/ Mental_Floss (2012). What do the Ms on M&Ms stand for, and how do they get them on there? Retrieved February 14, 2013, from http://mentalfloss.com/ article/30494/what-‐do-‐ms-‐mms-‐stand-‐and-‐ how-‐do-‐they-‐ get-‐them-‐there

CREDITS

GENERAL MEMBERS

1ST YEARS Maisha Chowdhury, Madalyn Fewster, Alexandra Garneau, Andy Ha , Farrah Hanif, Anthony Krystecki, Amber Lahav, Daniel Langsford, Christina Phan, Rodrigo Sanchez, Andi Talogo, Vinh Tran, Elizabeth Van Blyderveen

2ND YEARS Alyzeh Ahad, Jagvir Aujla, Shanfar Balsara, Marisa Blair, Brian Bostwick, Kenesha Campbell, Jason Chow, Sabihat Chowdhury , Alyssa Chung, Sheela Cruz, Marija Dikova, Mira El-Assi, Kristin Fong, Hannah Gonzalez, Jason Hang, Jennifer Huynh, Nicole Joczys, Deborah Kasongo , Jasmine Keo, Biata Kerr, Joyce Lee, Peter Mai, Aman Patel, Jim Phan, Rukshan Pulle, Adriana Sarmiento, Kim Sipkens, Rosa Sucilan, Anna Tang, Anjelica Tizon, Mirette Youssef, Ramage Zaki

3RD YEARS Sarah Aspler, Anna Avitsian, Jeff Bezbrozh, Megaera Bonsall, Arnold Chan, Jacqueline Chan, Angela Chau, Alex Chheun, Alina Esmatyar, Alana Ferrera, Emily House, Carmen Lam, Holly Lam, Scott Morgan, Amy Nguyen, Traci Phillips, Kurt Sagurit, Kelly Somers, Ellie Voutsinas, Keven Vu, Jessica Wong, Trevor Yeow

4TH YEARS Rhonda Atkinson, Martyna Benisz, Katherine Bermudo, Emma Blanchard, Basile Bonadeo, Kyle Chan, Philip Chan, Giustina Foschia, Melanie Green, Sara Hawary, Brian Hui, Ashley Lombardo, Anna Salazar-Tello, Alex TF Wong, Emily Wong, Mike Wu, Melissa Yung

EXECUTIVE TEAM

CREATIVE DIRECTOR Erin Luu This is my first year as Creative Director for RyeTAGA and I am sad to say that it is also my last. The year flew by so quickly and looking back on it now, it has been an amazing experience. I have learned a lot throughout my four years at Ryerson University as a Graphic Communications Management student. RyeTAGA gave me an opportunity to put all of that knowledge and skills to good use. With that said, it has been a very stressful year with many sleepless nights but well worth it for what we managed to accomplish. We did a great job with everything from fundraising to production. I would like to thank everyone on the team for all your hard work and I wish you all luck with your future careers. I hope we will be able to work together again. Hereâ&#x20AC;&#x2122;s to RyeTAGA bringing home another win this year!

COLOPHON