September 2019 - colorsync-users

Xrite i1Pro3+ review
by Scott Martin 15 Oct '19

15 Oct '19

Hey Colorsyncers. I’ve been testing the i1Pro3+ and i1iO3 table in a variety of situations for a good while now. I have a review of XRite’s i1Pro3+ you can find at: http://www.on-sight.com/xrite-i1pro3-review/ <https://www.on-sight.com/?p=71755&preview=true> I think you’ll find it addresses of some of questions I would imagine this group has: 1) When should I consider making polarized measurements? 2) What print quality differences can I expect to see with profiles made from polarized M3 measurements? 3) What print quality differences can I expect to see with the new transmissive profiling mode? 4) What's different about i1Profiler’s transmissive profiling mode that we haven’t seen before? 5) How consistent will print quality be relative to other measurement modes? 6) How consistent will print quality be if I made some profiles with the i1Pro3+, some with an iSis, and others with an i1Pro2? I’ve included some color gamut renderings and, IMO, some really important dialog about interpreting such renderings. I’ve also included some M2 and M3 profiles of the same glossy canvas media that you can download and compare for yourself. Let me know if you have any thoughts or questions about the i1Pro3+ and i1iO3 Table. Scott Martin www.on-sight.com Precise color science for printmaking professionals

5 12

Ryerson University Expanded Gamut Study
by Abhay Sharma 30 Sep '19

30 Sep '19

Dear ColorSync Community Ryerson University Expanded Gamut Study 2019 - Free 50 page report - tinyurl.com/ExpandedGamut <http://tinyurl.com/ExpandedGamut> I would like to bring to your attention a recent study we published - Ryerson University Expanded Gamut Study 2019 - in which we examined the color gamut of the Epson SureColor P9000 and also the HP Indigo 7900 and their ability to reproduce spot colors using CMYK OGV inks. We tested 6 color management solutions - Alwan, CGS ORIS, ColorLogic, GMG Color, Heidelberg, Kodak. Color management solutions are able to hit 90% of the spot color library to < 2.0 CIEDE2000. Download your copy of the report today. Abhay Sharma Ryerson University, Toronto

1 0

RE: Xrite i1Pro3+ review
by graxx＠videotron.ca 30 Sep '19

30 Sep '19

It's a good study, Martin. Thank you so much for sharing the link here. It's very interesting. I confess I never experienced with building backlit profiles for transparent media. I still have an old SpectrascanT table, here, but I never used it for that purpose. I was once asked to profile some transparency media and I was entertaining reading patches by placing the target on some kind of backlight, using my CS200 to make the readings. I'm sure, it would have worked well but I never got the "order" 😊 It would have take a hell of a long time to make all these readings but the idea of a large aperture system like Barbieri's and X-Rite's new i1Pro3 would seem more sensible, especially when having to make lots of backlit profiles. The two gamuts you present at the beginning of the article are interesting and show that the "absolute" size (volume) of the gamut is less important than its "relative" size. Kudo for digging into the use of a polarizing filter. Best / Roger Breton -----Original Message----- From: colorsync-users <colorsync-users-bounces+graxx=videotron.ca(a)lists.apple.com> On Behalf Of Scott Martin via colorsync-users Sent: Monday, September 30, 2019 9:14 AM To: 'colorsync-users?lists.apple.com' List <colorsync-users(a)lists.apple.com> Subject: Xrite i1Pro3+ review Hey Colorsyncers. I’ve been testing the i1Pro3+ and i1iO3 table in a variety of situations for a good while now. I have a review of XRite’s i1Pro3+ you can find at: http://www.on-sight.com/xrite-i1pro3-review/ <https://www.on-sight.com/?p=71755&preview=true> I think you’ll find it addresses of some of questions I would imagine this group has: 1) When should I consider making polarized measurements? 2) What print quality differences can I expect to see with profiles made from polarized M3 measurements? 3) What print quality differences can I expect to see with the new transmissive profiling mode? 4) What's different about i1Profiler’s transmissive profiling mode that we haven’t seen before? 5) How consistent will print quality be relative to other measurement modes? 6) How consistent will print quality be if I made some profiles with the i1Pro3+, some with an iSis, and others with an i1Pro2? I’ve included some color gamut renderings and, IMO, some really important dialog about interpreting such renderings. I’ve also included some M2 and M3 profiles of the same glossy canvas media that you can download and compare for yourself. Let me know if you have any thoughts or questions about the i1Pro3+ and i1iO3 Table. Scott Martin www.on-sight.com Precise color science for printmaking professionals _______________________________________________ Do not post admin requests to the list. They will be ignored. colorsync-users mailing list (colorsync-users(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/colorsync-users/graxx%40videotron.ca This email sent to graxx(a)videotron.ca

