Question about spot colors...

[gordo]

Just making sure that I'm on track with reality (at least as far as this goes LOL)

If I have a job in, say, InDesign, that has CMYK plus 3 specified spot colors (that should remain as spot colors) then, unless I do something goofy I will get 7 plates for the press (1 each for the CMYK and 1 each for the 3 spot colors) - i.e. it's a 7 color job with 7 plates for press.

However if I output the same InDesign file to a digital press (assuming the typical CMYK or LCCLMMYLK inks that are used) it really doesn't matter how many spot colors are specified in the source file. The file will get RIP'd and converted to output on that printer using those inks. In fact, I could have specified 50 spot colors in addition to CMYK and it really wouldn't matter to the inkjet press/printer whereas if destined for an offset press I would end up with 54 plates which probably wouldn't be appropriate.

Also, Spot colors overprinting each other or overprinting CMYK will probably not appear on offset presswork the same as they would appear as proofed on an inkjet printer.

Is that correct?

 

[dabob]

Gordo . . Yes you are correct but the last paragraph about inkjet printers would depend on your rip as to the accuracy of the overprinted colors. - you can use the color manager to turn all your spots to process also . . or selectively turn them into process.

 

[gordo]

Gordo . . Yes you are correct but the last paragraph about inkjet printers would depend on your rip as to the accuracy of the overprinted colors. - you can use the color manager to turn all your spots to process also . . or selectively turn them into process.

Thank you! That's very helpful.

 

[arossetti]

Also most digital front ends have a spot color library and depending on the color management settings will treat the spot colors differently then the CMYK tagged art. If the color is left as a spot and the DFE is setup to accept spot colors then if there is a matching pantone in the library it will bypass the source profile. Obviously if you convert all spots to CMYK before the DFE they will go through the source profile and as you already know that will clip the possible gamut.

 

[zoran]

Ditto to what Dabob and Arosetti said, just to add for example on Nexpress, even though it is digital device, you can utilize additional toner (5th unit) that will complement CMYK to expand colour gamut and create close match to actual Pantone colour, so leaving spot channel as such is desired.
This is controlled on the RIP itself as Arosetti pointed.

 

[gordo]

Thanks all for your input.

I can be more specific. The example I'm looking at is an EG image. This could be an RGB image with extra channels to expand the gamut or a CMYK image with extra channels to expand the gamut.

If I softproof the image in PShop it seems to give me a good representation of how the image will look printed on a buddie's 12 color Canon IPF6400 inkjet printer driven by an EFI RIP. So it seems that I can expand the gamut of an image by adding spot color channels to it. And if my final destination is that inkjet printer then all is good.

But, I'm not sure that the inkjet will properly proof what the color will look like if I take that same CMYK image with the extra channels and print it using CMYK plus spot color inks on an offset press.

Thoughts?

 

[Stephen Marsh]

Thanks all for your input.

I can be more specific. The example I'm looking at is an EG image. This could be an RGB image with extra channels to expand the gamut or a CMYK image with extra channels to expand the gamut.

If I softproof the image in PShop it seems to give me a good representation of how the image will look printed on a buddie's 12 color Canon IPF6400 inkjet printer driven by an EFI RIP. So it seems that I can expand the gamut of an image by adding spot color channels to it. And if my final destination is that inkjet printer then all is good.

#1: If this is the case, my guess is that it is probably pure luck, a freak accident! What about if you change paper and or settings on the RIP for the different paper?

This is something that only a few softproofing solutions appear to support. YMMV.

But, I’m not sure that the inkjet will properly proof what the color will look like if I take that same CMYK image with the extra channels and print it using CMYK plus spot color inks on an offset press.

Thoughts?

#2: Agreed, most inkjet RIPs only have limited info, such as the solid L*a*b* values of 100% of the ink, if you are lucky then you have tone curve info and the ability to input the L*a*b* value of the spot colour should at say 50% etc. Add to this the problem that there may be no accurate way for the RIP to know the opacity of the ink, lay down order or how the ink will interact with other spots or process colours with the opacity of the spot in question and other inks be they spots or process (solid or tinted ink trapping).

