Open EPS with Photoshop: Mode Grayscale Globally Dotgain 20% or Gamma=2.2 (doesn't matter for pure black on white) Choose resolution sufficiently high. I would start with about 600ppi. Uncheck Anti-Aliasing (to avoid edge blurr). Open. Downsample if required (no interpolation, "nearest neighbour"). Save as TIFF (still Grayscale) Paste TIFF into K-channel of new empty CMYK file. Save as TIFF (now CMYK, K-only). Best regards --Gernot Hoffmann ... View more

Rob, that's correct – just the nearest Pantone swatch is found. The swatch doesn't "match" the RGB color. http://docs-hoffmann.de/swatch16032005.pdf This doc contains many swatches with manufacturer's Names or Numbers and then sRGB, aRGB and Lab-values in a parallel representation. Each set of the two RGBs and Lab is consistent andmatches the appearance on a calibrated monitor (if in-gamut). The relation to manufacturer's swatches is nowadays outdated, but the catalog may help the OP to become familiar with the stuff, including the out-of-gamut problem. Best regards --Gernot Hoffmann Edited ... View more

As explained in the other thread I had done similar tests, which resulted as well in appearance errors in Illu. Working space in PhS is sRGB, working space in Illu is sRGB and the file is in sRGB with profile (TIFF) or without (PNG-8, PNG-24). Therefore no rendering intent is involved. Furtheron, in my opinion, for RGB-RGB-conversions rendering intents generally are not applied, as tested by PhS. Best regards --Gernot Hoffmann ... View more

Dave, thanks for joining the discussion. I had compared the PhS values 247/147/30 as measured in PhS with the somewhat random Illu values in the Illu program window, I didn`t measure not in screenshots. Reference ist still the image in Gilligan's post #1, but in my impression the effect isn't accurately reproducible. Best regards --Gernot Hoffmann ... View more

Dag, I think, Gillian knows this already. I've tried to reproduce his workflow, but with CS6 instead of newer versions: Synchronize Photoshop and Illustrator for sRGB. Embedded profile overrides working space (actually here nowhere required) View PhS and Illu on the same monitor, even at he same time. Make Image in PhS with orange square RGB = 247/147/30 Save as PNG8 or PNG24 or TIFF (PNG doesn't embed the profile, TIFF does) Prepare a document in Illu: Document space RGB, Settings as mentioned (sRGB). CMYK is nowhere involved. Place PhS-Image (one of them). Move PhS-Window and Illu-Window side by side. The colors look slightly different. Save Illu doc as PDF. Open this PDF in PhS. Measure RGB in PhS: 247/147/30 ...OK! I still don't know, how to use Eyedropper in Illu for measuring instead of transferring a picked-up color to another object. Applying Eyedropper(pick-up) delivers odd numbers e.g. 245/148/42 or 247/147/1 or some other random-like combinations with unknown source. In the moment I can see only two possibilities left: Illu misinterpretes the common monitor profile or uses something different. Or Illu is affected by a random error. PhS is in this test everywhere correct. So far, the problem could not be solved in the Illu Forum. Best regards --Gernot Hoffmann ... View more

Pipedreamer, please don't insist on your idea of levels with 8 units distance. This is merely a mathematical construct without practical relevance. Opposed to this, the mentioned conversions map the input range x [0 to 255 step 1] to an output range y [0,255] with k steps, e.g. k=32. The step size cannot be exactly the same, e.g. 8,  for the range [0,255] but it could be for [0,256]. The compromise is already contained in the practical solutions. Best regards --Gernot Hoffmann ... View more

If you just want to represent the colors reasonably by 32 levels per channel, then use Posterize (as already explained in #4) or save as BMP X1-R5-G5-B5 and reload into Photoshop for further handling. No need for other software. The difference between this high color version and your suggestion is mainly, that you cannot represent pure white. X1 is a single dummy bit, because available file formats require 16 bit per color (in this case). The quantization is done by 5 bit per channel, equivalent to 32 levels. Best regards --Gernot Hoffmann ... View more