2 1

RE: Slight color differences
by graxx＠videotron.ca 27 Sep '19

27 Sep '19

Hi Marc, Thank you for taking the time to post here. I failed to mention that these were i1Profiler custom output profiles created out of IT8.7/4R measurements. The two IT8s were measured a few months apart, I confess, the 7890 in early June and the 7890 yesterday, both using the same Minolta FD-9 in M0 mode -- I'd be surprised the instrument would have drifted that much over such a short period of time? The media was the same for both printers yet there could be some slight variations coming off the media, contributing to the differences I see? Marc, I did exactly what you suggested. To assess the resultant conversion, at the end of my work, to do "quality control", I always run an IT8 on the printer through the workflow to test the quality of the simulation. The difference between the two printers "simulating" GRACoL2006_C1 is 0.66 dE2000 with a max of 2.37, 95% of patches are below 1.45, see the following screen capture from PatchTool Compare function: https://1drv.ms/u/s!AkD78CVR1NBqkaJgOTwFNP1Mpd1O7Q?e=P4lSsR I feel I should be "resting my case" but I'm curious to dig into the possible spectral differences across both ink sets? To that end, I assembled a summary analysis of both proofs "simulations" here, in a PDF: https://1drv.ms/b/s!AkD78CVR1NBqkaJhE5Sy0r9fFZ2UQg?e=DhA18K I am pleased to find similar spectra across both printer simulations, the largest difference, spectrally, coming of the CMYKRGB set is on 100M, as William noted. Yet, in terms of color difference, that difference in spectral shape only translate into 0.56 dE2000? Coming back to the visual difference between the prints, the largest and most questionable visual color difference (so far) comes from CMYK values around 66c 72m 62y 60k or CIE Lab 21 7 1, some kind of "dark gray violet", you see? 21 7 1 is the Lab value coming out the 7890 IT874. While on the 7800, according to Photoshop color picker, the CIE Lab value for this build is 19 8 1, a 1.74 quite "reasonable" dE2000 color difference. Yet, visually, across both prints, under a decent ISO-3664:2009 viewing booth, the color difference in that area of the image seems way bigger than 1.74? It's not a difference that immediately jumps to the eye when glancing at the two prints, side by side, overall, but upon closer inspection, these two areas don't seem as close as the rest of the imagery are, suggesting some possible difficult selective color edits. At a packaging printer I sometimes consult with, in their ink room, I've seen color differences in some dark blues samples, if I recall, that, colorimetrically, were rather small, well under a typical 2 dE2000 tolerance yet visually, appeared quite different. (Next time I'll go, I'll be sure to grab some samples to measure and document...) In the IT8.7/4R measurements, I found a patch that's closest in term of CMYK build to the offending "dark gray violet", patch ID 1238, 70c 70m 70y 80k. It is interesting to note that on the 7800, this is CIE Lab 16 1 1 while it is CIE Lab 15.2 1.9 1.38 on the 7890, a dE2000 difference of only 0.96. Of course, both proof pass G7 tolerances with "flying colors". I think I'll leave my analysis at that and move on to some other project. Best / Roger -----Original Message----- From: colorsync-users <colorsync-users-bounces+graxx=videotron.ca(a)lists.apple.com> On Behalf Of Marc Levine via colorsync-users Sent: Thursday, September 26, 2019 3:32 PM To: colorsync-users(a)lists.apple.com Subject: Re: Slight color differences Hi Roger, It's been so long since I posted here. I hope this is the right place. ; ) >From a scientific perspective, you might run a small test. You could transform an IT8 using the workflow you described, and then compare the resulting Labeez of the 2 converted images. That would tell how ho far off you were before you even printed. It was unclear from your post whether the profiles you mentioned are "canned profiles" or custom profiles that you made. In the case of canned profiles, there is a whole host of things that could be different and unknown between the 2 profiles. If there are both custom, following the same exact process, I would expect the electronic result to be very close and for the printed result to present a greater-but-reasonable difference compared to the electronic files. It sounds like the differences you are see are pushing the boundary of "reasonable". Here I would fall back to what others are saying about the output colorspace and the differences in the inks (assuming that both printers are also moving the material at the same speed, laying ink down at the same rate, and that the ink is drying at the same rate). Different inks will produce different behaviors in the colorspace that the profiling engine needs to accommodate (which is why proofing software like GMG and CGS tools iterate). These different behaviors might be handled slightly different but the profiling engine and would likely introduce some of the effect you are seeing. If you are feeling adventurous, you might try to add some "training data" to your profiles. That p rocess would look like: -Prepare a "special target" (an Lab file with measurable colors that are important to you). Grays are always good. -Make your regular-ol profiles -Transform your special chart into the colorspace of the printer (gives you a set of device values) -Bolt those device values together with the original IT chart values to make a new super-special profiling target -Print that target and build a new profile -Rinse and repeat for "printer #2" -Compare -Enjoy ; ) -Marc _______________________________________________ Do not post admin requests to the list. They will be ignored. colorsync-users mailing list (colorsync-users(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/colorsync-users/graxx%40videotron.ca This email sent to graxx(a)videotron.ca

