Using i1 share > making LAB spot colors > making them work in CB

[Tovis]

So, after much reading I found out that CMYK and RGB are both Device Dependent Colors. Pantones and LAB are both Device Independent Colors. I've been experimenting with using my i1 in i1 share to find the LAB value, converting that to a spot color with lab in illustrator and running that color in colorburst.

My goal is to be able to scan something like a Benjamin Moore color and have our printers print that color.

So far its working with lighter and darker colors, but the the colors in the middle need some help.

Has anybody used this method?

So far I've profiled each substrate with the i1, and pantones and photos look beautiful. I just want to take the guesswork out of colors we have to reproduce from a sample.

 

[eye4clr]

I'm curious what rendering intent you're using during conversion in your RIP? In theory it should be relative colorimetric.

Also, how are you naming the spot color and does that name match up with a spot color reference in your RIP's spot color table? Most of the RIPs have a table that lists the spot colors by name and you have to have a name match to get the connection. Sometimes this is even case sensitive. Sometimes they let you put in custom spot names to help you keep track of things. If not, you may have to tweak the LAB of an existing spot within your RIP to get the right results.

 

[Rooster]

Define needs some help.

Is there anything consistent about the inaccuracies you mention?

 

[Tovis]

yes, I am using relative colorimetric.

i believe the when the rip finds a spot color it runs that particular color in absolute.

in illustrator I set the color as a spot color, select lab, and enter the lab values

the colorburst manual says it reads the values i enter from the postscrip file and puts it in, but perhaps I should test going into the rip and entering those values.

needs some help means the colors did not match up and are not going to be acceptable.

 

[eye4clr]

the colorburst manual says it reads the values i enter from the postscrip file and puts it in, but perhaps I should test going into the rip and entering those values.

I'm curious if it "reads" the lab values or the alternative RGB or CMYK values Illustrator puts in the postscript according to the color format of the overall file. The RIPs i'm familial with will not read the lab but either recognize the name and use the RIP's own internal lab reference or don't recognize and use the alternate RGB or CMYK values.

 

[Tovis]

the rip says it picks up the color in status.

i guess a way that I could test this would be to do spot colors and non spot colors next to them. if they don't match then i'll have to insert the lab values into the rip.

 

[eye4clr]

the rip says it picks up the color in status.

I don't know what this means.

i guess a way that I could test this would be to do spot colors and non spot colors next to them. if they don't match then i'll have to insert the lab values into the rip.

If i'm thinking right, they should NOT match. The CMYK or RGB would be limited by the working space and the spot would not. Can't say i'm confident in this, but that's my first thought.

Maybe the test is to do the lab in the RIP and test that against the spot performance in your file?
Or! make a custom spot in the rip that has a totally jacked lab value from that in the file. Maybe make the file's lab cyan and the rip's magenta. Then see how it comes out. If the rip reads the file's lab, you'll get cyan, if not you'll get magenta.

 

[Rooster]

needs some help means the colors did not match up and are not going to be acceptable.

This where I was looking for information about whether there's anything consistent with the inaccuracies. Like do you see a consistent move to cool or warm in the color that would indicate an overall color cast.

If you print an RGB grayscale, is it perfectly neutral across the entire tonal range?

How many target patches are you reading to create your profile?

 

[Tovis]

in the mids I don't see a particular bias with the prints from the scans, also grays are very neutral, that might be because i have gcr set to 100.

I wonder if it has something to do with i1 share.

 

[Rooster]

Even with GCR set to 100 it's probably still making up a considerable part of the grays with CMY. Unless you've set your black generation to start at 0% which very few do. The neutrality is encouraging. Rules out a number of possibilities anyhow.

How about tonality. Are they consistently lighter or darker than the samples?

Also, what kind of samples have you been reading with the i1 and iShare? Have they been a flat smooth surface to get a reading from. Or something textured where you may be getting highlights and shadows in your sample spot area?

Try this. Load your output profile into the profiles folder in photoshop and set it as your default CMYK profile. Then under view/proof setup/custom select it as the proof profile with the relative colorimetric intent.

You should now be able to create the new color based on the lab specs in photoshop (to the nearest whole numbers anyhow, stupid photoshop!). If you see a little yellow triangle with an exclamation point inside next to the color you've created. It's telling you that the color is outside the gamut of that printer profile. With the new color window still open you should be able to go back under the view menu and turn on gamut warning. It will gray out everything in the color selector window that's outside the gamut and show you just how far outside that profile gamut the color is.

Being able to determine we're dealing with an in-gamut color will prevent us from chasing a color you just can't hit.

 

[slangsc]

When using Wasatch Rip and Profilemaker, I measure the color chip, and then enter my lab values into Wasatch spot color replacement. Most times colors look better if I use the lab values and have Profilemaker calculate a device CMYK color for me and I enter that into the rip. Either way I'm at the mercy of the gamut of the profile, but this has resulted in closer, cleaner color matching.

 

[sjm]

I'm curious what rendering intent you're using during conversion in your RIP? In theory it should be relative colorimetric.

Also, how are you naming the spot color and does that name match up with a spot color reference in your RIP's spot color table? Most of the RIPs have a table that lists the spot colors by name and you have to have a name match to get the connection. Sometimes this is even case sensitive. Sometimes they let you put in custom spot names to help you keep track of things. If not, you may have to tweak the LAB of an existing spot within your RIP to get the right results.

