March 2016 - colorsync-users

Spectral sensivities
by Roger Breton 15 Mar '16

15 Mar '16

Is there a (simple) way to get at a particular camera sensors spectral sensitivities? / Roger -----Original Message----- From: colorsync-users-bounces+graxx=videotron.ca(a)lists.apple.com [mailto:colorsync-users-bounces+graxx=videotron.ca@lists.apple.com] On Behalf Of Ben Goren Sent: 15 mars 2016 12:07 To: ColorSync <ColorSync-Users(a)lists.apple.com> Subject: Re: Colorsync-users Digest, Vol 13, Issue 76 On Mar 14, 2016, at 6:08 PM, Graeme Gill <graeme2(a)argyllcms.com> wrote: > Certainly the issues that Anders Torger has tripped across in > developing his own DNG profilng solution (DCamProf) reenforce the view > that there is some odd hard coded behavior in DNG. Anders has independently developed a very similar approach to mine, starting with a spectral model of the camera sensor. If you want the least-worst DNG profile you're going to get, you need to use DCamProf. And it's an open source labor of love with not much attention paid to user friendliness -- good results if you're willing to invest in the learning curve. But even at its best...you're still stuck with the bizarre illuminant nonsense that does this automatically-estimated blending between A and D50 with no meaningful user control nor input. Again, horses for courses. I can't imagine a _Sports_Illustrated_ or _National_Geographic_ photographer having any non-idle interest in my workflow. Lightroom and DNG is a great solution for those photographers -- and, indeed, for the overwhelming majority of photographers. You'd be hard pressed to better, even theoretically -- and anything you could do better in theory would have been impractical or impossible with the hardware on the market at the time it was developed. But just because Adobe products are excellent for most photographers doesn't mean that they're excellent for everybody. And, indeed, when it comes to colorimetric reproduction, they're no more suited to the job than Excel is to being a distributed relational database server. Cheers, b&

4 3

Re: Colorsync-users Digest, Vol 13, Issue 76
by Ben Goren 15 Mar '16

15 Mar '16

On Mar 15, 2016, at 9:34 AM, Denis Potapenko <denis.potapenko(a)Mastech.com> wrote: >> ...you're again limited (at least in practice, if not theory) to 24-patch original ColorChecker targets with no ability to supply custom measurements. > > please finally read the dcamprof manual !!! DCamProf is not an Adobe product. And it's only about a year old. Yes, Anders is to be highly commended for his work, and I'd unhesitatingly recommend DCamProf for those stuck using DNG profiles in their workflow. But I wouldn't for a moment recommend using Camera Raw or Lightroom for a colorimetric workflow, and I'd strongly caution against DNG-based workflows in general -- if for no other reason than that everything else (monitor, printer, proofing, what-not) is ICC-based rather than DNG-based. DNG doesn't give you anything over ICC for colorimetric reproduction, so why mess with it at all? b&

1 0

RE: Reading textile samples
by Roger Breton 11 Mar '16

11 Mar '16

Max, > Computational M1 measurements (like those of the i1 Pro2) may not be appropriate. Is this something you personally witnessed? Anything in published form that you could share? Couldn't resist asking... / Roger

1 0

re: Reading textile samples
by Garth Fletcher 11 Mar '16

11 Mar '16

Lorenzo Ridolfi wrote: > I need to profile a textile media for a large format printer II have no specific experience with the instruments you mention, but I have worked with textiles. What differentiates textiles from most other media is their strong surface texture. The incident angle of the illumination can have a strong effect on the reading. For example, 45° illumination creates shadow areas behind each raised fiber which will be "seen" by a 0° sensor. Diffuse illumination may be useful to alleviate that problem. The above assumes that you are printing on fabric woven with a single weave - the usual case for printing. Dobby or Jacquard woven fabrics, which contain many different weaves, create a much larger set of problems... -- Garth Fletcher

7 7

Duotone workflow problem (Warning: includes printing with color management off)
by Richard D. Warner 10 Mar '16

