By CHRISTINE WHEATLEY
February 2005
Abstract
In the gravure industry, having a proofing press that is used to simulate the visual characteristics of a finished product has become common practice. It is often difficult to achieve accurate color on the proof press since it may use different substrates, inks, and run different speeds than the final press. The proofing press also introduces delays in the cylinder delivery process, typically adding half a day to the production cycle. One way to reduce the necessity of using the proofing press is to use color management to provide off-line, color-accurate proofs.
This research investigates some of the issues relating to proof-to-press color management for the gravure process. This research was completed on one of the current shortcomings in color management, specifically focusing on the gravure process. A set of four experiments was completed to obtain the information required to appropriately analyze this problem. Each experiment is designed to build on the information acquired in the ones previous in order to gather all the information necessary to analyze the problem of color differences between proof and press sheets.
In Experiment #1 – Reference Printing Conditions, data was collected on all of the primary printing specifications – SWOP, GRACoL, etc. The ISO standard values for the gravure process as detailed in ISO 12647-4 were also examined. The sample gravure press sheet from the WMU Printing Pilot Plant was then compared to these specifications.
In Experiment #2 – Gamut Comparison between the Fuji PictroProof and the Cerutti Press, an ICC profile was created for the gravure press at Western Michigan University and as well as a second profile for the Fuji PictroProof digital proofer. The difference in gamut size between the two devices was analyzed to determine if the PictroProof is suitable to be used as a proofer for the Cerutti press.
In Experiment #3 – Profile Comparison – Proof-to-Press Error, the error in the process based on an IT8.7/3 target was examined. The experiment was completed to determine how accurately the PictroProof could be used to proof our gravure sheet, the error shown in terms of Delta E.
In Experiment #4 – A Comparison of images printed on the Fuji Pictro and the images printed on the Cerutti gravure press, images from the press sheet are printed on the digital proofer, and the results were evaluated.
Introduction
Color variability
Color variability is a large dilemma in the graphic arts industry. The problem lies in the numerous variables that are required to consistently and accurately produce the same color at different times, in different locations, using different material, and by using different printing processes. To help ease the complexity of this problem, some form of process control needs to be used throughout each variable of the workflow.
Color management is the answer to this problem. The use of a color-managed workflow is a way to control color and view it in a consistently accurate manner throughout all devices and print. Color management is implemented through the use of profiles created specifically for each individual device within a workflow.
ICC profiles in the gravure industry
The International Color Consortium, the leading association in color management, has approved color profiles to be used to color manage a workflow. A profile is a set of data collected from a target chart printed on a proofer or press containing numerous color swatches that describes the color behavior of the device.1 The L*a*b* values of these swatches are measured, recorded, and then inputted into profiling software to create a profile for the specific device. When this profile is applied to the image, the result is a device-independent workspace allowing devices to communicate color in a controlled environment. This is done by creating a path through which images are converted into the L*a*b* values that will correctly portray the color of the image on the device or printed sheet.
When profiles are assigned to each device in a workflow, color can be portrayed in a consistently unified manner. Press profiles are created in order to portray all images within the restrictions of the final press’ gamut. A gamut is defined as the range of colors a device can produce.2 This ensures all prepress files are created and viewed within the gamut of the press. Monitors using the RGB color space can portray many more colors than the press’ CMYK gamut can produce. Applying a press profile to the prepress image allows the image to be viewed exactly the same as it will print. This is very important for color matching and adjusting before the press run.
Creating a gravure press profile
The process of creating a profile for a gravure press involves what is called “fingerprinting.” Fingerprinting a press is the process of optimizing the use of specific paper, ink, and gamma curve combinations. This optimization results in the printing of a color target that best fits the capabilities of that press, ink, paper, and gamma curve combination. This target will be used as the basis of the press profile.
There are many color targets, but the target used here is an IT8.7/3. A target is a compilation of a number of color swatches whose L*a*b* values are measured and recorded using a spectrophotometer and profilemaking software. Once the target is measured, the software compares the information of the printed target to the information of the target used as a standard. The result is a profile specifically rendered for that press, ink, paper, and gamma curve combination. This profile can then be applied to files in order to view the images in the color gamut of the press. (This is only completely accurate in a fully color-managed workflow.)
Another important benefit occurs when applying this profile to an image before printing a proof. This ideally creates a proof that has accurately portrayed the colors as they can be seen on the press sheet. The advantage is the ability to use this proof to match the color of the press sheets during the press run.
Experiment #1 Reference Printing Conditions
Introduction
Creating press profiles in the gravure industry is very time consuming and costly due to the high cost of engraved cylinders verses lithography and flexography plates. It is just not feasible for most gravure printers to fingerprint and create an individualized profile for every press, paper, ink, and gamma curve combination. To help relieve this burden, reference printing conditions have been created by a team of specialists representing various aspects of the industry.2 Reference printing conditions are lists of printing specifications based on ink densities. This list can be used to find a color profile that is most relevant to the particular press, ink, paper, and gamma curve combination. Reference Printing Conditions are an alternative to individually profiling a gravure press. Although the profiles chosen from these specifications will be close to that of a particular press and variable combination, the best possible results would be obtained by fingerprinting and individually profiling the press.
Methodology
Figure 1.1 is an example of a reference printing condition chart. Reference printing conditions contain measurements from a sample previously used as a standard. These measurements include the L*a*b* values of the paper and the ink solids, in addition to the density of the ink solids. Printers use these measurements to find a standard profile used for their particular press, ink, and gamma curve combination. A printer will measure
the L*a*b* values and densities of the ink and paper that they are using. If there is a condition whose measurements closely resemble the measurements from the sample, the printer can assume the corresponding profile would be a good match for their variable combination. In this experiment, two standards were chosen to compare to a sample that was printed on the Cerutti gravure press at the Western Michigan University Printing Pilot.
Figure 1.1 – Printing reference conditions for GRACoL, SWOP, ISO, and the Cerutti gravure press at Western Michigan University
Results
Figure 1.1 is the printing reference conditions containing the measurements for a GRACoL TR004 and a SWOP TR001 profile. Figure 1.2 is a graph showing the comparison among the various measurements of the three conditions. When looking at reference printing conditions, the things to look for are whether the paper type was typical of any specification and if the solid ink densities on the press sheet are near any of the specifications. If the press sheet is within any of the known specifications, then the generic SWOP or GRACoL profiles as representative of our process can be used. If this is the case, it is not required to custom profile the press, but, instead, the readily available generic profiles can be used. From these graphs we can conclude the best profile for the Cerutti press, paper, and ink combination used in this experiment would be the SWOP TR001. This can be seen in the fact these are the closest in relationship on all four graphs comparing the different paper and printing specifications.
Experiment #2 Gamut Comparison Between the Fuji Pictro and the Cerutti Press
Introduction proofer and press gamuts
A large number of commercial printers are now using digital proofs created on a special digital proofer device to match color during a press run. Problems occur when the proofer has a color gamut not including the entire gamut of the press. The result is the proofer cannot print the entire range of colors the press can print. When using proofs to accurately match color on press, it is very important the proofer has a larger gamut than
the press in all color aspects so the full range of colors can be produced on the proof.
Methodology
Profiles were created for the Cerutti gravure press and the Fuji PictroProofer in order to compare the gamuts of the press and the proofer. The following steps were taken to create these profiles:
- IT8.7/3 targets were printed on both the gravure press and the PictroProofer.
- These targets were measured using a Gretag Macbeth SpectroScan which was operated by Measuring Tool found in Gretag Macbeth’s ProfileMaker 5.0.
- After the measurement data was obtained, a profile was created in Gretag Macbeth’s ProfileMaker.
- The IT8.7/3 CMYK Ref.txt was selected as the Reference Data, and the data from the Measuring Tool was imported as the Measurement Data (see figure 2.1).
- The profile size was changed from Default to Large, and the Perceptual Rendering Intent and Gamut Mapping were not changed from the default values of Neutral Gray and LOGO Classic respectively (see also figure 2.1).
- In the separation area of the ProfileMaker, the Predefined setting was set at Dye Sublimation (nearest estimate to photographic) for the Pictro profile and Gravure for the Cerutti profile (see figure 2.2).
- The profile was then created by Profile-Maker.
- After both profiles were created, ColorThink 2.1.2 for OSX was used to obtain pictorial views of the color gamut for each device (Figures 2.3 – 2.5).
Figure 2.1 shows the settings used to create the profiles using Gretag Macbeth’s ProfileMaker 5.0
Figure 2.2 – This is the separations portion of ProfileMaker. It is important to set the predefined setting to the correct printing process of the press for which the profile is being created.