You may use Posterize with 32 levels per channel. The result contains level 0 and level 255 as well, not exactly what you want. Chr.Pfaffenbichler's tweaking might help. But in my opinion it's not reasonable to exclude the highest level. Here we have the full range but a slightly irregular sequence: 0 8 16 24 33 41 49 57 66  ... 247 255 Best regards --Gernot Hoffmann ... View more

Very good explanations, Dag. But eventually there is one thing not yet clear. Probably the OP refers to inkjet printers with RGB input, which are often  called – misleading – "RGB-printers". For instance Epson 7890 can be used as RGB-printer or, together with a RIP, as a CMYK- printer with CMYK input (what I'm doing). Here is a typical RGB profile for such an "RGB-printer": http://docs-hoffmann.de/Epson-RGB-Printerprofile.icc Original EPSON, with shorter name. ICC-Profile Inspector shows clearly: RGB-Input. Proof Setup for this printer profile and an RGB image offers indeed Preserve Numbers. In my opinion this should be ignored – it doesn't make any sense. Best regards --Gernot Hoffmann P.S.: Proofing Grayscale Images is eventually useful. This shows, how an image with defined Gamma (e.g. 2.2) or Dot Gain (e.g.20%) will look if reproduced by an output device with different Gamma or Dot Gain. ... View more

With special reference to Theresa, #2: Quote 1: https://www.w3.org/TR/2003/REC-PNG-20031110/#11transinfo Portable Network Graphics (PNG) Specification (Second Edition) The colour values in a pixel are not premultiplied by the alpha value assigned to the pixel. This rule is sometimes called "unassociated" or "non-premultiplied" alpha. (Another common technique is to store sample values premultiplied by the alpha value; in effect, such an image is already composited against a black background. PNG does not use premultiplied alpha. In consequence an image editor can take a PNG image and easily change its transparency.) Quote 2: https://en.wikipedia.org/wiki/Portable_Network_Graphics Transparency of image PNG offers a variety of transparency options. With true-color and grayscale images either a single pixel value can be declared as transparent or an alpha channel can be added (enabling any percentage of partial transparency to be used). For paletted images, alpha values can be added to palette entries. The number of such values stored may be less than the total number of palette entries, in which case the remaining entries are considered fully opaque. The scanning of pixel values for binary transparency is supposed to be performed before any color reduction to avoid pixels' becoming unintentionally transparent. This is most likely to pose an issue for systems that can decode 16-bits-per-channel images (as they must to be compliant with the specification) but only output at 8 bits per channel (the norm for all but the highest end systems). Alpha storage can be "associated" ("premultiplied") or "unassociated", but PNG standardized[17] on "unassociated" ("non-premultiplied") alpha so that images with separate transparency masks can be stored losslessly. End of quotes. Bold by me. So we see, transparency can be encoded by a special number per pixel (like in GIF, my guess) or by a adding an alpha channel value per pixel, this for continuous/variable  transparency. Only in the second case the image can be recovered in transparent areas. Best regards --Gernot Hoffmann ... View more

Use Pixels instead of Shapes Uncheck Anti-aliasing Especially the second line shows that Photoshop's line approximation is not optimal. Compare with correct lines (e.g. p.5): http://docs-hoffmann.de/drawf11102001.pdf Best regards --Gernot Hoffmann ... View more

An embedded CMYK profile would consume 1.5...2.5 MByte, e.g. 1.74 MByte for ISOCoated-v2-eci. The embedded profile will be part of the image file (PSD, TIFF, JPEG), it is not added as Metadata by XMP. The JPEG compression for RGB images is based on splitting RGB into YCbCr, where Y is the Lightness and Cb,Cr are residual colors (comparable to a and b in Lab). The numerical compression loss for Y is low, but very high for Cb,Cr. One reason is the so-called subsampling, where Cb (and Cr) for each group of 4 pixels is considered as one value. This CbCr com- pression doesn't affect the visual quality very much. But CMYK cannot be split into Y and residual colors, because this would destroy the black generation. Therefore each channel has to be treated as a Grayscale image, applying the same low-loss strategy as for Y. Therefore the filesize of a CMYK JPEG is not only larger by factor 4/3 compared to RGB, but in fact much larger for the same visual quality. Best regards --Gernot Hoffmann ... View more