1 0

Re: Slight color differences
by Marc Levine 26 Sep '19

26 Sep '19

Hi Roger, It's been so long since I posted here. I hope this is the right place. ; ) >From a scientific perspective, you might run a small test. You could transform an IT8 using the workflow you described, and then compare the resulting Labeez of the 2 converted images. That would tell how ho far off you were before you even printed. It was unclear from your post whether the profiles you mentioned are "canned profiles" or custom profiles that you made. In the case of canned profiles, there is a whole host of things that could be different and unknown between the 2 profiles. If there are both custom, following the same exact process, I would expect the electronic result to be very close and for the printed result to present a greater-but-reasonable difference compared to the electronic files. It sounds like the differences you are see are pushing the boundary of "reasonable". Here I would fall back to what others are saying about the output colorspace and the differences in the inks (assuming that both printers are also moving the material at the same speed, laying ink down at the same rate, and that the ink is drying at the same rate). Different inks will produce different behaviors in the colorspace that the profiling engine needs to accommodate (which is why proofing software like GMG and CGS tools iterate). These different behaviors might be handled slightly different but the profiling engine and would likely introduce some of the effect you are seeing. If you are feeling adventurous, you might try to add some "training data" to your profiles. That process would look like: -Prepare a "special target" (an Lab file with measurable colors that are important to you). Grays are always good. -Make your regular-ol profiles -Transform your special chart into the colorspace of the printer (gives you a set of device values) -Bolt those device values together with the original IT chart values to make a new super-special profiling target -Print that target and build a new profile -Rinse and repeat for "printer #2" -Compare -Enjoy ; ) -Marc

1 0

Slight color differences
by graxx＠videotron.ca 26 Sep '19

26 Sep '19

Hi everyone, I'm working on profiling two different Epson printers, a 7800 and a 7890. Even though both printers are using the same, exact substrate, is it to be expected that there would be some "slight (but visible) differences" between the two printers? (I know I can't show you the prints.) I convert from GRACol2006 to the 7800 and from GRACoL2006 to the 7890, using absolute colorimetry in both cases. Same conversion, same substrate, the only difference is the two printers. In your opinion, what could account for the slight differences I observe? Is it possible (that's my best take) that the difference in the generation of inks used on both of these printers could account for the slight differences I observe? Best / Roger Breton www.graxx.ca

