Printing gradients with 2 Pantone colors

Bento

Member
Hi everyone,

Its been years since I have worked in a pre-press environment, so my skills are a bit rusty to say the least. Here's the issue:

We have a 2-color cover with a gradient goinf from Pantone 180 (red/orange) to Panton 116 (Yellow). I originally left this built in InDesign not knowing that banding was still an issue (as it is in Quark-built gradients). Our printer and their service bureau has so far offeen no solutions. So, how do I build a 2/color Pantone gradient that doesn't have any banding. I have been told this can be done in Photoshop, involving Spot Channels, but I am not experienced enough in this to know how to do it. Thanks.

B
 
Re: Printing gradients with 2 Pantone colors

B,

First, you want to say that in many cases, going from one color to another makes the gradient look too light in the middle (where only 50% of each color is printing) to look good when printed. So in these times, you may want to print a gradient of the darker color over a solid of the lighter color. You want to try it both ways and soft-proof (turn on Overprint Preview and choose the output press ICC profile) to see which looks better. You can do this in InDesign or PhotoShop either one.

Second, I wanted to say that I am a prepress person that uses Nexus, and haven't had problems in over five years with gradients. But some that use older equipment still may have problems with gradients. So here's the directions I would give for when the gradient needs to be rebuilt in Photoshop:

You'll want to make a new CMYK PhotoShop document with the same dimensions width and height as your gradient in InDesign. Make it the resolution your service bureau uses in their rip (usually 300 unless specified otherwise by your service provider/printer).

Note: I'm doing this in CS2.

Go to Windows menu > Channels. Click on the arrow at the top of that Window. A flyout menu appears. Choose 'New Spot Channel...'. Another window opens. Click on the square to choose the spot color. Two icons under OK, if it says Color Libraries, that means you are in Picker and would have to click Color Libraries to choose a Spot color.
I keep mine on PANTONE solid coated library in the pull-down list, unless choosing color for job printing on uncoated stock, which I would then choose solid uncoated library, or if choosing a Spot to print in CMYK, choose solid to process coated library. Type in 180 and it will go to PMS 180 in the list. Click OK. (Note: If you wanted your color to be a custom Spot color, you would choose the color to make it display here, and then OK.). Click OK again.

Choose the PMS 180 channel (and make sure it's the only channel chosen), and do a gradation with the gradient tool, usually beginning at the far left of the document and going to the far right (or top to bottom, but always making sure that the foreground color is solid black and the background color is solid white, unless you want to go from a tint of one color to another). Now in channels window, drag the PMS 180 channel to the new channel icon at the bottom of that window to make a duplicate copy of PMS 180 channel. Double-click on that channel's icon, click the Color swatch, and type 116. Click OK. Click OK. Now go to Image menu > Adjustments > Invert. This will make an exact opposite gradation of the 180 channel.

Now you see one scenario. I also want you to make another scenario where the PMS 116 is solid. Drag the PMS 116 channel down to new channel icon in channel window to duplicate it. Make sure solid black is chosen as foreground color. Edit menu > Fill... Foreground color, Normal, 100%, click OK.

To soft-proof in PhotoShop: View menu > Proof Setup > Device to Simulate: choose an industry official ICC profile such as GRACoL2006_Coated1v2 if printing on coated, or choose ISO uncoated if printing on uncoated (or choose custom profile from your printer if you have one). Click Preserve CMYK Numbers. Check Simulate Paper and Simulate Ink (Click Save if you want to save the setup for easy choosing in the future). Click OK.

You now have two PMS 116 channels. Turn them on and off to see which looks better with the PMS 116 channel. Delete the one you don't want to use, as well as deleting the CMYK channels, save as .psd, keeping the Spot Colors. place in Indesign at same X and Y location as InDesign-built gradation, and then delete InDesign-built gradation after making sure placed gradation is in correct position.

The good thing about the solid, although it will change the other solid's color even if slightly, is that the gradation visually doesn't go too light in the middle. I use this almost technique always if I have a black gradation over a color. It just helps to make the black richer also, as well as fixing the "ghosted middle" look. If doing this technique in InDesign, you would simply:
Duplicate (Step & Repeat at 0,0 offset) the gradation, make the top gradation from one PMS color to white, the bottom object solid, and then overprint the top object with the gradation.