A RAW file contains groups of four adjacent sensor pixels R G G B with (e.g.) 14 bit for each sensor pixel. This group contains the information 4 x 14 bit = 56 bit. The reconstruction for an image (demosaicing) delivers four pixels with three channels R,G,B each RGB RGB RGB RGB with (e.g.) 16 bit per channel R, G, B, with the information 4 x 3 x 16 bit = 192 bit. The increased information, 192 bit instead of 56 bit, is a result of cute interpolation, taking into account as well neighbour sensor pixels beyond the actual group, and maybe inherent image information. If the reconstructed image is represented by 8 bit per channel (instead of 16 bit per channel), then we have this information: 4 x 3 x 8 bit = 96 bit, which is still more than the original 56 bit, therefore partly fantasy.  By the way: 14 bit per channel in the RAW image is by no means relevant in the sense of accuracy and precision. Best regards --Gernot Hoffmann ... View more

No deja modificar el tamaño del GUI y esto es muy molesto, para las personas que no tienen pantallas grandes Better translation: It (the system) doesn't let us change the size of the GUI, and this is very annoying for people who do not have big screens Best regards --Gernot Hoffmann ... View more

Don't worry about – the file can be recovered. Remember: RGB (3 channels), Grayscale(1 channel) or CMYK (4 channels)? Probably you didn't enter a header size, overwriting 0 bytes, did you? Remember the true or approximate size in pixels. In landscape orientation width is larger than height. In portrait orientation width is smaller than height. Please post a screenshot of the visible result. Best regards --Gernot Hoffmann ... View more

Choose correct values for all parameters: Width: in pixels Height: in pixels Channels: most likely 3 (RGB) Interleaved: default Depth: 8 bits (default) or 16 bits, per channel Header Size: 0 bytes (default) An ICC-Profile can be assigned during opening or later, for instance sRGB. The file contains only the sequence of triples R,G,B from the first pixel to the last. Best regards --Gernot Hoffmann ... View more

A device link profile works only in one direction: RGB ––> CMYK CMYK1 ––> CMYK2 Of course D Fosse is right - for any application of a device link profile except printing one has to know the source profile and the target profile (for printing these informations were used implicitly, making the device link profile.) Now let's assume AdobeRGB (aRGB) as source and ISOcoated-v2-eci as destination profile. An image in aRGB can be converted to ISO eventually faster and more accurate than by using the standard workflow aRGB ––> LAB ––> ISO. In fact the creation of such a profile, as explained for the RIP Colorgate Productionserver requires an optimization:   DEVICELINK PROFILER Modul DLPFM   Voraussetzung: Sie haben das Devicelink Profiler Modul (DLPFM) erworben,   den Proof Workflow gewählt und ein Referenzprofil des zu simulierenden   Drucksystems erzeugt.   Der DEVICELINK PROFILER ist ein eigenständiges und optional erhältliches Modul.   DeviceLink Profile enthalten eine Farbraumtransformation von einem Gerätefarbraum   (Simulation) in einen zweiten Gerätefarbraum.   Der DEVICELINK Profiler Assistent optimiert DeviceLink Profile für den Proof Workflow   in iterativen Schritten, wobei das Messergebnis über einen weiteren iterativen Schritt entscheidet.   Erst durch eine Optimierung des DeviceLink Profils erreichen Sie eine optimale Farbsimulation   im Proofergebnis. Der DeviceLink Profiler kann auch Multi-Color-Profile optimieren.   Hier sehen Sie eine Übersicht der Schritte im Assistenten: ... The creation requires as well Rendering Intents, in the mentionened RIP either equal or different for raster graphics and vector graphics. Furtheron some rules how to Preserve (or not) pure colors, especially the reproduction of RGB-black by K-only, as mentioned in the original post. Now let's assume, we have told Photoshop the RGB space (aRGB), the CMYK space (ISO) and the Rendering Intent (only one, for raster and vector). Then we can read, using the color picker, aRGB-values and ISO-values without any application of "convert to profile". The problem seems to be solved: get aRGB numbers from CMYK numbers. Unfortunately this task doesn't have a general solution. Some RGB colors might have been reproduced in CMYK affected by gamut clipping, especially if the source data were delivered by usually highly saturated web sites. Even if we knew that gamut clipping happened for a certain color, the source color could not be reconstructed. So far about a strict solution of the problem. If the conversion from aRGB to ISO had been executed professionally, e.g. for book printing, then images in RGB were prepared by Soft Proofing so, that no loss due to gamut clipping happened (RGB colors were somewhat desaturated, blues shifted towards cyan, for instance), and Rendering Intent is always Relative Colorimetric, then the method as mentioned above should work satisfying. The interpretation of vector graphics (lines, rectangles, text), which exist in PDFs as independent elements besides raster graphics, will still be doubtful. Best regards --Gernot Hoffmann ... View more

