Re: Black generation questions for large format inkjet
Re: Black generation questions for large format inkjet
- Subject: Re: Black generation questions for large format inkjet
- From: email@hidden
- Date: Sat, 21 Apr 2007 00:22:13 -0400
On 20-Apr-2007 Rich Apollo writes:
.
.
.
[Boll}
>>This delayed onset creates a very steep rise
>>in the 3/4 tones & shadows, a tonal region where
>>ink gradients should be lessening not increasing.
[Apollo]
>Which is where I need it. My proofer doesn't have light
>black and has pretty large droplets, so the idea is to
>use the black in areas where it doesn't look like pepper.
Understood, you want these large droplets to start in the presence of large
amounts of CMY so it masks their gross appearance.
> I'm sorry, Harold. I don't know what you mean when you say
> "colorimetric reversal" or "Monotonic behavior of all inks
> on neutral" or "forced reversal of the chromatic inks".
> Could you clarify, please?
OK, but let me start by reviewing some basic principles. GCR is an acronym
that stands for Gray Component Replacement which pertains to the fact that
their is a colorimetric equivalency between K and some mix of CMY inks.
That is to say, for a given Lab color, you can remove some amount of CMY
and replace it with some amount of K and the overall Lab color remains the
same in appearance or if measured. In principle, any in-gamut Lab color
has a range of inkings which yields the same color. Any color will have a
max-K solution and min-K solution. On neutral, a min-K solution would have
K=0 and create the neutral with only CMY; while a max-K solution would
have mostly K and a very small amount of CMY to coax it to appear neutral.
This range also implies that there will be an intermediate inking in which
the % value for K is approximately equal to individual CMY values. The
lesson here is the more K you use the less CMY you need and, of course,
vice-versal.
On the neutral axis its important that all 4 CMYK inks are always
increasing (monotonic behavior); the existence of reversals in the 3/4
tones or shadows can create visual artifacts like banding, shimmering and
loss of shadow detail. Reversals mean that some inks are going up while
others are going down in a region of color space where the gamut is
tapering - this mix of effects come together to increase the probablility
of these artifacts appearing. You'd rather have inks that are monotonic
with shallow gradients in dark tonal regions, so that any variations in the
printing system are more forgiving.
A late onset of K at 50% or 60% of black means that up to that point, the
neutral axis inkings consists exclusively of CMY inks. At that point, K
suddenly starts and increases rapidly (meaning steep gradients). Recall
the GCR principle, the more K used, the less CMY needed; K increases so
quickly that CMY begins to decrease in value relative to what it had been
before K onset. This means that colorimetrically the inkings should appear
neutral because of the principle of GCR, but real-world production needs
suffer because of this forced reversal in CMY inks due to the steep K
gradients as described previously.
> I'm not using this profile to create seps - this is only for proofing.
Doesn't matter - these "cardinal rules on neutral" that I've espoused apply
to either half-tone proofers or ink-jets. It pertains to generally
desireable profile characteristics.
> I don't need any of the benefits offered by GCR.
Any profile for any CMYK system has to deal with GCR. It relates to the
overall black generation strategy employed. The principle of GCR requires
that K be specified independently; once K is specified, CMY is then
computed to achieve the desired color. Its valid to speak of Heavy, Medium
& Light GCR which is reflective of where in the range of min to max K
solutions a Profile was built on. To speak of UCR versus GCR is a
misnomer; its a distinction which makes no sense at all. As implemented in
PM5, UCR represents a particular choice made for a set of neutral axis
parameers: black start, max black and the extremely concave shape of the
K-curve connecting these points, ( and perhaps TAC as well).
> The black provides beef and depth to the images, and that mostly in the
shadow end.
Agreed. My point is that this desireable effect is probably better
achieved with other settings available to you in the PM5 UI. I suspect
that you can employ approximately the same black start and max K settings
afforded you by UCR, but choose a K-curve (GCR in PM UI) which is not as
concave in shape to connects these. points. This may avoid the CMY ink
reversals and result in more stable shadows and enhanced detail.
If you send me your PM5 UCR profile and the characterization data used to
construct it, I'll do an analysis for you and see if my UCR suppositions
apply to your inkjet printer. In the past, I've used this setting on other
characterization data sets in PM and have observed the negative
characteristics that I have described. I'd be happy to try and come up with
a mix of settings which may produce a profile which could, in principle,
out perform your UCR profile.
Harold
TOSHIBA AMERICA BUSINESS SOLUTIONS 2 Musick, Irvine, CA 92618
Harold Boll | 1.781.856.5174 | email@hidden
" The sunlights differ, but there is only one darkness. "
Ursula K. Leguin
_______________________________________________
Do not post admin requests to the list. They will be ignored.
Colorsync-users mailing list (email@hidden)
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden