Digital BW, The Print

Hi,
   I have just set up an Epson 1400 with the MIS UT-14 inks, and
things seem to be working pretty well printing from Photoshop with
Paul Roark's profiles.  But, I would like to try using QTR to get
better control over the toning.  For instance, I find Paul's "warm"
too warm for my taste, so I would like to be able to mix a warm and
cool curve in QTR.

I think that I have managed to figure out how to make QTR curves, but
it isn't very clear to me how to mix the UT-14 inks for different
effects.  From Paul's description, it initially sounded to me as
though his profiles for glossy papers used only the non-black
channels.  I have tried making a QTR curve that just uses the C and LC
inks, for a cool tone.  It does work, but the prints clearly lack the
depth of those generated by Paul's ICC profiles, leading me to think
that there must actually be some Ebony black used.

If Paul or anyone else, could give me some guidance on how to make QTR
curves for this particular inkset, I would be very grateful.

Thanks,
David

David,


>    I have just set up an Epson 1400 with the MIS UT-14 inks,...
> I would like to try using QTR to get better control over 
> the toning.  For instance, I find Paul's "warm"
> too warm for my taste, so I would like to be able to mix a
> warm and cool curve in QTR.

That would be a useful.  QTR does give the most control.

Just for fun, you might see what happens if you take, for example, 
the warm and neutral PS curves that are included in the 1400-
Profiles.zip and put them on layers in PS and try blending them at, 
say 50% each.  See if they are close enough to re-linearize in an 
ICC.  Sometimes this combination of PS curves works, sometimes not.


> I think that I have managed to figure out how to make QTR
>  curves, but it isn't very clear to me how to mix the UT-14 
> inks for different effects.  

As noted on page 1 of http://www.paulroark.com/BW-Info/UT14.pdf, the 
C channel is cool, M is warm carbon.  C is, in effect, a PKN.  M is a 
standard PK.

> From Paul's description, it initially sounded to me as
> though his profiles for glossy papers used only the non-black
> channels.  I have tried making a QTR curve that just uses the 
> C and LC inks, for a cool tone.  It does work, but the prints 
> clearly lack the depth of those generated by Paul's ICC 
> profiles, leading me to think
> that there must actually be some Ebony black used.

No, look at the glossy curves in the Zip file.  The C and M curves 
(Red and Green) are close to -- but not at -- 100% at the black 
point.  The yellow ink (Blue curve) (Glop) is off.  To turn on the 
Eboni that Blue curve needs to be firing at least a little (I think).

I also do use the glop to suppress the reflective artifacts.

You might want to make a curve for QTR that has the C and M channels 
going from 0 to 100 and print a 21-step test strip to find the dmax.  
See if it is higher with both channels firing.

Hope this helps.

Paul
www.PaulRoark.com

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "pr_roark"
<pr_roark@...> wrote:
> 
> No, look at the glossy curves in the Zip file.  The C and M curves 
> (Red and Green) are close to -- but not at -- 100% at the black 
> point.  The yellow ink (Blue curve) (Glop) is off.  To turn on the 
> Eboni that Blue curve needs to be firing at least a little (I think).
> 
> I also do use the glop to suppress the reflective artifacts.
> 
> You might want to make a curve for QTR that has the C and M channels 
> going from 0 to 100 and print a 21-step test strip to find the dmax.  
> See if it is higher with both channels firing.
> 
> Hope this helps.
> 
> Paul
>
 
Paul,
   Thanks very much.  That does help clarify things.  I think that my
major problem earlier was much more basic: I was not using the maximum
printer resolution and the "uni-directional" option when printing. 
With these set, things look much better, and I am trying again with
the QTR curves.

David

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "pr_roark"
<pr_roark@...> wrote:
>
> 
> I also do use the glop to suppress the reflective artifacts.
> 
> You might want to make a curve for QTR that has the C and M channels 
> going from 0 to 100 and print a 21-step test strip to find the dmax.  
> See if it is higher with both channels firing.
> 
> Hope this helps.
> 
> Paul


Paul,
    A couple more questions:

1. At what levels is the Glop used? Is it constant with black density,
or should it decrease as more ink is applied?

2. Is there a way to see what the actual ink levels are from your ICC
profiles?  If I apply the curve to an image in Photoshop, I can see
what they are, but they are obviously expressed in terms of the
current color model (Grayscale, RGB, CMYK).  If I look at the CMYK
profile for the Kirkland Glossy (neutral), the yellow fixed at zero is
at zero, and the black is linear from 0 to 100%.  But image has to go
through the Epson driver as well, so I assume that the final ink
values are different.  

Thanks,
David

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "dpgoldenberg33" 
<dpgoldenberg33@...> wrote:
>
...
> 
> 1. At what levels is the Glop used?

You can see in those 1400 glossy curves how high the Blue curve goes
(yellow ink position = glop).  I think one the Iflord curve I opened 
it hits about a max of 40% in the mid-highlights.

> Is it constant with black density,
> or should it decrease as more ink is applied?

With just black firing the distribution might be different.  I just 
worked visually from 21-step test strips.  Where there was bad 
bronzing I added a little more glop -- very much a trial and error 
type of approach.

> 2. Is there a way to see what the actual ink levels are from 
> your ICC profiles?

The curves that are in the Zip file are the ones I used in the ICCs.  
I'm sure Roy could tell you what program is needed to open the ICCs 
and extract the curve information, but when they are opened with 
Notepad it looks like Romulan to me.

> If I apply the curve to an image in Photoshop, I can see
> what they are, but they are obviously expressed in terms of the
> current color model (Grayscale, RGB, CMYK).  If I look at the CMYK
> profile for the Kirkland Glossy (neutral), the yellow fixed at
> zero is at zero, and the black is linear from 0 to 100%.  
> But image has to go
> through the Epson driver as well, so I assume that the final ink
> values are different.  

Yes, and I suspect that in addition to the ink limit that is imposed 
by the driver, usually set by the paper type, the relationships we 
see in the curves is not applied to the inks in a linear fashion.  My 
general experience with writing these PS curves makes me think that 
the one unit moves near the top of the curve (highest ink output) 
adds much more ink per unit that near the bottom of the curve.  I 
suspect this is to offset the generally decreasing slope of the ink 
to density relationship as more ink is added.

Paul
www.PaulRoark.com

> ...
> > 
> > 1. At what levels is the Glop used?
> 
> You can see in those 1400 glossy curves how high the 
> Blue curve goes (yellow ink position = glop).  
> I think one the Iflord curve I opened 
> it hits about a max of 40% in the mid-highlights.
> 

I see that on some of the glossy curves I don't have any glop in the 
mix.  I, frankly, forget why.  I suspect I switched the printer to 
Eboni-6 before I'd finished the curves.  It'll simply take soem 
experimentation to see how much glop is best for different papers.  
Just print 21-step test strips and see where the bronzing or gloss 
differential are and how much different amounts of glop help or don't 
help at that point.

Paul
www.PaulRoark.com