4 3

RE: Slight color differences
by graxx＠videotron.ca 26 Sep '19

26 Sep '19

Claas, [Ahtung! Long message] You raised the most interesting question, "how far is it worth questioning the observable color differences"? If I was doing this for litho shops, I know I'd be glad to live with the difference I'm getting between the two printers, a difference nothing a printing press would not be able to match to a demanding client's expectations. Aber (But), since someone raised the question with me, and tested my humble expertise (indirectly), I'm tempted to hypothesize, as a good, amateur color scientist, that the difference that exists between the inks set, with the Magenta ink in particular, as William noted, *may* explain the small, lingering differences visible between the two outputs we see. It's possible. This would be an interesting analysis to make but I only have the IT874 for one of the two printers? I've seen my share of not so good color matching in my life and this is not bad color matching -- to my eyes (hard to relate to you in words the visual differences I observe between the prints?). I have to be willing to live with the small lingering differences anyone would get by using the same, basic ICC output profiling process. I find ludicrous the idea that creating the output profile across a number of profilers would make any differences, in AbsCol? And I'm not about to try it. "Trust the numbers" seems to be the bottom line. MfG / My two pesos / Rojelio (aka Roger Breton) -----Original Message----- From: Claas Bickeböller <lists(a)bickeboeller.name> Sent: Thursday, September 26, 2019 10:57 AM To: Roger Breton <graxx(a)videotron.ca>; 'colorsync-users?lists.apple.com' List <colorsync-users(a)lists.apple.com> Subject: Re: Slight color differences Hi Roger, there are some reasons that come to my mind: 1) The profiles do not represent the actual printer status anymore (or never did) - printer(s) changed after having done the measurements - measurement errors - inaccurate sampling of the individual printers’ non-linearities during characterization 2) Nozzle issues (which is basically another reason for 1) :-) 3) Out of gamut colours (unlikely with the combination you mention) Did you do a simulation of the result without printing? Is the dE-distribution similar for both printers? Just an idea to start digging into it. Best regards Claas > Am 26.09.2019 um 15:57 schrieb Roger Breton via colorsync-users <colorsync-users(a)lists.apple.com>: > > Hi everyone, > > > > I'm working on profiling two different Epson printers, a 7800 and a 7890. > > Even though both printers are using the same, exact substrate, is it > to be expected that there would be some "slight (but visible) > differences" between the two printers? (I know I can't show you the > prints.) > > > > I convert from GRACol2006 to the 7800 and from GRACoL2006 to the > 7890, using absolute colorimetry in both cases. Same conversion, same > substrate, the only difference is the two printers. > > > > In your opinion, what could account for the slight differences I > observe? Is it possible (that's my best take) that the difference in > the generation of inks used on both of these printers could account > for the slight differences I observe? > > > > Best / Roger Breton > > www.graxx.ca > > _______________________________________________ > Do not post admin requests to the list. They will be ignored. > colorsync-users mailing list (colorsync-users(a)lists.apple.com) > Help/Unsubscribe/Update your Subscription: > https://lists.apple.com/mailman/options/colorsync-users/lists%40bickeb > oeller.name > > This email sent to lists(a)bickeboeller.name

1 0

IS&T Webinar on Augmented Reality & Color/ CIC27 Early Registration Deadline
by Roberta Morehouse 19 Sep '19

