Non-Pantone spot colors - from a noob

[Plaza]

Hi all, noob question here: it is my understanding that if someone specifies a Pantone color for a spot, the printer knows how to mix up that ink, because he's got a recipe from Pantone. However, what if you specify a spot by using L*A*B values? How does the printer match it - does he make up a custom mix of ink through trial and error? Or does he find the closes match he can in Pantone and then use that, even if it isn't exactly what you had asked for.

 

[Sherbert]

http://printplanet.com/forums/color-management/15207-l-b-values-pantone-485-a

Sort of dicussed there.

 

[almaink]

Either use the pantone swatch book to find the nearest equivalent color or plug the LAB values into Photoshop's color picker to find a match.

 

[Plaza]

Thanks to both of you. Are you saying that printer wants/needs the Pantone number - to save him from trial and error ink mixing??

 

[Werby]

Lab

Lab

The problem with LAB as a way of specifying a color is that it doesn't really exist in the physical world.

For instance, how are you picking your LAB values? Based on what you see on your monitor? Are you sure your monitor is displaying that LAB value correctly? Even with a perfectly calibrated wide-gamut monitor, there are lots of LAB values outside the gamut. A given LAB value on your monitor will probably look different on a different monitor.

Or are you using a spectrophotometer to get LAB values off a printed piece or a real-world object? This is dependent on the kind of device you have and how well calibrated it is. Two different devices of different ages can give you pretty different numbers, so your LAB value for a physical sample might be significantly different from someone else's.

Or are you getting LAB values from a look-up table in your RIP (or photoshop or some other software) that gives you the LAB values of various flavors of spot colors like PANTONE or TOYO? If that is the case, you should just use the spot color designation, as that corresponds to something in the real world. Different software and rips can give you different LAB values for spot colors, as there is no "official" spot color to LAB table from PANTONE.

However flawed swatch books might be, they are REAL ink on REAL paper and they all look the pretty much the same from one swatch book to another, so if you specify a PANTONE or TOYO color, the printer (or the designer or whoever) can look to a physical sample to see what it is supposed to look like. Maybe they can or can't match it, but at least everybody knows what the goal is. Not so with LAB values.

 

[Plaza]

Or are you using a spectrophotometer to get LAB values off a printed piece or a real-world object? This is dependent on the kind of device you have and how well calibrated it is. Two different devices of different ages can give you pretty different numbers, so your LAB value for a physical sample might be significantly different from someone else's.

Yes, that is exactly what I am doing. With a well calibrated spectro. While I appreciate your input, I can't believe that two spectros could vary as much as two swatchbooks possibly/probably printed in different years.

And I can send the printer a swatch of the color to reference.

What I am trying to understand is how things work at the printers. Do they/will they only mix to Pantone colors (because they come with recipes) or will they/can they tackle custom mixes as well.

 

[Luc St-Pierre]

Many of my customers (flexo printers) will mix any feasable ink color from any sample submitted. They have so-called ink guys that are used to these custom mixes. They generally come under the 2 dE depending on the sample they are trying to match. In some occasions metamerism occur and the spectro is useless. They just can in the recipe when the match is done.

 

[Plaza]

Many of my customers (flexo printers) will mix any feasable ink color from any sample submitted. They have so-called ink guys that are used to these custom mixes. They generally come under the 2 dE depending on the sample they are trying to match. In some occasions metamerism occur and the spectro is useless. They just can in the recipe when the match is done.

Thanks for that information. Do you think it is something that should be dealt with on a case-by-case basis? In other words, I need to speak to each printing company and ask them if they will do custom mixes for non-Pantone colors?

 

[Luc St-Pierre]

Correct. This is the way I work.

 

[Plaza]

Thanks! For the printers who are willing to do custom mixes for non-Pantone colors - should I except this service to come at additional cost over and above the other costs associated with using spot colors?

 

[Luc St-Pierre]

No general rules here. Mixing is mixing. I often get away with the extra costs through good old crying and begging. The end-user that insists for some color to be matched out of the limb is the one I usually turn around the extra cost when there is an extra cost.

 

[Werby]

... I can't believe that two spectros could vary as much as two swatchbooks possibly/probably printed in different years.

And I can send the printer a swatch of the color to reference.

I hate to burst your bubble, but I've seen "well calibrated" spectro's disagree on certain patches by as much as 5 dE. My brand new i1 can have a dE 2 just between consecutive measurements of the same sample. And just because your spectro is well calibrated doesn't mean the printer's is. I'd trust a swatch book before I'd trust LAB numbers from two different spectros any day.

Of course all this is moot because you can supply a physical swatch, which solves the specification problem. But don't just give them the numbers and hope for a perfect match!

-Werby

 

[Plaza]

I hate to burst your bubble, but I've seen "well calibrated" spectro's disagree on certain patches by as much as 5 dE. My brand new i1 can have a dE 2 just between consecutive measurements of the same sample.

No offense, but if this is the case, then your unit, although it may be recently calibrated, is not well calibrated. The spec for variation for the i1 pro is 0.1 dE 94. Send it back.

 

[michaelejahn]

Color communication

Color communication

No offense, but if this is the case, then your unit, although it may be recently calibrated, is not well calibrated. The spec for variation for the i1 pro is 0.1 dE 94. Send it back.

HA HA HA HA - oh, now THAT was funny. I will make a guess that you do not have hundreds of i1 Pros to prove this to yourself.

1. Okay, lets say that somehow X-Rite could actually manufacture anything that would meet that specification, and you owned 200 i1 Pros. If you think they will all be able to measure the same patch identically with all 200 of these i1Pros and have them report identically, you may not fully understand how these things work - this is especially true of deep/darker colors.

2. If you want to ask someone to create special color for you, I would suggest that you may save a lot of time by specifying a Pantone color. Pantone Goe has a new straight forward method - 10 mixing bases are use to create 2,068 colors - and you can 'see' what these colors look like on coated and uncoated stock (if you buy the Goe guides)

3. Even if you send a Lab value to your vendor, they have no method to take that value and create that color onto something - there is no "Lab to color /substrate" print gadget, and even if there were, you do not have any hope of dealing with the many factors like paper color, opacity, metamerism, color temperature of the light - people tend to use spectophotometers to measure things to see if they fall within a tolerance, and they often do that by measuring two samples with the same spectrophotometer in rapid succession -m this is what they are designed to do - but not so much for 'communicating' color - Pantone is your best bet.

4. If for some odd reason you can't find your special color in that Pantone Goe Guide, then by all means, if you have a sample and were considering measuring that sample and sending that measurement, I would suggest that you send the sample you measured with (or instead) and do explain what you paper you will be printing on.

Okay, that is what I do, and others may do other things.

 

[michaelejahn]

just to pile on...

just to pile on...

CCInspector2

 

[rich apollo]

Plaza,
The printer will most likely send the sample to his ink vendor (unless he has an in-house installation). The vendor will measure and do some "trial and error". The final color determination will come from the draw-downs.

Yes, you will pay for the color matching. How much depends on a few things - 1) the billing practices of your printer, 2) how much is this job worth, 3) how much work do you (or your client) bring in to the printer, and 4) the ingredients of the ink. Spot colors vary WIDELY in price. You might get a yellow for $6/lb while a blue/purple might be $20/lb. If you aren't a valued customer and this is not a long run you may get "go-away pricing".

Also, if the color will coated, the ink has to be formulated differently.