Neil, thank you for your response. I'm a little late because of travelling. In the assessment of the problem, we agree: After replacing an Epson cartridge by one of the mentioned company, the tubes are still filled with Epson ink and the print will be correct. After an unspecified time the print changes (but there may be several reasons). Immediate recalibration is not recommended. Rather, one must wait until the tubes are filled completely by new ink. It's my impression that the inks are not entirely compatible, but I cannot prove it. That's why I am interested in the experiences of other users. The mentioned company doesn't provide reports from independent laboratories or opinions of satisfied users as references. Best regards --Gernot Hoffmann Sorry, if I marked your answer as correct, then this would imply problem solved, which is not the case. ... View more

soft_pilot, if all of your your swatches are looking considerably wrong, then please refer to jdanek`s advice. If only some of them (the most vibrant swatches) are affected, then you may read these explanations: Nowadays Pantone Swatches have to be defined by Lab. Any discussion about Swatches in RGB or CMYK would be quite misleading. Let's start with this doc: http://docs-hoffmann.de/swatch16032005.pdf p.4 – p.14 show a set of more than 1100 swatches. All are defined by Lab. Any page can be placed (opened) in Photoshop in mode Lab (or any other, which would be quite wrong for following discussion). Each swatch contains besides Lab values two triples of numbers, indicated by (s) for sRGB and by (a) for aRGB=Adobe RGB(1998). If any of these numbers is either clipped at 0 or at 255, then the color will be out-of-gamut for either sRGB or aRGB. Please let us ignore the rare case of colors exactly at the gamut boundary. If the monitor profile is near to sRGB, then aRGB colors will be further clipped, if in-gamut for aRGB but out-of-gamut for sRGB. A wide-gamut monitor is mostly near to aRGB. Then only those swatch colors will be clipped which show in the column (a) numbers  0  or 255. So far it is assumed, that the placing or opening in Photoshop happens as Lab page. We can see immediately from these first 11 pages, that many swatches are out-of-gamut for sRGB or aRGB, thus showing a wrong appearance on monitors. The following pages in the doc demonstrate this by other presentations, which can be explained on request. Which role play the global color settings, like North American General Purpose 2 ? This combination contains sRGB as RGB-working space. As long as  the source and the Photoshop document are still in Lab, these global settings are meaningless. Best regards --Gernot Hoffmann ... View more

According to your original post you're working in Illustrator. 1. You can use RGB swatches (Process RGB) unchanged, if your document color is RGB. If your Document color is CMYK then an RGB color is converted immediatedly into CMYK and from there back to RGB, if you read the swatch RGB numbers. This results in a severe change if the color is out-of-gamut for CMYK. Most likely for your green! 2. Now let's asasume that your document color is any RGB. The RGB numbers are to be considered as fixed, but the appearance depends on the RGB space (sRGB, AdobeRGB, ProPhotoRGB), as already explained. 3. Hex-Numbers (h) mean exactly the same as RGB decimal numbers (d)  – simply numbers: 0...9 (h) = 0...9 (d) A B C D E F (h) = 10 11 12 13 14 15 (d) The weights for the two digits in e.g. D4 are 16 and 1. 09 D4 04(h) = 0*16+9*1 / 13*16+4*1 / 0*16+4*1 (d) = 9 / 212 / 4 (d)  Thus, hex is just a very uncomfortable notation, exept for programming in assembly language . 4. Don't ever use so-called Web-safe colors. These are a small subset (only 216 colors) of 16.7 millions of colors as available by R,G,B in the range 0...255 each. The full set is safe anywhere, if the channels R,G,B are represented by one byte each. Best regards --Gernot Hoffmann ... View more