10 Mar '16

Hello all, A customer came in recently with a duotone JPEG image to be printed. The image opened in Photoshop CS2 on one of our Mac workstations without any profile mismatch complaint. Our PS Color Settings are U.S. Prepress Defaults with its AdobeRGB (1998) editing space, and likewise our Onyx RIP is configured to "expect" arriving files to be in the Adobe RGB (1998) color space, although others, such a sRGB, ProPhoto RGB, etc. can be chosen if needed. Without regard to the image's duotone mode, I selected the desired media in Photoshop's print dialog box and printed it from the Mac workstation via Ethernet to our Onyx Productionhouse RIP. Where it arrived in the RIP queue it was composed with other jobs and printed. After the file was printed we could see that something had obviously gone wrong, as evidenced by scum dots in the image's white border, and less obviously in the image itself. But it gets worse... All subsequent images printed from the aforementioned workstation displayed the same problem. The RIP's Preflight displayed the values of the scum dots in white borders as: 2.0% Cyan, 3.1% Magenta, and 2.4% Yellow. I finally discovered that the duotone image had changed Photoshop's "Output" from "No Color Management" to "Let Printer Determine Colors", a setting that proved to be "sticky", as it was applied to all subsequent prints. The change wasn't visible, as it was hidden under the drop-down "Output" setting. In my defense, I can't recall ever printing a duotone image before this unhappy event. Has anyone else had this occur? Thanks, Richard -- Richard D. Warner Warner Graphics, Inc. 22 Washington Street Camden, Maine 04843 Tel (207) 236-2065 Tel (800) 875-2422 Fax (207) 230-2422

3 5

Re: Reading textile samples
by Antonios Granis 10 Mar '16

10 Mar '16

Hi all, just to share my experience. I am using a Konica Minolta CM2600d, mounted on a Colorscout A3+. With this combination i get to have remarkable repeatability and very good color accuracy. I ended up with this solution because i could not achieve proper Pantone hues reproduction on textile, using an i1, or a Spectrolino. I invested on this solution for the textile market and i can tell for sure it was more than worth it. Regards

5 6

Re: Colorsync-users Digest, Vol 13, Issue 72
by Mike Strickler 10 Mar '16

10 Mar '16

It sounds as if this client has wandered quite far into strange woods. Are they really converting process CMYK into 4 spot channels? Do they not know the difference? Of course it is possible to design with CMYK as though they are spot colors, though as you point out, tints and overprints, which are well characterized in process color via the active CMYK profile, suddenly loose their standardized interpretation and reproduce ability. We have trouble enough predicting and proofing real spot color tints and overprints; no need to induce this problem with process CMYK. So this would seem to be part of the problem here. Perhaps in addition to telling your client that what they are asking for (special "Pantone" process inks) doesn't exist, you might say that their insistence in renaming their process colors is actually making it likely that they will have inaccurate proofs and on-screen images. Mike > Message: 1 > Date: Wed, 09 Mar 2016 15:47:55 -0500 > From: Todd Shirley <todds(a)urbanstudionyc.com> > To: "'colorsync-users?lists.apple.com' List" > <colorsync-users(a)lists.apple.com> > Subject: Re: "Pantone Process" Inks > Message-ID: <7825B208-3855-4571-AD8F-EFB2397D3223(a)urbanstudionyc.com> > Content-Type: text/plain; charset=utf-8 > > Thanks for all the feedback so far. > > After doing some testing with our RIP, we think we know where this misconception may have started. If you take a GRACol CMYK image and relabel each channel in Photoshop as PANTONE Cyan, Pantone Magenta, etc., the colors proof richer and more saturated because the RIP doesn't know anything about how the colors should overprint, it's just doing it's best guess based on the solid inks and and dot gain curve that may (or may not) be applied. I used the Pantone Process colors and got a richer more saturated proof. This is of course not color-managed in any way, but it looks richer. > > This channel substitution is surprisingly easy to do in Photoshop by just selecting the “PANTONE solid coated” library. The first four choices are PANTONE Process Yellow C, PANTONE Process Magenta C, PANTONE Process Cyan C & PANTONE Process Black C. To a novice, this looks like legitimate inks that could be chosen to be used in an image. And it seems that this is what the other vendor may be doing - when they sub in “pantone Process” inks, they get a richer proof, ergo, these are “richer inks”. > > This has been going on with this client, vendor & printers for years now. We are in fact required to label all our proofs with swatches that say PANTONE Process Yellow C, PANTONE Process Magenta C, etc. so that the printers know to use the “pantone process” inks. It is happening with a job that we are working on right now. It appears to be an emperor has no clothes situation, but there doesn’t seem to be any great advantage to us pointing this out to any of the parties. It is unclear if this is active deception or if everyone truly believes these are actually different inks. > > The reason I’m writing all this is that I need to be 100% clear that “pantone process” inks are not a real thing. As has been pointed out, Pantone doesn’t actually make ink, and this is what they said when I inquired about this 2 years ago. There is no formulation of process inks that is different, correct? Does anyone have a similar story or have any further insight on this situation? > > Thanks again! >