Hope this all helps.

Don
 
Re: Printing gradients with 2 Pantone colors

Don,

Wouldn't Illustrator do the trick as well; perhaps easier if you want shaped objects?

Can one *safely* add noise to gradient spot channels in Photoshop? Now that might be a plus.

inez
 
Re: Printing gradients with 2 Pantone colors

Inez, you are right. Of course, people have used adding noise to prevent banding in the past, and this would be done in Photoshop (some rips have it as an option also and do it on the fly - I have the option in my rip but haven't needed to use it). A small amount of noise can help especially if doing a gradation over 7 inches in length (I believe it was that I read), where the problems become more apparent because of very small percent of changes over longer distance (e.g. a gradient of 0-10% will of course band more than a gradient of 0-100% in the same distance. At some point the banding will become apparent especially if the rip is rounding percentages. Nexus doesn't round percentages and so the gradations are smoother IMO). Someone else will have to talk about noise settings perhaps, since I haven't had that particular problem in years, and my post focused on how to make the gradient in Photoshop. Thanks Inez for adding that info.

Don
 
Re: Printing gradients with 2 Pantone colors

-To avoid the light "valley" in the transition area, rather than creating a gradient from PMS 116 to PMS 180, have it go from PMS116 to a combination of 100% PMS 180+100% PMS 116.

To do this, first make a Mixed Ink that consists of 100% of both colors: Choose Mixed Ink swatch from the Swatches platte menu fly-out and check both spot colors and set the value to 100%.

Then, create a gradient that's PMS 116 on one end and the new mixed ink on the other.

This alone may cheer up your banding problem.

-a PostScript Level 3 RIP shouldn't be having banding issues with this strong a gradient. Are you seeing this in inkjet proofs? It may be a linearization issue with that device. Or is this happening with final plate and print? Or is this job printing digitally?
 
Re: Printing gradients with 2 Pantone colors

As Claudia pointed out (maybe in a little different lingo), if the final printed piece looks like that, the plates may have curves that are not smooth, which would be the printer's responsibility to fix. As we see, there are multiple places for it to go wrong and band. Hope we have given enough info to help.

Don
 
Re: Printing gradients with 2 Pantone colors

Hey, this is a great help. I thought I'd share some tricks I picked up, with doing gradients.

One of the brands I work on has a thing for subtle background gradients. Their favorite background color is currently a soft blue, in the neighborhood of PMS5523. It's a very light color, I've included an image for reference.

I found the easiest way to build a background file for this, is to build a CMYK gradient, in Photoshop. Instead of using the gradient tool, however, I place each color in a fill layer, using layer masks to build the gradient. Someitmes I build the color out of multiple layers of the same color.

I then add noise to the layer masks (Filter > Noise > Add Noise), using a percentage between 1 and 5 percent, with Gaussian distribution. The percentage really depends on the size of the ramp, and the number of colors. It's something I play around with until the proofs look good.
 
Re: Printing gradients with 2 Pantone colors

Thank you all so much for your explanations. I now see how it should have been done and can be done. Much respect for you talents and knowledge.
 
Re: Printing gradients with 2 Pantone colors

After trying/soft-proofing the first 2 methods mentioned above, the second, involving the solid fill on the bottom channel appears much richer. The first method had the fading issue in the middle where they meet. I also did the mixed-ink method in InDesign. Excellent suggestion. I have yet to try some of the other ideas, but will just so they are in my mental library of tricks.

Claudia, to be honest I don't know what types of proofs these were, as they left the building last week along with the person who had that answer. I believe they were digital but other than that I don't have any more details. Since this job is off to be printed (and fixed at the printer) there isn't much I can do now. But, I will most certainly have a good handle on it for the next time. Thanks again to everyone.
 
Re: Printing gradients with 2 Pantone colors

OK, so I presented these concepts to the designers here so they are more knowledgeable and was presented a question I didn't know how to answer. When using the method Claudia mentions using the mixed inks in InDesign, what is actually going to happen when it goes to print. The designer here was under the impression that this would create and actual mixture of the 2 inks when printed. I imagine this is taken care of on the RIP side of things involving overprinting where the mixed ink is interpreted a certain way by the RIP. Any quick explanation would be appreciated.
 
