Digital BW, The Print

Carl/Roy

Some observations after another evening playing around with this and hopefully a lot 
better understanding.

Using EEM and just focusing on the warm curve for now, Default Ink Limit set at 75% 
(tough on the eyes to place this one) and capping the LK at 60%, I get a much lower 
cross over point into the use of the LK ie 25 rather than 35.  (Capping LK at the 
default 75% makes little difference to this number, ie 26%.)  I also set boost at 95 
rather than 90.

Am I right in seeing the selection of this 25% point as being the point at which QTR 
switches from using LK to K to produce shades of gray, ie only LK is used from 0 to 
25 and only K is used from 26 to 100, or is it not an abrupt switchover? 

The unlinearised density values were:
   
0.039 0.087 0.153 0.224 0.305 0.375 0.464 0.548 0.632 0.705 0.783 0.862 0.949 1.
041 1.124 1.210 1.295 1.374 1.456 1.548 1.666

And the linearized density values were:

0.040 0.089 0.138 0.189 0.243 0.293 0.361 0.420 0.479 0.541 0.610 0.689 0.770 0.
855 0.938 1.031 1.148 1.242 1.361 1.496 1.623

I was surprised to see the dMax difference but both of these seem to be jumping 
around with different readings and as the ink settles.  I would expect the two 100% 
figures to be the same.

Wouldn't the lower use of LK make it easier to cool this curve rather than one with 
Gray_Ink_Val LK set to 35?

Cheers

Steve



--- In DigitalBlackandWhiteThePrint@yahoogroups.com, Carl Schofield <scho@m...> 
wrote:
> Steve,
> 
> On Monday, December 22, 2003, at 01:00  PM, Steve Kale wrote:
> > There could be printer differences, Eye-One (or other instrument)
> > differences, or differences in drying time before measurement.  This
> > why it is a good idea to do your own linearization.>
> >
> > Ok but you would agree that I am at least following the right procedure
> > here?
> 
> Yes.
> 
> >> 3) Why is LK limited to 56 (cool)/ 60 (warm)?
> > You answered this question in #4 below.
> >
> > Sorry my point was why around this level in the first place (in 4 I was
> > simply referring to the cool curve having a slightly lower figure than 
> > the
> > warm).
> 
> As I recall the specific values were determined some time ago using an 
> earlier beta version, prior to the implementation of the linearization 
> feature.  The numbers were arrived at by trial and error, with some 
> guidance from Roy.  The iterative process involved setting limit values 
> for LK, LC, and LM, printing and measuring step wedges, plotting the 
> density curves, and finally evaluating the curves for linearity.  We 
> were also trying to equalize the densities of the cool and warm curves 
> and adjust them, using the ink limit values.  A mid-tone (50%) density 
> of about 0.6 was also a target value for adjustment.  Roy may wish to 
> clarify, but I think that the linearization makes the absolute limit 
> values used less critical in this regard.
> 
> >
> >> 4) I assume the limits for LC and LM need to be the same to balance
> >> against
> >> metamerism and that LC and LM are used to cool the Epson K and LK (and
> >> I guess it
> >> follows that one might lower the LK limit a little in the cool file).
> > The idea was to adjust LC and LM, relative to LK,  to produce a very
> > cool curve that could be used for blending with the warm curve to
> > produce the desired tone.
> >>     Is a cool curve is
> >> one that takes the warm and at 50:50 is neutral?
> > It could be adjusted to do so, but perfect neutrality at 50% would be
> > fortuitous.
> >>
> >> 5) If so, how does one measure neutral?
> > You could use the LAB or XYZ values from the Eye-One to evaluate
> > neutrality.
> >>
> >> 6) Was the cool file created by steadily increasing the limit on LC
> >> and LM until this was
> >> achieved?
> > I wasn't shooting for neutral with the cool curve - just a very cool
> > curve that could be subsequently blended with warm to get the desired
> > tone.
> >
> >
> > I guess I need a little more guidance on how you determine you have 
> > obtained
> > a cool curve (or neutral combination).  That is, when you are working 
> > on
> > _adjust[ing] LC and LM, relative to LK_ how do you know you have 
> > reached the
> > goal of a very cool curve (and so there is no need to go further)?
> 
> I just did it by eye to get a curve that printed the step wedge a cool, 
> blue color.  It is then much easier to use the blender in the print 
> driver to get close to "neutral" or whatever other hue you prefer for 
> your prints by blending the cool and warm curves.
> 
> >
> > <snip...
> 
> Carl