Of course, there have been proprietary solutions for proofing to overcome these issues, mostly in packaging workflows.

Until recently, there has not been an open standard for communicating this info among different parties (brand, design, prepress, ink lab, pressroom). Now there is:

ISO 17972

And the first product to market supporting this new standard:

http://www.cgs-oris.com/index.php/en...is-cxf-toolbox

Again, YMMV!


Stephen Marsh

 

[Deleted member 16349]

But, I'm not sure that the inkjet will properly proof what the color will look like if I take that same CMYK image with the extra channels and print it using CMYK plus spot color inks on an offset press.

Thoughts?

Isn't this the heart of the colour management problem? People can not be sure how the print will look.

Again, as I see it, this lack of predictability is due to the methods used. Trying to predict how a printing device, either digital or analogue, will print with information that is not complete, will always tend to give results that are not predictable. This should not be the case and it does not have to be.

Ideally, a printing device should be a device that produces repeatable outcomes. Printing devices only have a specific number of combinations that it can print. The number of these combinations can be very large but they should ideally be repeatable.

Of course ideally it should be repeatable and the more repeatable it is, the better. It is the job of the print device designers to work towards making these devices repeatable.

If one had a device that is not repeatable, then that device can NOT be a printing device.

IMO the practical and the most accurate way to deal with this problem is to map printing colours of the repeatable printing device and use that information to predict the colours that will be printed. Of course, the question is how should this be done. That is the direction that I think needs to be developed.

In an earlier post, I commented on a colour measuring device that could read up to 10,000 patches at one time. It is that kind of capability that is required to help map the total output of a printing device to give one the capability to predict how that device will print any colour.

Trying to model the output of a device without the required information, will aways be a problem and as has been suggested in an earlier post, if you get a good result sometimes, that might only be due to random luck.

 

[Possumgal]

Then you've got the problem of some spot colors that are really hard to replicate on an inkjet, like the perennial Reflex Blue problem. When you're using a spot that you know is a problem, you may just have to explain to the customer, perhaps offering a press proof (at his cost), if he's really that picky. Odds are, he won't be.

 

[Werby]

#2: Agreed, most inkjet RIPs only have limited info, such as the solid L*a*b* values of 100% of the ink, if you are lucky then you have tone curve info and the ability to input the L*a*b* value of the spot colour should at say 50% etc. Add to this the problem that there may be no accurate way for the RIP to know the opacity of the ink, lay down order or how the ink will interact with other spots or process colours with the opacity of the spot in question and other inks be they spots or process (solid or tinted ink trapping).

Of course, there have been proprietary solutions for proofing to overcome these issues, mostly in packaging workflows.

Until recently, there has not been an open standard for communicating this info among different parties (brand, design, prepress, ink lab, pressroom). Now there is:

ISO 17972

And the first product to market supporting this new standard:

http://www.cgs-oris.com/index.php/en...is-cxf-toolbox

Stephen, can you explain more about ISO 17972 and how the ORIS CFX Toolbox overcomes the problems of opacity, lay-down order or how spots will interact with process and other spots? I understand that it can accurately communicate the colors you WANT, but how does it help generate a proof of what you are actually going to GET? We are often mixing multiple spot colors with process in images and what we see on screen in photoshop looks one way, what comes out of our inkjet proofer looks another, and what happens on press is another thing entirely.

We've experimented with tools that purport to soft proof spot color overprints better and it has improved the situation, but without an ink draw down from the printer on the specific stock in question is very hard to know exactly how spot inks are actually going to look. I'd love to hear more about solutions!

