Spot Color Delta E

[ripit]

What would the industry standard tolerance (delta E) be for a printed spot color to the pantone book be? I have a drawdown from the press that is a delta E of 5 compared to the book, is this within spec? Would the same spec be used for process colors as well?

 

[gordo]

What would the industry standard tolerance (delta E) be for a printed spot color to the pantone book be? I have a drawdown from the press that is a delta E of 5 compared to the book, is this within spec? Would the same spec be used for process colors as well?

I don't believe there is a delta E tolerance for PMS colors. Even the latest Pantone GOE library does not mention tolerances or CIEL*a*b* values.
Typically swatches/draw downs are acompanied with a "Hi-Lo" density reference.

BTW a Delta E of 5 can be a good match as the visual perception of difference depends on the hue you're measuring and how the Delta is being calculated.

best, gordon p

my print blog here: Quality In Print

 

[dimitri]

There is no spec though for spot colors... only two questions: what is achievable by the press, and what is acceptable by the customer. You got to strike a balance there.

We've found that a DEcmc(2:1) of 2.00 overall and 1.50 for pastel colors , measured at D50/10 satisfy most of our customers and is within the variation of the presses, making it achievable.

A drawdown at 5.00 DEab could well have a value of 3.50 DEcmc(2:1) or any of the other formulae (DE94, DE2000)... but still, it seems a bit high to be coming from the ink labs (i assume a drawdown is done with the amazing device called the 'little joe' that services our industry so well). when you throw that color on press, the 5.00 DE could easily go 7ish, with the added press variation, or you might be lucky if it is only a lightness issue and you can actually decrease the color difference.

How old is the pantone book is another question.

 

[gordo]

Just a sidebar:
Not sure if it's a general truth, but I've never seen an ink room that uses a spectro for checking the CIEL*a*b* values of PMS mixes or drawdowns. It seems always done by weights and eye.

best, gordon p

 

[dimitri]

In packaging I would expect it to be the norm. Else, what? Too risky! Relying on subjective measurement without any documentation can be disastrous.

I know ink suppliers when formulating a spot color ink, they measure the DE of a drawdown, check whether it is light or not, or whether there is a hue difference, and then make formulation adjustments, or lightness adjustments to match a pms book or a chip. There is software like ColorMaster by X-Rite that makes this easy.

 

[arfamoe]

Our process is as follows. Ink room try and match PMS to a DE of 1 or less, we use DEcmc D50/10. this can be a challenge sometimes. Proofs or drawdowns are made using an IGT Orange Proofer which does a great job, we use both the litho and Flexo versions for pulling drawdowns. Press has a pass/fail target of 2.5 DE using the same X-Rite equipment the ink room use, Some colours of course dont follow Spectro rules too well but they are the exceptions.
We find this works very well. We are a Packaging plant and have around 800 specials both PMS and customer specific colour.

 

[Luc St-Pierre]

If you want to create your own answer to this question, scan for deltaE your Pantone swatchbook. You will be amazed. My PMS 021 swatch scans with a deltaE of 12, and my swatchbook is brand new and bought from Pantone. Correct me if I am wrong but Pantone does not publish Lab or Lch values for their colors, I suspect because they would then have to produce their own product line on spec ($$$$$).

 

[dimitri]

some variation is to be expected... even in the stock they are using. They are after all printers. But 12 deltaE is too much.

But to their defense: they do publish the LABs of their swatches on their digital libraries.... isn't that something?

 

[Luc St-Pierre]

Where can those LAB values be found?

 

[dimitri]

published LAB values?

published LAB values?

I guess 'publish' is not a right way to put it, and you also got me questioning myself.

However, if you setup photoshop in LAB mode and pick a color from the pantone library, then it would provide you with an LAB value. I was not sure if that LAB value (even if it is meant to be device independent) was somehow tied up to the CMYK profile loaded in your color settings. So I made this small exercise as seen in the PDF.

The image on the left is with the GRACOL2006 profile, and the image to the right with the EuroscaleUncoated. Point sampler #1 has the LAB value of Pantone Purple, that is not reproducible in any of the color spaces. As you can see, the LAB value is the same with both profiles, and as such device independent: someone needs to have known and provided this LAB value, and it better be Pantone... unless I was wrong and each third party application uses some arbitrary LAB value...

Point samplers #2 and #4 refer to switching to Color Picker mode and hitting the exclamation mark. The LAB and associated CMYK values are changing based on the CMYK profile.

So, I do not think anyone has published a book/manual/ with all the LAB values of all the pantone books, but these values must somehow be made known to third party applications... or not?

-D

 

[Luc St-Pierre]

I believed that too . . . but third parties may have different LAB values. Leads me to think that no Pantone official LAB values exist. If it does exist, please give me a reference.

 

[gordo]

The CIEL*a*b* values for PMS colors are available to extract from most(?) workflows as a text/excel/spreadsheet file. Check your documentation/ask your workflow vendor.

best, gordo

my print blog here: Quality In Print

 

[Lukas Engqvist]

(Exclamation marks in PS are to show out of Gamut colours)

 

[encompus]

5 is a bit high but then again if the client is happy with the match..... Like Gordon said, there is no industry spec especially when it comes to which Formula everyone should use. We still use DE2000. I always take my i1 and Laptop to Press Checks. We first agree along with the Art Director and/or Creative Director that it's a good match visually. We then measure just to know DE. We hover from DE 2-6. Lately, we have had problems with our printer mixing ink to match Goe colors since PANTONE changed the base inks. In fact we had to bring our own Goe Uncoated Swatch Book because our local printer had none.

