QTR-Quadtone RIP

--- In QuadtoneRIP@yahoogroups.com, "Tom Moore" <r.t.moore@...> 
wrote:
> It sounds to me like you have a pretty good grasp of the toner
> concept. 

I'm SLOWLY getting there ;-)

> If I may use a very simplistic analogy, QTR is like a toolbox
> with a number of tools. The grey, toner and toner 2 curves are 
like 3
> wrenches. They are normally used for adjusting bolts. Sometimes you
> need to pound a nail, and they will do that too. Maybe not as 
easily
> as using a hammer, but you can make it work.
> 
> The three curves in QTR (grey, toner and toner 2) provide 3 
parallel
> means of partitioning the inks over the greyscale range. 

This is what I understand, I hope to get some more insight when to 
the wrench and when to use the hammer...
 
> For example, the density information for ink
> influences where in the tonal range the ink is used. The limit
> controls how much is used. The relative densities of each ink 
controls
> the order of use of the inks assigned to that curve over the tonal 
range.
> 
> One interesting feature of QTR is that the partitioning is applied
> even if an ink is limited to 0. So, for example, that means that 
you
> can apply an ink A to the deep shadows by assigning  another
> (otherwise unused) ink B position with a 0 limit and lower density
> than the ink you want to use. The density value of this lower ink B
> can be arbitrarily set to control where in the tonal range, Ink A
> starts to come in. If you look at the curve for Ilford Smooth Pearl
> warm or cool you'll see an example of this.

I partly understood this, but you definitely give some new insights 
here.
 
> I haven't used the UT-3D inks or seen any curves, so I can't 
suggest
> the "best" way to use QTR for this inkset. What ever way you 
decide to
> proceed, sharing the results with with us all will help other get 
over
> the hump you are climbing.

I have some first results now. I made (first versions of) the 
following basic curves (all for the 2100 on EEM):
- warm (relinearization of the UT7 curve)
- cool relinearization of the UT7 curve)
- black only
- neutral (an approx. 65/35% mix of warm and cool, making the 
midtones as neutral as possible)
Actually, these curves were pretty straightforward. 

More complicated are the curves involing the lab a/selenium toner. I 
tried the following:

- selenium 1: a K3 approach, Eboni for K, the neutral 65/35 
warm/cool mix as mentioned above for LK and the selenium/Lab a toner 
for LLK. So, the lab a toner is part of the gray curve. This is 
actually quite simple to make since the lab a toner has the same 
density as the warm & cool LLK toners. To me the disadvantage is 
that the shadows remain neutral and that only the higlights are 
toned.

- selenium 2: a "K3+toner" approach, with eboni, mix of warm & cool 
toners as K, LK and LLK + applying a QTR toner curve to the lab a 
toner. It is a bit more tricky to get this one right. I have 
something working now, with more tone in the shadows. Probably it 
can be better. But frankly, I don't know how to define "better" 
here. I can plot the a and b values over the range, but I'm not sure 
what an optimal plot should like like.

In addition, I made some split tone curves, mostly following the 
straightforward K3 approach as above, having either a warm or cool 
LK for the shadows and a warm, cool or selenium LLK toner for the 
highlights. I'm not too satisfied with these. The split tone effect 
is not very strong, since the tone of the shadows is much less clear 
than that of the higlight. To overcome this I am experimenting with 
two "cool2selenium" approaches:

- a K3+toner approach as above, but with pushing the selenium toner 
far to the highlights to leave some room for the cool LLK in the 
upper mid and lower higlights. 
- a "quasi K4" approach with for all toners as "grays": eboni for K, 
dark cool for LK, light cool for LLK and and selenium for LLLK 
(actually designed as LLK toner)

The first one works decently. I do not have a decent result with the 
second one. Since the densities of the LLK and LLLK toners are 
actually almost the same, I get a very sharp transition between cool 
and selenium. Based on Tom's suggestion I should perhaps 
artificially change the densities in the curve definition.

That's it till so far. Hope it is of any use to someone. Perhaps I 
am stating the obvious. Any suggestions are welcome!

Joost