Dan Margulis Applied Color Theory

Crazy Gray Ramp with Custom CMYK

Crazy Gray Ramp with Custom CMYK dot gain curves?
    Posted by: "George Machen"
    Date: Mon Jul 10, 2006 4:29 am (PDT)

I'm having printed some wedding invitations on thick uncoated paper, presumably with dot gain close to that of newsprint, e.g., 30%.

So I tried making custom dot gain curves in Custom CMYK like the ones in Dan M.'s Professional Photohop 7 (2002), p.291.

Whereas a normal dot gain compensation curve of such a large move at the 50% control point insufficiently handles more apparent gain in the quarter-to-midtone range - especially in magenta - and plugs the shadows, Dan's curves raise at the 30% control point, and lower & staighten in the threequartertone-to-shadow range.

After doing this, I noticed that the Gray Ramp showed an enormous height of the cyan curve, towering over the other three. Yow! Is this right, or is the Gray Ramp lying to me?

I had made similar curves for all four, with the highest move in black, the center average move in cyan, halfway in-between the latter two in magenta, and the lowest in yellow. All four moves were no more than six percentage points from each other, not enough, I would think, to put the cyan Gray Ramp so high.

I know that the weak cyan ink has to be higher, but this seems ridiculous, and in any event much higher than the cyan over the others in all the other normal or canned dot gain curves.

Maybe I just don't understand what the Gray Ramp portrays, but can anyone provide me some insight?

George Machen
________________________________________________________________________

Re: Crazy Gray Ramp with Custom CMYK dot gain curves?
    Posted by: Stephen Marsh
    Date: Mon Jul 10, 2006 5:09 pm (PDT)

George Machen wrote:

I know that the weak cyan ink has to be higher, but this seems
ridiculous, and in any event much higher than the cyan over the others
in all the other normal or canned dot gain curves.

Maybe I just don't understand what the Gray Ramp portrays, but can
anyone provide me some insight?

George, forget that for now and just accept it - when you go to convert  or after the conversion with this custom setting, run your cursor over the images neutral areas (or use a neutral test image like a gray gradient/step wedge if the real image is hard to evaluate) with either RGB or LAB info palette readings. If R=G=B and AB in LAB are 0 (zero) then the separation is neutral (as far as Photoshop is concerned, the actual press/stock/inks etc may not be close to this though).

There are "proper" newsprint profiles to be found, so you could also compare or use them as an alternative.

Stephen Marsh.
___________________________________________________________________________

Re: Crazy Gray Ramp with Custom CMYK dot gain curves?
    Posted by: Dan Margulis
    Date: Mon Jul 10, 2006 5:53 pm (PDT)

George writes,

Whereas a normal dot gain compensation curve of such a large move at
the 50% control point insufficiently handles more apparent gain in the
quarter-to-midtone range - especially in magenta - and plugs the
shadows, Dan's curves raise at the 30% control point, and lower &
staighten in the threequartertone-to-shadow range.
After doing this, I noticed that the Gray Ramp showed an enormous
height of the cyan curve, towering over the other three. Yow! Is this
right, or is the Gray Ramp lying to me?

Under these conditions--which are newsprint-specific, and you aren't using newsprint--yes, the cyan should be much higher than the magenta in the midtone range. Because newsprint is second cousin to toilet paper, it doesn't support a dark shadow--the ink just bleeds into the paper. But it does have heavy dot gain. Under these circumstances, yes, you need a bigger cyan-to-magenta ratio to make a midtone gray.

You, however, are not using newsprint, just an uncoated paper. Sure, it has high dot gain, maybe as high as 30%, probably not. But there's no reason to put an eccentric curve on it as if it were newsprint. Just open the cyan curve, set 50%=78%, click All Same, then go in and raise black to 50%=80%. That ought to get you fairly close.

Dan Margulis
___________________________________________________________________________

Re: Crazy Gray Ramp with Custom CMYK dot gain curves?
    Posted by: remaleydan
    Date: Tue Jul 11, 2006 5:19 am (PDT)

In the R-G-B world, equal amounts (=) neutral (without a 'cast'). In the
C-M-Y world equal amounts (=) BROWN!
So we always ned more Cyan than  Magenta and Yellow (Mag. & Yello are
usually equal in value)
Example: SWOP specified (in the past) 1/4 tone - 25%C 16M 16Y 1/2 tone -
50%C 39M 39Y 3/4 tone 75%C 63M 63Y.
These values should represent 25K - 50K - 75K.
Add 0% and 100% and you have a 5-point curve for gray balance, throughout
the tone scale. These numbers are for 'white' paper not Newsprint, their gains are the same for all colors (30%).
Their inks are a different color, likewise the paper.
Now here is the "classic secret" - create a plate - (or film) curve to
print, that will alow the uncoated to print with midtone gains like coated (18Y-29M-20C-22K) use the densities for uncoated
.95 Y 1.00C 1.12M 1.25K (these are 'dry' densities). Results- your uncoated will look gorgeous - remember
-almost no one seperates for 'uncoated' so a coated 'scan' will look GREAT!

Dan Remaley
Process control Manager
412.259.1814
___________________________________________________________________________

Re: Crazy Gray Ramp with Custom CMYK dot gain curves?
    Posted by: "Henry"
    Date: Tue Jul 11, 2006 11:00 am (PDT)

On Jul 11, 2006, at 9:28 AM, Dan Remaley wrote:

Whoops . . . Typo . . . . 18 Yellow - 20 Magenta - 20 Cyan - 22 Black
. .
.midtone dot gain for SWOP.

Dan,

I am glad you pointed this out.  There is usually some questioning/confusion expressed when advising that separations destined for uncoated should be made same as coated.  It's as if you are advising something that sounds inherently wrong, until they see the results.

Henry Davis