19 Sep '19

The next in IS&T's series of complimentary webinars: Augmented Reality & Color Perception Challenges, offers amazing potential for mixing virtual content into the real world, but many tricky perceptual questions (displays, optics, software) remain. Dr. Michael Murdoch of RIT's Munsell Color Science Laboratory discusses these questions and more during his webinar. Color and Object Appearance in Augmented Reality Dr. Michael J. Murdoch, Rochester Institute of Technology Wednesday, September 25, 2019 10:00 - 11:00 AM EDT Register: http://bit.ly/Color_AR_Webinar Interested in more color science and imaging topics? Registration is still open for the IS&T Color and Imaging Conference (CIC27), but there are only 4 DAYS LEFT to save $100 off the full conference registration. Join us for a full week of color science and imaging across applications through keynotes from Facebook Reality Labs, Apple, and L'Oréal, invited talk on "Scientific Imaging for Cultural Heritage", 71 papers presented in 8 technical sessions, 15 Short Courses and 4 workshops and plenty of opportunities to engage in the many exciting topics related to color and imaging. IS&T Color and Imaging Conference (CIC27) October 21-25, 2019 Paris, France Learn more and register at http://bit.ly/CIC27 ADDITIONAL DISCOUNTS *Student rates - save more prior to the early registration deadline *Group rates for 5+ registrants from the same institution. Contact IS&T at registration(a)imaging.org for details. *Not a member? Include IS&T membership with your registration plus get a journal subscription - all for the same price as the non-member fee. *Can't attend the full week? Register at the one-day rate for the day that fits your schedule and interests. Connect with our community! LinkedIn IS&T CIC Group Twitter @ImagingOrg #CIC27

1 0

RE: New NEC MultiSync monitor
by graxx＠videotron.ca 13 Sep '19

13 Sep '19

Andrew is right, the NECs are no-brainers. I have mine, here, a PA271w, and it's a breeze. / Roger -----Original Message----- From: colorsync-users <colorsync-users-bounces+graxx=videotron.ca(a)lists.apple.com> On Behalf Of Dean Mackey via colorsync-users Sent: Thursday, September 12, 2019 4:28 PM To: ColorSync <colorsync-users(a)lists.apple.com> Subject: New NEC MultiSync monitor We need to replace the display at one of our graphic design stations; and are trying to narrow down the best choice. The rest of our stations are running NEC MultiSync PA242W displays; which we have been quite happy with. However, our IT department is telling us that those are no longer available. They claim that the closest thing to the PA242W are the following: NEC MultiSync EA271U ($849.00 USD) MultiSync PA271Q-BK ($1,299.00 USD) The PA271Q-BK appears to have much of the same features that I remember looking at when we purchased our existing 24” displays. Is this display a newer generation; just in a 27” version? There is a distinct price difference between the two; but I’m willing to fight for the PA271 if it is the best choice from a color-management standpoint. Sincerely, Dean Dean Mackey, Senior Graphic Designer Office of Public Affairs | Rice University, MS 95 | Allen Center, Suite 200 | 6100 S. Main St., Houston, TX 77005 | Office 713.348.6767 | Fax 713.348.6757 | dmack(a)rice.edu _______________________________________________ Do not post admin requests to the list. They will be ignored. colorsync-users mailing list (colorsync-users(a)lists.apple.com) Help/Unsubscribe/Update your Subscription: https://lists.apple.com/mailman/options/colorsync-users/graxx%40videotron.ca This email sent to graxx(a)videotron.ca

2 1

New NEC MultiSync monitor
by Dean Mackey 12 Sep '19

12 Sep '19

We need to replace the display at one of our graphic design stations; and are trying to narrow down the best choice. The rest of our stations are running NEC MultiSync PA242W displays; which we have been quite happy with. However, our IT department is telling us that those are no longer available. They claim that the closest thing to the PA242W are the following: NEC MultiSync EA271U ($849.00 USD) MultiSync PA271Q-BK ($1,299.00 USD) The PA271Q-BK appears to have much of the same features that I remember looking at when we purchased our existing 24” displays. Is this display a newer generation; just in a 27” version? There is a distinct price difference between the two; but I’m willing to fight for the PA271 if it is the best choice from a color-management standpoint. Sincerely, Dean Dean Mackey, Senior Graphic Designer Office of Public Affairs | Rice University, MS 95 | Allen Center, Suite 200 | 6100 S. Main St., Houston, TX 77005 | Office 713.348.6767 | Fax 713.348.6757 | dmack(a)rice.edu

2 1