Re: Printing gradients with 2 Pantone colors

You're welcome Benton. I learned something from this too. Although I knew Quark had multi-ink option in the Swatch palette, I didn't know InDesign did as well. Learn something new everyday.

Don
 
Re: Printing gradients with 2 Pantone colors

Benton,

I just did a test and sent the multi-ink InDesign doc to my rip, and also saved as PDF/X-1a and dropped on a hotfolder to my rip. Just as you see in InDesign with Separation Preview is how it ripped - the yellow is solid and the red gradation overprints it.

Don
 
Re: Printing gradients with 2 Pantone colors

The designer here (who likes to play devil's advocate—to put it politely) argues that the mixed ink swatch and overprinting the orange will cause it to not look like it would were it not mixed/overprinted. Will the orange (defined as a mixed ink swatch of 100% of the 180 and 116) overprinting the yellow (116) still closely match the 180 in the swatch book? This will be my last question on this issue as I know everyone is busy. Thanks again.
 
Re: Printing gradients with 2 Pantone colors

Benton,

As said before the first time I replied:
"The good thing about the solid, although it will change the other solid's color even if slightly, is that the gradation visually doesn't go too light in the middle. I use this almost technique always if I have a black gradation over a color. It just helps to make the black richer also, as well as fixing the "ghosted middle" look."

It may not match PMS 180 but will be a darker (IMO better) red.

Having said that, if you must match 100% PMS 180 on one end (or as close as possible), you may want to build it in InDesign and have a swatch of 100% PMS 180 at the PMS 180 end of the gradation, and build a mixture of a screen of 180 and solid PMS 116 until you get as close to the actual appearance of 180 as possible. What do you think about that Claudia? Or would it be better to just let it be 100 over 100?

Don
 
Re: Printing gradients with 2 Pantone colors

Well, since Claudia hasn't replied, you really have two choices:
1. Use noise to help banding problem.
2. Use the technique of gradient of one ink over the other ink solid.

With number 2, realize that even at the 180 side of the gradation that there is some of the other ink mixed in (since it is a gradient), and know that none of the gradient looks like PMS 180. Unless the PMS 180 end of the gradation is touching a solid PMS 180, then no one will likely know that the two are made up of two inks, and will just think it's a mighty nice gradation. Also note, that unless you have the "ghosting" problem in the middle of the gradation, then you won't need to use #2 option. Maybe I was wrong in posting it, since in this situation, yes option #1 is probably better.

Don
 
Re: Printing gradients with 2 Pantone colors

Well, the printer worked on it, since they know their equipment best. I found out that the printing was, and will be done digital, and have read digital can be even more difficult with regards to good gradations than traditional printing. But, they did a good job of making a satisfactory blend. We used a photoshop file with some noise and they took it from there. Thanks for your help and explanations. We have archived these techniques for future reference.
 
Re: Printing gradients with 2 Pantone colors

Well there is still the option of creating 2 gradients from 100%-0%, laying them on top of one another and either overprint or multiply. This used to be the OLNY way you could get 2 PMS gradients in PS level 2. Theoretically Indesign should work this way by default, but some rips still interpret blends differently so its worth a shot. My 2¢.
 
Re: Printing gradients with 2 Pantone colors

Why not just move the 2 centers of each gradients toward their respective zero points to make the gradient darker in the middle?

*On the banding problem*: this is a mathematical formula from adobe:

