RE: Ink limiting on Ultrachrome
RE: Ink limiting on Ultrachrome
- Subject: RE: Ink limiting on Ultrachrome
- From: "Abe Hayhurst" <email@hidden>
- Date: Thu, 27 Feb 2003 09:58:33 -0800
- Organization: We Do Graphics Inc.
Hi Thomas,
I would recommend one of two methods for determing what percentage to
limit each ink channel to on any almost any device (including the Epson
7600/9600). I sometimes compare results between both methods. If I
arrive at the same conclusion using both methods, I know that I have a
good answer as to where to limit the ink.
Method 1
Measure tint patches output on the device. Look at the spectral
reflectance curve or spectral reflectance values. Find the wavelength
(550 nm for example) that is consistently the lowest on the reflectance
curve/values for each tint patch. As the percentage of ink increases,
the reflectance will decrease at this wavelength. There will be a point
where the reflectance at this wavelength is not decreasing any more and
may increase. Any additional ink layed down beyond this point will not
be increasing the absorbtion of light in the low area of the reflectance
curve, but will mainly be decreasing the reflectance of light in the
high areas of the reflectance curve, thus causing a hue shift. Even most
"black" pigments and dyes have a low point on their reflectance curve.
Method 2
Measure tint patches output on the device. Look at the LAB and LCH
values for each tint patch. For any chromatic ink (C, M, Y, O, G, etc.)
find the tint patch with the maximum C value. If the Chroma increases,
then decreases, and increass as you move your way up the tint values,
use the first (lowest) percentage of ink that produces the maximum
Chroma value. For any achromatic ink (K or light black/gray) find the
tint patch with the lowest L value, preferably that does not shift H
values or have reverse A and/or B values. For example on the Epson 9600
with Ultrachrome inks, the black starts out with a yellow/orange cast in
lighter tint values, becomes more red/magenta in the midtones and darker
tints. There is a point where the black makes a radical shift in hue
from a magenta/purple cast to a significantly blue cast. This also
corresponds to the point where the lowest reflectance value on the
spectral reflectance curve does not go any lower in method 1, resulting
a significant hue shift.
In my opinion, the goal of ink limiting (besides preventing bleeding and
smudging) is to make the device behave in a way that predictable and
easily characterizable (is that a word?) to the profiling software.
Ideally every ink should decrease in luminance and increase in chroma as
the percentages of ink that are layed down increase. I have yet to see a
combination of device, ink, and paper that if ink limited as described
above can not handle a Total Ink Limit of 300%.
I think it is worth mentioning that if you are using BestColor, I
believe you get significantly better results limiting the ink channels
as described above, NOT linearizing them with BestColor, and NOT using
the black generation and ink limiting in BestColor and instead setting
your total ink limit in your profile.
Good luck,
Abe Hayhurst
We Do Graphics Inc.
_______________________________________________
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.