Figure 2.3 – A three-dimensional graph depicting both the Cerutti press and Fuji PictroProofer gamuts. The Cerutti gamut is the innermost solid area and the Pictro gamut is the semi-transparent outer area.
Figure 2.4 – The outermost lined space is the gamut of the Fuji Pictro, and the inner solid gamut is that of the Cerutti press. This two-dimensional view portrays the proofer gamut as larger than the press gamut in all color aspects.
Figure 2.5 – This three-dimensional view portrays the proofer gamut in red and the press gamut in full color. It can be seen that the press gamut is larger in the bottom (darker) region of the color gamut
Results
The gamuts of the Cerutti press and the Pictro proofer could be compared using the profiles created for each device. Figure 2.3 is a threedimensional graph of these gamuts, the Cerutti gamut being the solid area and the Pictro gamut being the semi-transparent image. Figure 2.4 shows the gamuts in a two-dimensional fashion, the proofer gamut being the lined item and the press gamut being the solid center gamut. From this picture, it seems the proofer gamut is larger than the press gamut in all color areas; however, this is not the case. Figure 2.5 is the same three-dimensional graph as Figure 2.3, but viewed from the bottom right corner. The figure has also been altered to portray the Cerutti gamut in full color and the Pictro gamut in red in order to see where the Cerutti gamut is larger than the Pictro gamut. From this graph, it is seen that the Cerutti gamut is larger at the bottom of the gamut area. Because of this location, it can be concluded the Cerutti press can print a wider range of dark colors than the Pictro proofer. This situation would result in color matching difficulties when comparing dark color ranges on the proof sheet printed on the Pictro and the press sheet from the Cerutti.
Conclusions
This experiment illustrates the importance of viewing and comparing both proofer and press gamuts before making proofs to be color matched on press. Gamuts should be compared in both a two-dimensional and three-dimensional form so all areas of the gamuts can be seen. This is the only way to accurately determine if there is any area where the press gamut is larger than the proofer gamut.
Experiment #3 Profile Comparison – Proof-to-Press Error
Methodology
To analyze the difference between a profiled and unprofiled workflow, the comparison of three profiles were needed. These were the Cerutti press profile, the Fuji Pictro profile (both created in Experiment #2), and the profile of a target with both the press and proofer profiles applied to it before printing it on the Fuji Pictro. The following steps were taken to print this new target.
- The IT8.7/3 target was opened using Adobe Photoshop CS.
- The Cerutti press profile was applied to the target using the following steps:
- Image > Mode > Assign Profile.
- Selected the Cerutti profile from the list of profiles.
- The Fuji Pictro profile was also assigned to the same target using the following steps:
- Image >Mode > Convert Profile.
- Selected the Fuji Pictro Profile from the list of profiles.
- The target was printed on the Fuji Pictro.
- The image was printed in a TIFF format with no compression and no embedded profiles.
- The target was measured and a profile was created.
In order to compare the color accuracy of a profiled workflow versus one not profiled, Gretag Macbeth’s Measuring Tool (part of ProfileMaker) was used to compare two sets of profiles. The first set compares the profile of the Cerutti press to that of the PictroProofer. The results will show the difference (Delta E) between the profiles and will also portray the areas where there would be differences or problems if an unprofiled proof from the Fuji PictroProofer were compared to the same image printed on the Cerutti gravure press. The second set of profiles compares the Cerutti profile to the profile of the proof created on the Pictro, with the Pictro simulating the Cerutti. The results of this comparison will show the Delta E and problematic areas in a profiled proof when compared to the same press sheet. The results of these experiments will show the difference between a profiled (color-managed) and unprofiled proof-to-press workflow. The effect of color management in this process can be determined by the result of these experiments.
Results
Figure 3.1 shows the results from the comparison of the Cerutti profile verses the Pictro profile. This case represents a non-color-managed workflow. In this figure, the IT8.7/3 target is portrayed with yellow boxes around the worst 10 percent of the swatches that did match both profiles. It can be assumed these would be where color problems would occur between the proof sheet and the press sheet. These problematic areas seem to be scattered randomly across the sheet, which signifies the problems are not completely due to an overlap in color gamut. The final result is a Delta E of 13.69 with a standard deviation of 5.79.
Figure 3.1 – This is a snapshot of the results from the comparison of the Cerutti profile and the Fuji Pictro Profile. The areas of the IT8.7/3 target that are outlined in yellow are the areas where color differences would occur in the two devices. This is representative of a non-colormanaged proof-to-press workflow.