Creating smooth gradients for PostScript output 12/3/07 10:29 AM
http://kb.adobe.com/selfservice/viewContent.do?externalId=tn_3668&sliceId=2 Page 1 of 5
ID: tn_3668
Products affected
FreeHand
Adobe
TechNote
Creating smooth gradients for PostScript output
Issue
Sometimes when I print graduated fills from FreeHand I get large bands of
color, not smooth gradations.
Reason
Postscript actually creates gradations of color by combining small areas of
single color. In most print output these bands of color are so small that your
eye blends them together. But sometimes Postscript can cause these
bands if it does not have the correct information. However, there are steps
which you can take to reduce the bands so the gradation seems smooth.
Solutions
The simplest solution is to set File/Output Options/Maximum Color Steps to
256.
Further Information
It is possible to calculate the size of tint bands in a gradient (i.e., gradient
fill, whether radial, linear or logarithmic) to predetermine banding and
assure efficient gradients for PostScript output.
A correct understanding of the mathematics involved in creating smooth
gradients will provide the tools for producing the desired effect without
compromising quality.
A. Creating Gradations Digitally
B. Formula for Calculating Tint Bands in Gradients
C. Determining the Available Number of Tints& Band Size
D. Working with Process Colors
TechNote Details
Did this resolve your
problem?
Yes
No
300 characters
Send
Take a Survey
Take a survey about
Adobe's resources for
learning and problem
solving.
Type feedback here
Creating smooth gradients for PostScript output 12/3/07 10:29 AM
http://kb.adobe.com/selfservice/viewContent.do?externalId=tn_3668&sliceId=2 Page 2 of 5
E. Adjusting Known Variables to Reduce Banding
F. General Tips
A. Creating Gradients Digitally
Banding is a result of producing halftones digitally. It is possible to calculate
the band widths before imaging. These calculations will aid in creating
efficient gradients with the optimum number of steps to prevent noticeable
banding and optimize imaging time.
The results of the calculation will provide:
1. How many different tints the printer can print.
2. What range of those tints the color change can use.
3. Whether the resulting tint bands are fine enough to create the illusion
of a smooth gradient.
The following information is needed to calculate tint band widths:
1. The printer resolution in dots per inch (dpi).
2. The screen ruling in lines per inch (lpi).
3. The change in color between the beginning and ending tints in the
gradient described as a percentage.
4. The physical distance between the beginning and ending tints of the
gradient described in points.
B. Formula for Calculating Tint Bands in Gradients:
([Resolution(dpi)]/[Screen Ruling(lpi)] [Squared] +
1= [a]
Example:
A printer with a dpi of 600 printing a file with a Screen Ruling set at 85 lpi
would yield this equation:
600/85= 7.0588 (round off to 7) 7 squared= 49 + 1 =
50 [a] = initial number of tints available to be printed.
[a] x [amount of tint change in gradient] =
Example:
An object in FreeHand has a gradient set that starts at 90% black and ends
at 10% black. The change in the tint is 80%, and would yield this equation:
50 x 80% (or .8) = 40
Creating smooth gradients for PostScript output 12/3/07 10:29 AM
http://kb.adobe.com/selfservice/viewContent.do?externalId=tn_3668&sliceId=2 Page 3 of 5
is the actual number of tints available in the current tint change.
[length of the area to be filled] / = [c]
Example:
The object in FreeHand is a horizontal rectangle 4 inches wide (288 points)
by 1 inch (72 points) high, and the gradient fill above is applied at an angle
of 0 degrees (the fill is horizontal over the 4 inch distance) would yield this
equation:
288 / 40= 7.2
[c] is the size in points of each band, or how visible banding will be and the
minimum width of objects used to create the gradient.
C. Determining the Available Number of Tints & Band Size
The first part of the calculation determines exactly how many different tints
the printer can produce at the screen ruling chosen. A certain tint is based
on the size of the halftone dot the printer creates, which in turn is based on
the combination of printer resolution and screen ruling chosen. The higher
the resolution of the printer, the more tints it can potentially create.
1. Divide the printer resolution by the screen ruling chosen in setting up
the PPD.
2. Square the integer value (discard the remainder or decimal), then add
1 to the result. This is the approximate number of tints the printer is
capable of producing with these two variables.
3. If the number calculated for the initial number of tints available is
higher than 256, reduce that number to 256 because the PostScript
Level 1 graphics language cannot create more than 256 different tints
per ink. (Some PostScript Level 2 devices may allow more than 256
tints per ink.)
([Resolution (dpi)]/[Screen Ruling (lpi)])[Squared]+1 = [initial number
of tints available]
For example:
If printing a job at 150 lines per inch (lpi) at 1270 dots per inch (dpi),
this formula yields roughly 72 different tints available. If the resolution