Pantone Colour look-up tables in a rip by pass profiling. i-share offers a best match against various output profiles.

A bit of code from an Illustrator generated file.

%%DocumentCustomColors: (PANTONE 485 C)
%%CMYKCustomColor: 0 0.9500 1 0 (PANTONE 485 C)
%%RGBCustomColor:

 

[rfulford]

I used to use ColorBurst quite a bit but it has been at least 2 years since I have touched the pro version. Consequently, my terminology might be a bit rusty.

My preferred method was to use device color for my custom spots. I would read a swatch with ProfileMaker's Color Picker application and use the device/media profile to calculate a match using the device color. There are a few ways to add the spots to ColorBurst. You can go into the spot color dialogue and enter the name of the spot and the color values you want to use. Also you can just open the .spt file directly and add to it. Last, you can just create a file from scratch with data exported from your color management app. (Not sure what your options are here) Whatever you do, be sure your environments are using the right .spt file and that you click, save and over write the existing spot file as you edit it. It can get confusing.

The disadvantage of using device color for the spot file is all the separate color libraries to maintain. The advantage for me was that I could use ColorPicker to adjust colors if I did not like the mix provided by my icc profile. Additionally, I found it easier to tweak my replacements over time using device color.

If done correctly, there should not be any need to tweek the spot color in the dtp app. Just be sure it is named exactly the same as the name in your .spt file. Actually, you can assign a color that is easy to identify if you want to test the replacement. Just pick something obviously wrong. Watch the console as you rip and look for the messages about "spots found" and "replacing?" Again, I do not recall much here but the console is pretty verbose

 

[eye4clr]

Pantone Colour look-up tables in a rip by pass profiling. i-share offers a best match against various output profiles.

A bit of code from an Illustrator generated file.

%%DocumentCustomColors: (PANTONE 485 C)
%%CMYKCustomColor: 0 0.9500 1 0 (PANTONE 485 C)
%%RGBCustomColor:

Both work the same way, they use the LAB->CMYK conversion provided by the printer icc. True they bypass any sort of input profile since it is not a true source->destination conversion like the typical RGB->LAB->CMYK or CMYK->LAB->CMYK. But they both certainly use the output icc.

Each workflow is going to give different numbers because the RIPs and i-share use different "calculators" to do the conversions. What i've seen is if one is off they both tend to be off. I find the more out of gamut the color the more they tend to differ as well.

 

[sjm]

Both work the same way, they use the LAB->CMYK conversion provided by the printer icc. True they bypass any sort of input profile since it is not a true source->destination conversion like the typical RGB->LAB->CMYK or CMYK->LAB->CMYK. But they both certainly use the output icc.

Each workflow is going to give different numbers because the RIPs and i-share use different "calculators" to do the conversions. What i've seen is if one is off they both tend to be off. I find the more out of gamut the color the more they tend to differ as well.

Sorry the Pantone Name remains unchanged. This value will

%%CMYKCustomColor: 0 0.9500 1 0

 

[rfulford]

Sorry the Pantone Name remains unchanged. This value will

%%CMYKCustomColor: 0 0.9500 1 0

SJM, Eye4Color is right. In a case where the rip has spot color replacement and it is properly configured, named colors defined as spots are identified by the postcript interpreter and replaced with the value from the spot color table. Furthermore, if the replacement if a lab value, device color is calculated based on the icc profile for the environment using the absolute colormetric rendering intent.

 

[sjm]

SJM, Eye4Color is right. In a case where the rip has spot color replacement and it is properly configured, named colors defined as spots are identified by the postcript interpreter and replaced with the value from the spot color table. Furthermore, if the replacement if a lab value, device color is calculated based on the icc profile for the environment using the absolute colormetric rendering intent.

rfulford,

Pantone Colour Libraries are supplied as LAB colours, agreed. Other colours depend on the rendering intent setup on the rip. Perceptual, Relative, Absolute or Saturation. The rendering intent presumes you are doing Color Management on the RIP and not via the host.

 

[rfulford]

rfulford,

Pantone Colour Libraries are supplied as LAB colours, agreed. Other colours depend on the rendering intent setup on the rip. Perceptual, Relative, Absolute or Saturation. The rendering intent presumes you are doing Color Management on the RIP and not via the host.

Yes, I agree and I see that we are touching on two methods to the same end. Tovis's method: Enter the lab values in illustrator and pass the color along to the rip without adding to the spot color entry.

Spot replacement method: Name the color in illustrator and add the name and color values to the spot table in colorburst.

 

[eye4clr]

%%DocumentCustomColors: (PANTONE 485 C)
%%CMYKCustomColor: 0 0.9500 1 0 (PANTONE 485 C)
%%RGBCustomColor:

The RIP reads through the postscript from top to bottom. If it matches the DocumentCustomColors name to the built in library it uses the LAB reference it has for that color. If not, it flows to the next row(s) to find and alternative. This would use the rendering intent as setup in the color management policies in the RIP.

 

[sjm]

The RIP reads through the postscript from top to bottom. If it matches the DocumentCustomColors name to the built in library it uses the LAB reference it has for that color. If not, it flows to the next row(s) to find and alternative. This would use the rendering intent as setup in the color management policies in the RIP.

Assuming colour management is done in the RIP. MicroSoft has ICM and Apple ColorSync.