Ernst, the process you describe would be the logical choice, if the z3100 was infact doing this.
The way the z3100 currently mixes colours is a key to the problem that the HP engineers who are coming to see us are looking at. Currently the device always mixes the additional 3 components with the cmyk base components making it a heptachrome printer.
This has lead to a scientifically provable loss in maximum saturation and the fact that the z3100 achives its best a,b, red value at too higer L values.
I have been in the commercial printing game designing colour workflows for over 17 years and the best approach to n colour printing is to create a base cmyk gamut and then create a secondary gamut using the additional colour which is what we used to do with litho when we played with 8 colour printing (C,M,Y,K,LC,LM,O,G).
If you start trying to profile cmykrgb you can get overlaps where the machines colour mixing will intervene and possibly give you adverse results as a long and exhaustive testing program has shown to be the case with the z3100.
[{POST_SNAPBACK}][/a]
Robert,
The Wasatch SoftRip that Marc also uses (I have 4.5, he has the latest version) has in the advanced color control section a choice to get into the Color Separation Rule, something that is normally a black box in RIPs (and drivers) or something that is entirely defined in the CMYK(RGB) profiling. Here it allows you to make the N-color Printer + RIP a "CMYK device" or "RGB device" and it defines how the extra hues RGBwhatever are integrated in the CMYK mixing model. With that rule used you can apply "CMYK" or "RGB" profiles on an N-color printer (CMYKRGBwhatever). Not so different from HP's APS solution.
The 5 choices are:
CMY Dominant which will cause the extra hue inks to be used in areas of strong color saturation but prefers the use of CMY mixes in areas of low color saturation.
The Hue Angle selection assigns inks to colors on the basis of their Lab hue angle, color saturation isn't a value there.
There are the Full Ink versions of the two above where the ink limits are increased for textile printing mainly.
There's a Pass Through choice that will use the separations as specified in the CMYK profile and it will not use the HiFi inks but they are available in the RIP for spotcolor areas.
What I do not see in my version is the setting of the Lab hue angle numbers for the RGB inks but it could be a black box thing that Wasatch sets per printer.
If you can influence the color separation rule to the same degree for N-color in N-color profile creation packages like the Wasatch SoftRip allows I will be surprised but I am not familiar enough with them. I do not know what other CMYKRGB profilers have in choices but setting the Lab hue angle values for the extra hues is a thing they have in common. At least that is what I have seen. I still wonder why they didn't add that Lab description for all the CcMmYRGB hues in that case as it wouldn't surprise me that the CcMmY hues in a set like that are not the same as used in a CMY/CcMmY set. Even the Epson K3 magenta has a huge shift compared to the UC magenta so it is also happening in conventional inksets. Not to mention the warm black, grey inks of the K3 set compared to the Z3100 near neutral monochrome inks. The last must be ideal for any profile creation software.
According to your message HP's choice for their driver has been the Hue Angle variety. I'm sure they are way more knowledgeable in this color mixing science than I am and I suspect that the inkset's RGB hues are specifically selected to be at the right hue angles within the CMY hues they have selected. Depending on their Chroma + transparency that Hue Angle integration choice may be the best choice in this case, not fully exploiting better RGB saturation at the extremes but cleaner RGB colors, more harmony in the gamut. Other benefits could be: lower metamerism, better fade resistance, more consistency in color. In my view this is still a true CMYKRGB mixing but not one that aims for an extended gamut.
The extended gamut would be possible with higher chroma RGB inks but it would show as bubbles on the gamut on the RGB spots (in this case orange, green, violet, spots). The extended gamut may be possible with higher Chroma CMYKRGB inks but that could influence fade resistance, there is a compromise to be made.
The Epson kind of dark gamut is the result of a cmyk inkset and the resulting CM, MY, YC, mixes that get their highest saturation at lower L values as that kind of mixes take out more light. With more ink transparency and more chroma a good saturation is possible. The ink volume used at those spots will be higher. The mixes less clean than the HP mixes. The other advantages mentioned for the HP CMYRGB mixes are not available. The Epson gamut doesn't have the higher L value saturation of the Z3100. So which one has the best gamut isn't clear to me.
It has to be seen whether the CMY dominant separation rule (as described above) for the Z3100 inks has the ability to get that lower (Epson) L value saturation as well for the gamut. If so it probably will be a subtle mixing of CMV, MYO, CYG, instead of straight CM, MY, CY, depending on the transparency and chroma of the CMY inks. More ink will be used then. Let's say something of the Full Ink variety mentioned above. There are cons to that choice like detail loss, gloss changes, ink use. But right now after calibration on the Z3100 I see nice saturated colors on several choices of target sheets (other profilers included) that make me think there's right now still a lot to gain with good profiling without changing the separation rules and with the ink limits and the ink specs as they are right now. Using the latest media profiles.
I'm not unhappy with the existing separation rule in the driver for for example Matte Litho (one of the media profiles that got better with the upgrades). Better than my Epson 10000 with MIS 7600 produced, equal to Epson 4000 prints I have to match but not a 1:1 match. I have to tweak the profile for the blue range where the dark blues tend to shift to violet but the blue saturation is better than what I have to match. One can expect some difficulty there where one Cyan (+ Violet and Green) replaces an LC an C. But what am I aiming at, something Epson created over the last 8 years or the ideal gamut ? I see shifts to my Fuji test image in both inksets. I like the overall result of the Z3100 more right now.
The normal offset printing gamut usually lacks a lot more saturation in art reproduction etc than normal inkjet printing does. Total gamut is smaller. The aim there is to get saturation of specific colors (often specified per painting or groups of paintings) correct and keep the CMYK mixes in harmony. In the inkjets we discuss here the gamut is already way beyond offset gamut and we have to use the complete set for everything that comes through the shop: gloss, matte, on any media, whatever. In offset the media selection is either limited or the ink set is changed to the media to be printed. I think that you can not translate offset practice to our inkjet printing.
Ernst Dinkla
try: [a href=\"http://groups.yahoo.com/group/Wide_Inkjet_Printers/]http://groups.yahoo.com/group/Wide_Inkjet_Printers/[/url]