 

[prepress@labels]

Eye One Share

Eye One Share

I have recently begun using the iShare software with a photospectrometer to evaluate color at press. Color has always been approved by a couple of pairs of eyes and a Pantone swatch.

I ran into an instance recently where the Pantone Coated Library that comes with the software was way off from the Swatch book. It gave me a DeltaE value of 12.

Our swatches are less than 6 months old and I tried two different books to be certain. They both read 12.

Anyone have any idea why the Pantone Library in the software wouldn't match the swatch?

 

[gordo]

I ran into an instance recently where the Pantone Coated Library that comes with the software was way off from the Swatch book. It gave me a DeltaE value of 12.
Our swatches are less than 6 months old and I tried two different books to be certain. They both read 12.
Anyone have any idea why the Pantone Library in the software wouldn't match the swatch?

There can be many reasons.
Remember that, AFAIK (working with Pantone techies), the CIEL*a*b* values (in RIPs and instruments) associated with Pantone colors are a courtesy to enable better CMYK simulations or to determine the closest PMS color to an unknown sample. The PMS library/swatchbook itself is not based on CIEL*a*b* values, but on recipes using X amount of the base colors. That's why there are no CIEL*a*b* values printed on the swatchbooks themselves.
That's also why there is no tolerance/DeltaE for the color match. It's a recipe book.

best, gordon p
my print blog here: Quality In Print current topic: the creative design/production process

 

[michaelejahn]

Let me say this about that.

1. Gordon is 'spot on' (please excuse the pun) - Pantone sells a book and a system to mix inks so you can order special colors. They have never - as far as i know - ever published or licensed LAB values of the Pantone color library.

2. using a single 'delta E differences' is not used when approving or rejecting for color match as a this shifts depending on hue. For example - differences in the center of the spectrum (example yellow) are very hard for the eye to perceive - one can not really see shifts in yellow that are as high as 10. In other hues near the edge of the spectrum, even a difference of 2 is noticeable.

3. To make matters worse, the formula that are used in calculating these difference is quite a tangled mess. We try to help people understand this using a small tool we built that you can view here;

IQ Colour, Inc.

if you want more details on this tool - visit here;

CIE Standards Comparison Tool...

4. I happen to think spectrophotometer are useful when comparing two color samples, and happen to think LAB color descriptions are very useful for that. I do not happen to think LAB is a very good color transformation space, nor is it very useful to use for Pantone colors in the first place, so I would not actually welcome anyone creating a "Pantone LAB guide book"

here is a small example of one colors LAB value differences that I would hope helps this argument;

Pantone 151 PC - L72 a32 b69
Pantone 151 C - L67 a51 b80
Pantone 151 M - L67 a56 b80
Pantone 151 U - L72 a28 b61

So, we have one single color (Pantone 151) in 4 (four) different swatch libraries in one single third party application (Adobe Photoshop) - there are MANY more. And before we go with the "well, of COURSE they are different, the print condition and paper changes, so the color will shift too!" I would like to point out that since most people view in D50 light booths - and most spectrophotometers using a D65 light source (or use an 'assumption' that the whitepoint is D65) - well, I just am not sure I would line up behind any LAB value of a pantone color in the first place as 'standard' for most printing conditions that are encountered in our industry - except for the printing presses that are used at pantone, which are quite different than most everyone elses !

 

[meddington]

I would like to point out that since most people view in D50 light booths - and most spectrophotometers using a D65 light source (or use an 'assumption' that the whitepoint is D65) - well, I just am not sure I would line up behind any LAB value of a pantone color in the first place as 'standard' for most printing conditions that are encountered in our industry - except for the printing presses that are used at pantone, which are quite different than most everyone elses !

Not to argue your overlying point, but where would you get your assumption that most spectrophotometers are using D65 for the whitepoint?

 

[dimitri]

I guess there is none. Using 'common sense' I thought that they wouldn't differ from application to application.

Do the third parties create their own LAB libraries or does PANTONE provide each one with different values would be one question. The second, you would expect that when talking for visual appearance you care about the end result, the color (in a standardized viewing environment); if we were to assume that each ink manufacturer has the same spectral properties in their bases they use, then you could, possibly, expect to get the same color every time.... otherwise we are talking about secret recipes.

I cannot see how this helps the industry, other than providing a starting point for customer-client communication.

 

[gordo]

To Dimitri:

The CIEL*a*b* values likely change from application to application because Pantone updates the values from time to time (paper changes, pigments may be reformulated etc.). AFAIK third parties receive the CIEL*a*b* values from Pantone and they may be slight differences depending on which revision of the library they receive.
There are no "secret" recipes. The recipes for creating each Pantone spot color, as you know, is included with each swatchbook. As I wrote before, the inclusion of CIEL*a*b* values in, for example RIPs, is AFAIK, a courtesy to allow for simple conversion from specified spot colors to CMYK simulations in cases where the creatives simply use a drop-down menu to specify a color - i.e. they don't specifically want PMS 354 - they just want a green something like PMS 354.
Ink suppliers are supposed to make up the PMS spot colors using inks that meet Pantone's standards - i.e. the Pantone base inks. If substitute pigments are used then it is the ink supplier's responsibility to contact Pantone for technical assistance.
If you want to formulate a Pantone color so that it appears the same on different colored substrates then you are in the realm of experimentation and you will need to be sure that is what your customer actually wants as well as whether it is technically possible (as it may not be).

best, gordon p

my print blog here: Quality In Print current topic: Halftone screen angles