Re: Spot Color tables in Best ColorProof
Re: Spot Color tables in Best ColorProof
- Subject: Re: Spot Color tables in Best ColorProof
- From: Roberto Michelena <email@hidden>
- Date: Sun, 01 Jun 2003 01:35:09 -0500
The problem lies in the gamut mapping built into the profiles.
Provided your proofer (meaning BestColor+printer+media) is well profiled,
the Lab value you ask for should be the Lab value you get, within a
tolerance (for example 2 dE), *if* it is within the gamut. That means, for
the in-gamut Pantones, filling their Lab values into Best's table should be
enough.
But if it is not within gamut, there comes the problem. Strategies for gamut
mapping differ from one profile building application to another, but
generally they are optimized to best retain the relationship between close
colors than the accuracy of each. In short, for continuous tone images, not
for spot colors.
Furthermore, the Lab values will usually work only for the "well within
gamut" colors, because even in the colorimetric intent, most profiles begin
to show the effects of gamut mapping (or should I say distortion) not at the
border but before. Understandably so, in order to keep a smooth transition.
Again, for the benefit of continuous tone images.
Even if you were using the "saturation" intent, the colors will not be as
close as they could, because depending on the color it would be preferable
to maintain hue than saturation; as usually happens in yellow.
Graeme Gill hints that his profiling routines do have a "keep hue" mapping
style or option or a "least dE" one, I don't recall which.
Probably the best solution would be for Best to implement an algorithm
similar to Praxisoft's ColorCompass Xtension for Quark, in which for spot
color it iterates Lab values in an out of the profile until it finds the
minimum possible dE. It is easy and fast given that you're looking for a
single point.
Meanwhile you could try a Photoshop trick in which you paint Lab gradients
around the value you want, then transform to the device profile and back to
Lab, and substract that resulting image from a flat tint of the Lab value
you wanted to achieve. After that, there should be some channel ops that
might transform this "dL-da-db" map into something resembling dE, and find
the lower value there, and by its position know which were the Lab values
(of the painted gradients) that produced it. Too late to think right now...
One problem is that you'll not be able to use Gretag's CMM (the one inside
Best), you'll have to resort to ACE or Apple's CMM and that will bring some
extra innacuracy.
Best regards,
-- Roberto Michelena
EOS S.A.C.
Lima, Peru
On 30/5/03 4:10 AM, "Marco Ugolini" <email@hidden> wrote:
>
Does anyone know of any existing software tool that allows one to profile
>
the full color gamut used by the Best ColorProof RIP in its separated spot
>
color function (for any of the inkjet printers it supports--say, the Epson
>
7600 or 9600 or 10000), and then chart the true L-A-B coordinates of a given
>
PANTONE color (as found in Photoshop 7's color swatch library) to the
>
colorimetrically closest L-A-B values within that profiled color space?
>
>
(Guessing the correct values for a device via manual measurements and
>
subsequent trial-and-error adjustments is insanely time-consuming, specially
>
when trying to build a whole viable library.)
>
>
I hope my question is clearly formulated. Please let me know if it's not.
_______________________________________________
colorsync-users mailing list | email@hidden
Help/Unsubscribe/Archives:
http://www.lists.apple.com/mailman/listinfo/colorsync-users
Do not post admin requests to the list. They will be ignored.