Preserving color bar with ink optimization

gordo

Well-known member
I haven't been able to figure out how to preserve the integrity of the color bar in a workflow that does reseparation for ink optimization. The 3/C grey patches in the color bar get K added and CMY tones are reduced. This is a Panther RIP/Binuscan workflow but I assume the same problem applies to other ink optimization solutions.
Doing a double burn isn't practical.

Any suggestions?

Thanks, gordo
 
I am not experienced with this workflow but there are a couple of different angles to approach this:

  1. Reseparate via preflight processes instead of at the RIP
  2. Depending on the system the colorbars/marks could be flagged with metadata that tells the RIP to leave them alone and not convert or curve them. For example: I have a plate control bar with embedded metadata that instructs to produce patches that are linear and patches utilizing the curve used on the separation exposed onto the plate.

Have you reached out to the support of the software company that makes the RIP?
 
Last edited:
So far the software company that makes the RIP has not been responsive.

best, gordo
 
I haven't been able to figure out how to preserve the integrity of the color bar in a workflow that does reseparation for ink optimization. The 3/C grey patches in the color bar get K added and CMY tones are reduced. This is a Panther RIP/Binuscan workflow but I assume the same problem applies to other ink optimization solutions.
Doing a double burn isn't practical.

Any suggestions?

Thanks, gordo

Is this related or similar to the potential of the color bars being modified when tone curves are applied?
 
Is this related or similar to the potential of the color bars being modified when tone curves are applied?

No. A 3/C CMY grey patch in the color bar goes through the ink optimization process and comes out as a 4/C grey patch rendering it more than useless.

gordo
 
No. A 3/C CMY grey patch in the color bar goes through the ink optimization process and comes out as a 4/C grey patch rendering it more than useless.

gordo

Even if you find a way to preserve the grey patch, how useful is it since much of the grey produced by cmy is removed in the image?

If it is just for process control, can you not just put in a grey patch with the correct values after the ink optimization step?


Another thought. If you leave the CMYK grey patch in, then it can be useful because it would show if the ink optimization works well. Does this make sense? For process control, it would also show some change in gray due to some of the CMY remaining in the patch.
 
If you can get images at platesetter resolution after screening, I would try replacing the patches in the colorbar with prescreened high resolution images. Then to the RIP, no pixels would be 50/40/40/0, only different combinations of 0 and 100 percent of cyan magenta and yellow. You'd probably have approximately 8 percent coverage of 100/100/100/0 pixels, and it may convert those. If so, perhaps there are configuration settings that would allow you to leave that color alone (e.g., maybe you can tell it to leave colors that are 100% of any ink alone). Of course, if you have lots of different screens in use, that would mean lots of different colorbars.


If the ink optimization comes after everything is flattened into a single contone image and before screening, then the way you build the colorbar (other than pre-screening the gray balance patches) probably can't make any difference. If the process happens earlier, maybe there's a way to trick it. I would also try making three separate N-channel contone images (not screened - just a flood of 50 or 40 percent) - one each for cyan, magenta and yellow, then setting them to overprint. You could also try setting the blend mode to multiply or interleaving them with white transparent images to try to confuse the RIP.

This raises an interesting question: how useful is a gray balance patch with max black generation? It would probably help keep the ink densities in proper ratio to each other, but wouldn't really show the actual gray balance since you're mostly locked in with all that black. Hope that doesn't derail the thread.
 
Even if you find a way to preserve the grey patch, how useful is it since much of the grey produced by cmy is removed in the image?

If it is just for process control, can you not just put in a grey patch with the correct values after the ink optimization step?


Another thought. If you leave the CMYK grey patch in, then it can be useful because it would show if the ink optimization works well. Does this make sense? For process control, it would also show some change in gray due to some of the CMY remaining in the patch.

A 3/C grey patch is completely useless for process control in an ink optimized workflow. What I'm experimenting with is the notion of using it as a SID reference instead of solid color patches. This is for application in a newspaper printing. I'm also trying to use it as a training tool for the press operators so that they can get a feel for how their SID moves affect (or more accurately don't affect) different types of imagery.
Adding the patch after the optimizing process adds complication and time to the workflow. I'm wanting to avoid that.
An optimized CMYK grey patch would have less than no value for the reasons you and I gave.

best, gordo
 
If you can get images at platesetter resolution after screening, I would try replacing the patches in the colorbar with prescreened high resolution images. Then to the RIP, no pixels would be 50/40/40/0, only different combinations of 0 and 100 percent of cyan magenta and yellow. You'd probably have approximately 8 percent coverage of 100/100/100/0 pixels, and it may convert those. If so, perhaps there are configuration settings that would allow you to leave that color alone (e.g., maybe you can tell it to leave colors that are 100% of any ink alone). Of course, if you have lots of different screens in use, that would mean lots of different colorbars.

I've tried sending prescreened images through the optimizer. but it does "see" wherever three solid colors overlap and does its optimizing on that so I get a peppering of black pixels introduced.


If the ink optimization comes after everything is flattened into a single contone image and before screening, then the way you build the colorbar (other than pre-screening the gray balance patches) probably can't make any difference. If the process happens earlier, maybe there's a way to trick it. I would also try making three separate N-channel contone images (not screened - just a flood of 50 or 40 percent) - one each for cyan, magenta and yellow, then setting them to overprint. You could also try setting the blend mode to multiply or interleaving them with white transparent images to try to confuse the RIP.