 

[Bill W]

In regards to number of spot colors that will be converted in a 7 color strategy, at least on an Indigo using the Esko DFE, the most can be 27. I belief that this is a limit imposed by the Adobe RIP that is being used. Additionally the Esko DFE does not have the capability to turn RGB images into EG. This has to be done before and then sent through a strategy that will honor what had been done upstream.

Cxf toolbox - I have starting using this and it requires a set of tints be printed on the press and then scanned into the Oris Rip. It also requires one to save the files in a certain PDF flavor. In my limited tests the proof looks very close to the tint ramp that was printed on the press. I have not tried color over color yet.

 

[D Ink Man]

Just making sure that I'm on track with reality (at least as far as this goes LOL)

If I have a job in, say, InDesign, that has CMYK plus 3 specified spot colors (that should remain as spot colors) then, unless I do something goofy I will get 7 plates for the press (1 each for the CMYK and 1 each for the 3 spot colors) - i.e. it's a 7 color job with 7 plates for press.

However if I output the same InDesign file to a digital press (assuming the typical CMYK or LCCLMMYLK inks that are used) it really doesn't matter how many spot colors are specified in the source file. The file will get RIP'd and converted to output on that printer using those inks. In fact, I could have specified 50 spot colors in addition to CMYK and it really wouldn't matter to the inkjet press/printer whereas if destined for an offset press I would end up with 54 plates which probably wouldn't be appropriate.

Also, Spot colors overprinting each other or overprinting CMYK will probably not appear on offset presswork the same as they would appear as proofed on an inkjet printer.

Is that correct?

My SWAG is that you are 99.44% correct; for whatever that's worth.

D

 

[D Ink Man]

Isn't this the heart of the colour management problem? People can not be sure how the print will look.

Again, as I see it, this lack of predictability is due to the methods used. Trying to predict how a printing device, either digital or analogue, will print with information that is not complete, will always tend to give results that are not predictable. This should not be the case and it does not have to be.

Ideally, a printing device should be a device that produces repeatable outcomes. Printing devices only have a specific number of combinations that it can print. The number of these combinations can be very large but they should ideally be repeatable.

Of course ideally it should be repeatable and the more repeatable it is, the better. It is the job of the print device designers to work towards making these devices repeatable.

If one had a device that is not repeatable, then that device can NOT be a printing device.

IMO the practical and the most accurate way to deal with this problem is to map printing colours of the repeatable printing device and use that information to predict the colours that will be printed. Of course, the question is how should this be done. That is the direction that I think needs to be developed.

In an earlier post, I commented on a colour measuring device that could read up to 10,000 patches at one time. It is that kind of capability that is required to help map the total output of a printing device to give one the capability to predict how that device will print any colour.

Trying to model the output of a device without the required information, will aways be a problem and as has been suggested in an earlier post, if you get a good result sometimes, that might only be due to random luck.

So in the end. printing still remains intact as an ART.

D

 

[Johu]

Stephen, can you explain more about ISO 17972 and how the ORIS CFX Toolbox overcomes the problems of opacity, lay-down order or how spots will interact with process and other spots? I understand that it can accurately communicate the colors you WANT, but how does it help generate a proof of what you are actually going to GET? We are often mixing multiple spot colors with process in images and what we see on screen in photoshop looks one way, what comes out of our inkjet proofer looks another, and what happens on press is another thing entirely.

We've experimented with tools that purport to soft proof spot color overprints better and it has improved the situation, but without an ink draw down from the printer on the specific stock in question is very hard to know exactly how spot inks are actually going to look. I'd love to hear more about solutions!

I know it's an old thread but I thought I would still answer the above question. CxF toolbox needs the information of a CxF wedge which in this case usually is a wedge with 0 -100% with 10% steps containing just the spot color and a second row with the spot color printed on top of 100% black again with 10% steps. This way you not only get measured information of the pure spot color and it's gradation but the same thing on top of another color ie trapping of the spot color. With this information it is much more likely for one to be able to produce more accurate proofs for example even with overprinting spot colors.