You'll need a so-called Multicolor profile. In ProfileMaker 5  profiling was available for CMYK+6 arbitrary inks or for 10 inks even without CMYK. http://www.photoshoplab.com/gretagmacbeth-releases-profilemaker-5-publish-plus.html Lc and Lm don't count as extra inks. I think ProfileMaker is no more available, you may ask X-Rite. However, the question is, whether Photoshop's Gamut Warning can handle Multicolor profiles. I don't have respective experiences, but my ProfileMaker package is still working perfectly. My inkjets (including an Epson inkjet) and toner printers are profiled as CMYK printers (opposed to RGB printers). Best regards --Gernot Hoffmann ... View more

Dimension in inches, pixel numbers and resolution are not independent. Of course 96 pixels and one inch should result in 96 ppi. With these settings (with resampling) I get exactly the intended result. The image shows the settings before the conversion. The state after the conversion is exactly the same. In #8 Resample is not checked. Best regards --Gernot Hoffmann ... View more

In which context is this deviation of 0.012/96 = (about) 10 to the power of minus 4 anywhere of any relevance for building a page? Best regards --Gernot Hoffmann ... View more

And don't forget all these trackers: Best regards --Gernot Hoffmann ... View more

Here we can see a "CMYK color wheel": http://docs-hoffmann.de/hlscone03052001.pdf p.16 The meaning of "CMYK color wheel" is explained there. Such a color wheel contains at the circumference CMYK values with maximal saturation. 0...1C, 0...1M, 0...1Y, K=0, but one value of a set CMY is always 0. The color wheel was generated by PostScript. RGB values are created in HLS (not in HSB, but that doesn't matter), and immediately converted into CMY values by the generic transform C=1-R, M=1-G, Y=1-B. Alternatively one might think of creating such a CMYK wheel by applying an ICC-profil to an RGB wheel. But this would deliver nothing else but garbage, because saturated RGB colors in any common RGB space are out of gamut in common CMYK spaces. This illustration is as well an answer to questions in this thread: Re: 12 Step Color Wheel Unfortunately not only the mathematics but also the underlying color theory aspects are somewhat complicated. Best regards --Gernot Hoffmann ... View more

PostScript Language Reference Manual, here Second Edition: https://www.google.de/url?sa=t&rct=j&q=&esrc=s&source=web&cd=10&ved=0ahUKEwjO1bLljL_JAhVCYA8KHdswBnkQFgheMAk&url=https%3… Page 305 PostScript had been invented earlier than color management systems (CMS) by ICC profiles. PostScript contained all necessary tools of a CMS, but it proved to be too complicated. For simple applications one can find generic transforms between DeviceRGB, DeviceGray and DeviceCMYK (see page 305...). These transforms may look nowadays quite undsatisfying, but in fact they are useful for tutorial purposes, e.g. for the construction of color wheels and spectrum bars. Best regards --Gernot Hoffmann ... View more

In addition to the already mentioned PostScript Language Reference, third edition https://forums.adobe.com/thread/1754351 this Technical Note #5600 about Smooth Shading might be useful: http://www.google.de/url?sa=t&rct=j&q=&esrc=s&source=web&cd=2&cad=rja&uact=8&ved=0CCoQFjAB&url=http%3A%2F%2Fpartners.adobe.com%2Fpublic%2Fdeveloper%2Fen%2Fps%2Fsdk%2FTN5600.SmoothShading.pdf&ei=ccwWVcqAG9HmaNP0gvAG&usg=AFQjCNEkcJZiNrzrhyW7gj44AHrQ4-xRzA&bvm=bv.89381419,d.d2s Please use simple examples with understandable numbers. Personally I'm not familiar with smooth shading. Just wanted to recommend the doc. Best regards --Gernot Hoffmann  ... View more