Digital BW, The Print

Hi Steve and Carl,

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, Steve Kale <stevekale@=
b...> 
wrote:
> 
> <Seems low for EEM with matte K, but not impossible.  Did you arrive at
> the 25% by  finding that the measured density of the 100% patch of the
> LK wedge was equal to the measured density of the 25% patch of the K
> wedge on the print of the ink separation file (printed in calibration
> mode with default ink limit set to 75 for all inks in the calibration
> profile)?>
> 
> LK¹s Ink Limit was set at 60.  I went through the process in the
> documentation for setting the default ink limit and decided that it was a=

> very arbitrary call up in the 70 to 90 range.  I then decided that I woul=
d
> maintain your existing ink limits for K and LK.  (Even though I note that=

> Roy references a higher range than a default set to 75 in his
> documentation.) The documentation implies (but is not explicit) that thes=
e
> limits should be determined prior to determining the partitioning.

The 2200 ink limits were done a little more adhoc rather than strictly by
my orginal documents.  The idea for cool 2200 prints is to have a
combination of LK,LC,LM play the role of one light black ink. (Basically,
instead of mixing the inks manually to get a neutral gray, the idea is
to mix them using software curves).  However the difference is that you
now have 3 inkjets delivering ink so it made sense to lower the individual
limits to make sure we don't flood the paper.

> 
> 
> <No it is not an abrupt switch, the two inks are blended and you can see
> the transition by running the graph_curve script with your .quad warm
> curve file (from the curvedropbox).>
> 
> Got it.  What is the x axis on this chart?  Ink percentage? I am surprise=
d
> to see (in the graph of the unlinearized curve file) LK peaking way up in=

> the 63% range and K not beginning until the high 40s, despite partitionin=
g
> at 25%.  I would love an explanation of this in layman terms.

I think you have the right idea about the chart -- its rotated 90deg CW
from the more natural view.  Down the page is gray values from 0K to 100K.
Across is ink percentage -- i.e. proportional to how many dots of each
ink are layed down.

Partitioning is a done by "weighting" the inks and limiting the total amoun=
t
of ink rather than by crossover points.  Secondly, because of dotgain
and the logarithmic nature of density, all the curves are pushed over
quite a bit.  I.e. where you'd think K might come in at 25% it comes in
much later like 40%.

> 
> 
> <Mine usually are about the same dmax or higher after linearization.>
> 
> Hmmmm...I would expect the linearization curve to use a fixed dMax readin=
g
> and that the movement we see is down to reading variance/drying effects b=
ut
> I guess that depends on the math.

They ought to match.  Drying/aging the inks is significant.

> 
> 
> The ink limit variables and boost variable still baffle me.  Presumably,
> right up to the boost level the maximum ink deposit is 75% K coverage and=

> maybe the very last remnants of LK. At the boost level we kick in more in=
k.
> Is the boost variable the level it kicks in or the ink coverage to which =
it
> climbs?  If the latter, then why would boost not be equal to 100?
> Presumably the argument is that we get good coverage at 75 (hence the
> default ink density) and a push to 90/95 is plenty at the very top end.  =
If
> this is to combat dot gain I would have thought that this would already b=
e
> factored into the underlying driver such that 100% gave maximum and optim=
al
> coverage.  Perhaps Roy can comment on these points.

The Boost number is another ink limit percentage.  E.G. if LIMIT_K=75 and
BOOST_K=90,  the K ink will start at a slope based on the 75 value, however=

towards the top it will "smoothly" transition into an ink limit of 90.  Thi=
s
gives an extra umph of black ink right at the end for a greater dMax.
Putting it too high though would make you reach dMax before the darkest
gray -- no separation from there on.  The underlying driver has no
builtin dotgain or optimal coverage factors -- it's totally "raw" dots.

Roy


> 
> Steve
> 
>