Digital BW, The Print

"Olivier" wrote:

>  Now one thing I don't fully understand is the multi-pass 
> to lay down GO. At the Dmax point I understand but in 
> other parts of the greyscale I feel you can process 
> GO as an additional partition.

We're using Glop (the MIS version seems to do a much better job than 
Epson's according to Carl's experiments) both in the first pass and 
as a second-pass overcoat.

We tried various approaches -- no Glop at first, a flat amount, and 
various curves for the Glop during the first pass.  The reasons for 
the Glop during the first pass are to help "nail" down the matte ink 
at least somewhat to avoid abraision marks and to result in an even 
final gloss.  Particularly before we were using PK, the end point 
required a very large load of Glop to get an even gloss.  With PK, 
the matte-compatible (not glossy) inks still need a relatively heavy 
load of Glop to avoid a gloss differential when the very glossy PK 
takes over.  It'll take a few iterations at the cross-over point to 
get that transition looking good.

On the other hand, I think it's possible, if PK is used, that one 
could use no glop at first and just pass the print through a second 
pass -- maybe using a different printer --  for the glop overcoat.  
But the risks of marks and gloss differential would be increased.  
I'm just not sure if one could feather in the PK and overcoat enough 
to get a good transition from the matte to gloss ink.

 
> Say in a simplified example you have at a given output 25% dark ink 
> and 75% light ink for a 100% coverage. The light ink is 50% density 
> of the dark one. At this specified level you need to apply 20% GO 
> (empirically determined and that's probably a bigger issue). As you 
> know there is not a single mix to deliver this output density : 25% 
> dark and 75% light with light density at 50% dark density is 
> equivalent to 62.5% dark alone. If you have to limit total 
> light and dark coverage to 80% to allow 20% GO and full 
> 100% coverage then you 
> can output 35% light and 45% dark (same 62.5% dark equivalent) 
> together with the 20% GO. That trades GO against granularity
> but it's  a single pass.

In the 3MK setup, we never run into paper load issues until the dmax 
point.  The 3 MK channels are so "dry" that there's lots of room for 
Glop.

With the Eboni-6 setup I experimented with briefly -- which is 
essentially a K6 using only Eboni and a matte dilution base -- I'm 
not sure I ran into any ink limit issues either, even though, 
presumeably like the typical K6 and K7 profiles, I use a lot of light 
inks.  I think the addition of glop does not just add to the total 
load and result in the same resulting visual problems that one would 
see with an overloaded paper.  But, I understand the risk of 
overloading the paper.  

With a K6 setup, I think the main thing you'd have to do is 
substitute a PK for the combination of inks Jon uses at the dark 
end.  (Note in the k6-7 test strips I've seen, the 100% is already 
bleeding badly from overloading.)

In my latest QTR profile, the Glop curve is as follows: 0;0	25;5
	50;10	75;24	80;27	85;30	90;30	92.5;27	95;22	100;0
	
I split this between 2 channels limited to 50% each.

My PK curve is as follows: 0;0	10;1	20;1.5	30;1.5	40;2	50;1.5
	60;1.5	65;2	70;1.5	75;1	80;4.25	85;8	90;15
	92.5;22	95;35	98;60	100;100	  (Ink limit 100)

Note that with the overcoat, the PK does not reach it's dmax until 
beyond 100%.  But the marginal increase may not be worth the effort 
it takes to hide the gloss differential.

Note also that I use the PK not only at the end but also in the light 
midtones to keep the print from being too  cool. (It's odd to think 
of a carbon print going cold, but it does.)

With the 3MK setup, the underlying Eboni curves are just about where 
they were with a matte print except for the end point.  I roll all of 
them off to (100,0).  In general, I think it's a mistake to have any 
other inks in the 100% spot aside from the MK or PK.

The second pass is done with a 100% Glop overcoat, split between 2 
channels.  With the 1800 I was able to see some microbanding in the 
glop if only one channel was used.  That would probably not happen in 
a 3.5 pl printer.

So, bottom line, with Piezo (I assume that is what you're using), I 
think you could use the glop curve above while printing if a PK was 
substituted in, and then do a 100% overcoat, and probably have a very 
interesting glossy print -- generally better than what glossy inks 
produce.  Just keep an eye out for possible incompatibilities between 
the Piezo matte base and Glop.  I may have run into such with the 
Eboni-6 and its matte base (which I'm guessing is close to what Jon 
uses).

If you see evidence of overloading the midtones, you might be able to 
back the Glop off and still get a good final gloss.  Particularly in 
the light midtones, not much is needed.

I hope this is somewhat responsive to the question.

Paul     
www.PaulRoark.com