2 1

Turn off color management!
by Ben Goren 09 Mar '16

09 Mar '16

I’ve got a Canon iPF8100. In the past, I’ve done all my printing via Canon’s plugin for Photoshop. For lots and lots of reasons, that’s no longer an option. I’ve switched to Affinity Photo, the version of the plugin for Canon’s Digital Picture Pro won’t work with images over 16” wide (at 600 ppi), an old laptop that I could run Photoshop on doesn’t have nearly enough RAM to work with large images, and on and on and on. In short, I need to be able to print from within arbitrary applications. The plugin made it trivial to turn off all color management. But, for the life of me, I can’t seem to get El Capitan to stop “helpfully” distorting print colors beyond all recognition. All I want is for the print driver to send the data unmodified to the printer. Is that really too much to ask for? Can anybody suggest a way to do it? (I do all my color management manually with Argyll. To build profiles I obviously need to turn off OS-level color management. And it also needs to be off for final printing, as well, since I’m using Argyll again to do all the transformations. Yes, most people need their hands held. No, I’m not most people, and I need my hands to not be tied.) Thanks, b&

18 120

Re: "Pantone Process" Inks
by Mike Strickler 09 Mar '16

09 Mar '16

It's hard to say where customers get these ideas, but we haven't as an industry been very successful at communicating about color to others. One could even say that it is not helpful that the current Pantone libraries still include a "Pantone Black" and a "Pantone Yellow" that are not ISO 12647 colors at all (the yellow being especially different). But perhaps that's no worse than changing the Lab specs if not the composition of base mixing colors (e.g., Reflex Blue, Rubine Red) and colors mixed from them while leaving their names unchanged. I recently became aware of a very large discrepancy between the Lab values in the database for Pantone solid colors of a large ink company and those in any of the last three versions of the Pantone Solid libraries. Where did the ink company get those values? Nobody seems to remember. This is the world we inhabit. Mike Strickler MSP Graphic Services > > Message: 2 > Date: Mon, 07 Mar 2016 15:18:05 -0500 > From: Michael Eddington <meddington38(a)gmail.com> > To: Todd Shirley <todds(a)urbanstudionyc.com> > Cc: "'colorsync-users?lists.apple.com' List" > <colorsync-users(a)lists.apple.com> > Subject: Re: "Pantone Process" Inks > Message-ID: > <CAJk2YzAcL1PvJ5VqPmHH5vDLuwcutapLe-eWswgYq4ksrW-FMA(a)mail.gmail.com> > Content-Type: text/plain; charset=UTF-8 > > Your interpretation is correct…Pantone CMYK primaries are governed under > ISO 2846-1. > > http://www.pantone.com/help/?t=CMYK-Primaries-not-found-in-PANTONE-FORMULA-… >

7 8

Reading textile samples
by Lorenzo Ridolfi 09 Mar '16

09 Mar '16

Hi, I need to profile a textile media for a large format printer. Unfortunately, I don't have money for a Barbieri LFP, so my options are an XRite IO2 table, an Eye-one Isis and a Colormunki. Using the Isis does not seem to be a good choice, I don't think I'll be able to feed the textile into the instrument properly. The Colormunki 6mm aperture is appealing but reading manually thousands of patches is a daunting task. The IO2 apparently is the best option. Surely I need to use larger patches to deal with the larger dot from the large format printer. Does anyone have experience in reading textile samples with these equipments? Best Regards, Lorenzo

1 0