Standard Finishing
4Over

Announcement

Collapse
No announcement yet.

Tints of L*a*b* colours

Collapse
X
  • Filter
  • Time
  • Show
Clear All
new posts

  • Tints of L*a*b* colours

    Here's a question for you colour-savvys out there. If one is dealing with a LAB colour (or, you've measured a Pantone colour), how could one determine what a tint of that colour would measure in LAB, without having to actually print it out. Obviously depends on your curves, substrate etc., but, are there any thoughts on this?

  • #2
    Re: Tints of L*a*b* colours

    Can you create a image in ps and fill with % of lab color and then read the values. Is that what your looking for?

    d

    Comment


    • #3
      Re: Tints of L*a*b* colours

      > If one is dealing with a LAB colour (or, you've measured a Pantone colour), how could one determine what a tint of that colour would measure in LAB, without having to actually print it out.

      For solids, you could use a round trip (Lab->CMYK->Lab) through an ICC profile using absolute rendering. but for tints, well, there's a problem there. Pantone does publish tint books, but I don't trust them as they're all over the place for gains. , though you could expect similar Lab values for simlar tints using simlar substrates.

      Other than that, this is kind of a black art. We've run into this problem from a proofing perspective and the best results, short of a press test of the spot color with tone scales, came from dialing in the solid to the closest delta E and then ensuring a dot gain curve representative of the printing is applied...assuming 20% gain for example. We're using GMG for proofing and spot color accuracy, including tints, come out fantastic with this method. A few curve balls though...gamut limitation of the proofer of course, and when tints mix with process colors, things get tricky, but this is still the best approach we've used, short of multicolor profiling.

      Comment


      • #4
        Re: Tints of L*a*b* colours

        >create a image in ps and fill with % of lab color and then read the values

        Or you could do this, which is a bit more elegant a solution. However, keep in mind that if you want the Lab at 50% as it will print, you'll have to account for dot gain...so instead of cutting back to 50%, cut to 70% for 20% more gain. I'd still feel more comfortable with a proof though, despite possible gamut limitations.

        Comment


        • #5
          Re: Tints of L*a*b* colours

          Hmm... okay. What, specifically, should one understand a tint of a Lab colour to be?

          If I had a greenish Lab colour, 50 -40 30, and I told a program such as Indy that I want a tint of it... what then? But I understand your response, take it into CMYK, tint that, then back into Lab. Theoretically, though, does it make any sence even talking of an Lab tint?

          Comment


          • #6
            Re: Tints of L*a*b* colours

            >If I had a greenish Lab colour, 50 -40 30, and I told a program such as Indy that I want a tint of it... what then?

            I guess this depends on how you create that color. If your creating it as process, but defining by Lab, there should be a "conversion" to the working space CMYK. If you creating it as a spot, then the Lab that it should be treated as a spot.

            This is easier (for me) to visualize in Photoshop...where you fill an image with a Lab value, then curve it back 50% and render a "new" lab value.

            >Theoretically, though, does it make any sence even talking of an Lab tint?

            stop...your freakin' me out.

            Comment


            • #7
              Re: Tints of L*a*b* colours

              > stop...your freakin' me out.

              Sorry 'bout that. (It's the end of the day for me here in sunny SA, so this is getting a bit heavy for me too. )

              I was just intrigued about how the idea of a tint doesn't apply to pure device independent colour. A tint is necessarily a product of halftoning, so is device +dependent+. It can't be figured without a trip via CMYK.

              Do you know what I mean?

              Comment


              • #8
                Re: Tints of L*a*b* colours

                >I was just intrigued about how the idea of a tint doesn't apply to pure device independent colour. A tint is necessarily a product of halftoning, so is device dependent.

                Well said. I feel a little better now.

                Comment

                UltimateDuploSmartsoft (Presswise)Standard FinishingKBA4Over
                UltimateDuploStandard FinishingKBAKBA GTIChili publish4Over

                What's Going On

                Collapse

                There are currently 4584 users online. 108 members and 4476 guests.

                Most users ever online was 6,151 at 11:36 AM on 11-17-2017.

                Working...
                X