Figure 3.2 shows the results from the comparison of the Cerutti profile versus the profile created from the profile that was the Pictro simulating the Cerutti. Notice the majority of the problematic areas are in the darker colors. This is due to the overlap of the two gamuts in the darker region described in experiment #2. The final results from this comparison are a Delta E of 3.80 with a standard deviation of 2.06.
Figure 3.2 – This is a snapshot of the results from the comparison of the Cerutti profile and the target with both the Cerutti and Fuji Pictro profiles applied to it. This is representative of a color-managed proof-to-press workflow.
Conclusion
By comparing these results, it is clearly shown that the color-managed workflow had significantly fewer problem areas as well as a smaller Delta E of 3.80 as compared to the 13.69 of the non-color-managed workflow. This experiment shows the extreme benefit of using a color-managed workflow.
Experiment #4 A Comparison of Images Printed on the FujiPictro and the Images Printed on the Cerutti Gravure Press
Methodology
In order to visually compare the color of a profiled image printed on the Pictro proofer verses the same image printed on the Cerutti press, two duplicate images were opened in Adobe Photoshop. The Cerutti profile (created in Experiment #2) was applied to the first image, which altered the colors to look as they would if they had been printed on the Cerutti press. A second image is shown where the Cerutti is being simulated on the PictroProof. This altered the colors to look as they would if the image had been printed as a proof on the PictroProofer. The color of these images was then visually compared.
Results
Figure 4.1 is the image with the Cerutti profile applied to it, and figure 4.2 has the Pictro and Cerutti combination image applied to it. The biggest color difference in these two images can be noticed in the flesh tones of the female on the left of the picture. The image from the Cerutti has much more natural-looking flesh tones, and the proof has very red-looking tones, which looks less realistic. The other place where there are noticeable differences is on the blue portion of the press in the middle of the image. The image printed on the Cerutti has more dark detail in the blue tone than the proof image. This can be explained by the gamut differences in the dark color regions found in Experiment #2. Even though these images were created in an entirely color-managed proof-to-press workflow there are still noticeable color variations between the proof and the press sheet.
Figures 4.1 and 4.2 – The top image (4.1) has the Cerutti profile applied to it and is representative of the color of the same image printed on the Cerutti press. The bottom image (4.2) has the profile created from the target with both profiles applied to it. This is representative of a proof printed on the Fuji Pictro with both the Cerutti and Pictro profiles applied it.
Final Concluding Remarks
Experiment #1 explained reference printing conditions, how to use them, and the importance of printing to a standard in the color-management process. In Experiment #2, two profiles were created, one for a Fuji PictroProof and a second for the Cerutti gravure press at Western Michigan University. These profiles were then used to compare the gamuts of both devices, and it was determined that the Fuji PictroProof is not the best option for a digital proof for the Cerutti press because its color gamut does not fully encompass the gamut of the Cerutti. Experiment #3 was designed to show the difference in the color accuracy of a colormanaged proof-to-press workflow versus an unprofiled workflow. The difference in this situation was a Delta E of 9.89, signifying large color differences between the two processes. Experiment #4 showed the visual difference between a Cerutti simulated proof from the Fuji PictroProof and an image printed on the Cerutti press. The results show that even though the process was accurately color managed, there were still noticeable variances in color between the two images.
The results of these experiments reveal color variation between proof sheets and press sheets is a problem even in a correctly color-managed proof-to-press workflow. Although this research does not include any solutions, the following is a recommendation on how to help this problem. When working in a fully colormanaged proof-to-press workflow, create soft proofs by applying the press profile and the press and proofer profile (as done in Experiment #4) on the computer before printing a proof. Color variances can be seen in this stage of the process and can be manually fixed by adjusting the color in Adobe Photoshop before sending the image to the proofer. Although this is an element of additional time in the overall process, it will make color matching on press easier and more accurate.
References
- Gravure Education Foundation and Gravure Association of America. (2003).Gravure Process and Technology. Gravure Education Foundation and Gravure Association of America
- Sharma, Abhay. (2004).Understanding Color Management. Thomson Delmar Learning.
- Bridg’s. (2005). “General Printing Guidelines.” Bridg’s.
- International Organization for Standardization. (2004). “ISO 12647-1.” International Organization for Standardization.