is raised to 2540 dpi, then 287 different tints are available, but since
PostScript Level 1 is only capable of 256 tints, the number 256 must
be used.
4. Calculate the percentage of available grays or the actual number of
tints available.
Express the "percent change between start and end" as a decimal
(e.g., if the gradient goes from 20% to 80%, the tint change is 60% or
Creating smooth gradients for PostScript output 12/3/07 10:29 AM
http://kb.adobe.com/selfservice/viewContent.do?externalId=tn_3668&sliceId=2 Page 4 of 5
(e.g., if the gradient goes from 20% to 80%, the tint change is 60% or
.6).
[Initial number of tints available] x [amount of
tint change in gradient] = [actual number of
tints available]
Note: This step is necessary because a tint change of less than 100%
uses less than 100% of the total available range of tints (e.g., 60% of
256 tints, or 153 tints).
5. Measure the object in the direction of the gradient and convert that
length into points (there are 72 points in an inch).
6. Divide the length by the actual number of tints available to determine
the size of each tint band. This number reflects the exact width of any
banding that might occur in the gradient, and is the minimum width of
objects used to create the gradient.
[length of area to be filled]/[actual number of
tints available] = [size of each tint band]
For example: If the fill is 4 inches (or 288 points) long and has 153
tints available, then each tint band will about 1.9 points wide. If the
calculation specifies the band size at less than 1 point wide, banding
will not be visible. If between 1 and 2 points, this is adequate for most
circumstances. If over 2 points, it is up to the discretion of the
designer as to whether the quality will be suitable for the type of work
produced. In most cases, as long as there are 256 total tints
available, the human eye is less likely to perceive any banding once
the final work is printed.
When predetermining banding that may occur in process color
gradations, it is important to remember that four separate inks are
involved, each having its own degree of banding. It is best to
calculate banding for the ink that has the most change in color, and
the ink with the least change. This gives the complete spectrum of
what banding to expect. Just run the color change in each ink through
the formula above.
Heavy banding in a lighter color (i.e., yellow) will be less evident when
the final product is printed than banding in a darker color (i.e., black).
E. Adjusting Known Variables to Reduce Banding
The variables that effect banding are the four numbers needed to
predetermine banding: the device resolution, screen ruling, tint change in
the gradient, and the length of the gradient.
To adjust a gradient that is banding:
Creating smooth gradients for PostScript output 12/3/07 10:29 AM
http://kb.adobe.com/selfservice/viewContent.do?externalId=tn_3668&sliceId=2 Page 5 of 5
1. Determine the variables that can be altered:
a. Are 256 tints already available from the first equation? Raising
resolution or lowering screen ruling after passing 256 tints is
pointless because PostScript Level 1 can only produce 256 tints
per ink. If the 256 limit has already been reached, adjust the
remaining variables.
b. Does the nature of the job dictate a strict requirement for screen
ruling? If it cannot be changed, adjust other variables.
c. Adjust the remaining variables, adjusting first the variables that
have the least effect on the visual composition.
2. Adjusting variables to reduce banding:
a. Raise the resolution. Raise until the formula shows the gray bands
are fine enough or until the number of initially available grays
passes 256.
b. Lower the screen ruling. Lower the screen ruling value until the
formula shows the bands are fine enough or until the number of
initially available tints passes 256
c. Increase the difference between the starting and ending tints in the
gradient.
d. Decrease the length of the gradient.
F. General Tips
Do not throw in extra gradient steps to create a smoother effect. The
number of tints produced is related to the results of this formula. Enter only
the number of steps needed to match that result. Specifying a number
significantly more than the available tints is extra information that the printer
cannot use but still has to process (RIP), making the print time considerably
longer with no increase in quality.
TechNote Details
Last Update: 07-06-2006
ID: tn_3668
Permanent Link: http://www.adobe.com/go/tn_3668
Products Affected:
FreeHand
 
Re: Printing gradients with 2 Pantone colors

Be careful using the indesign method. Depending on the colors and the type of press this may not work very well.

If you have a common blanket press...this may not be the route to use, as laying 100% each of 2 PMS colors on top of each other might "muddy up" the color.

Edited by: Brent Baillie on Aug 7, 2008 5:38 PM
 
Re: Printing gradients with 2 Pantone colors

Hey Claudia,

Haven't run across you on the groups.
Great to see you.

Waiting for the intensive CS4 week long seminars to begin.

MSD
 

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