Yep I'm looking to try and trick the RIP.

This raises an interesting question: how useful is a gray balance patch with max black generation? It would probably help keep the ink densities in proper ratio to each other, but wouldn't really show the actual gray balance since you're mostly locked in with all that black. Hope that doesn't derail the thread.

The gray balance patch in a color bar with live image area separated with max black generation is completely useless. I've tested this extensively recently on a coldset web press to help the press operators understand that their SID ink key moves weren't really affecting the imagery. I.e. don't waste your production time fiddling with SIDs.

best, gordo
 
A 3/C grey patch is completely useless for process control in an ink optimized workflow. What I'm experimenting with is the notion of using it as a SID reference instead of solid color patches. This is for application in a newspaper printing. I'm also trying to use it as a training tool for the press operators so that they can get a feel for how their SID moves affect (or more accurately don't affect) different types of imagery.
Adding the patch after the optimizing process adds complication and time to the workflow. I'm wanting to avoid that.
An optimized CMYK grey patch would have less than no value for the reasons you and I gave.

best, gordo

OK, I understand that if there is ink optimization by taking out some of the grey cmy, the printed image will not be so affected by changes in solid density. That would mean that the accuracy of the final separations are very important in obtaining the desired colour of the printed image since there is not so much the operator can do in that situation.

But I hope you are not suggesting that any SID values are OK. I would think that at least the SID values need to be in a specific range. Especially where one has no grey in the print but only combinations of one or two screens of the cmy channels. With only one or two channels I would think that density control again becomes important or at least more important than in an area that has the ink optimization applied to remove some of the CMY inks by reducing the dots sizes.

Also I would add that one of the purposes for aiming at SID values, even if those inks are not used in line with those patches is due to the benefit of having a reasonably consistent ink film across the form rollers. An uneven ink film can cause ink to move laterally from an area that does not use all the ink to an region that does use the ink and can affect its control.

Even within an image one can have small areas where cmy grey exists and where it does not exist. This can be next to each other or in line with each other. Removing cmy grey will only affect the parts of the image. Density control is still required to maintain the non cmy grey parts of the image.

Another dynamic issue is caused by the reduction of ink needed in line with a control patch. This dynamic exists for not only grey removal but also for going to FM screening from AM screening. The reduction of the ink required will move the dynamics of the ink feed to a condition that looks like a lower coverage condition and lower coverage is more difficult to control density. With FM, the additional situation where screens are not affected by changes in ink film on the form rollers means that the more solid or high screen values areas are more affected by variations in ink feed. These are probably not so significant but they are something to think about.

Also specifically with newspaper inks, which are weaker than commercial or packaging printing inks, similar moves with ink keys would not result in the same amount of density change. It would look less sensitive.

As you know, my view is that ultimately control of density is what is critical and should be solved. When that happens, the other issues would be much less critical and easier to deal with. Just my view. :)
 
Last edited by a moderator:
In our workflow, up until a month ago, was mapped to a screen angle in Esko that the plate compensation curve treats as linear. As of a month ago, our Director of Color Management spent a week with the good folks who brought you Gracol and G7, and they recommended to us that the color bars should have the DGC applied to them. The DGC will only adjust for dot gain, not reseperate the files. It sounds like your optimizing the files (applying GCR/UCR) in your plate rips, which is not what Esko does.
 
I like the metadata approach, however it of course depends on the workflow/RIP. Why not try a plate resolution pre-separated 1 bit tiff colour bar art. It can be recombined and overprinted in InDesign to produce a composite workflow friendly piece of art. I know that this will be a bit fiddly if they use different screen rulings, dot shapes etc.


Stephen Marsh
 
Hi Gordo,
If your Workflow is
A) go through ink optimizing
B) data is going to RIP

Then my first instinct is to add the bar in the rip. Now Panther RIP Is unknown to me so I have no clue if this can be done there. If it is Hq based you can do this, as well as in many others but as said above I don't know Panther RIP.

-Johannes
 
Hi Gordo.
Can you overlay 3 color bars on top of each other (one with your Cyan build, one with your Magenta, one with your Yellow), set them to overprint, and set two of their densities to "0" so the RIP misses the ink optimization setting? Just spitballing here.

"Just wanted to tell you both good luck. We're all counting on you."
 
Hi Gordo.
Can you overlay 3 color bars on top of each other (one with your Cyan build, one with your Magenta, one with your Yellow), set them to overprint, and set two of their densities to "0" so the RIP misses the ink optimization setting? Just spitballing here.

Consider the ball spitted. Binuscan, their ink optimization software, clobbers everything where three colors are used. But they are now using FM screening so the SIDs aren't such an issue anymore in terms of variation in presswork. The only practical issue they have in running without color bars is that if they have an ad that the advertiser is not happy with they don't have a way to confirm that they ran to SNAP and so therefore the problem is with the source file rather than their presswork. That technical problem is solved via advertiser discounts and preferred positioning in the newspaper.
 

PressWise

A 30-day Fix for Managed Chaos

As any print professional knows, printing can be managed chaos. Software that solves multiple problems and provides measurable and monetizable value has a direct impact on the bottom-line.

“We reduced order entry costs by about 40%.” Significant savings in a shop that turns about 500 jobs a month.


Learn how…….

   
Back
Top