QTR-Quadtone RIP

I am attempting to create a profile with my 1400 using UTFS inks.

Looking at a printed 21 step wedge, the lighter end seems to rise smoothly to about 60% where it flattens out, and from there on there is not much difference between the darker steps.

The prints look as you would expect, not too bad until the part-shadow and shadow.

I have an idea that the dark grey ink is being set as lighter than it really is.
When I print the first calibration print, the K ink limit comes up at 45%. Now, should I adjust the limit and then print another ink pattern page, without being in calibration mode, adjusting the setup of max ink limit until the printed ink limit is up around 100% instead of 45%.
If I don't, how then can the density of the dark gray ink be set accurately? eg. in the calibration scan, the 100% patch of the K row is the same density right down to 45%. That means that I am attempting to find the 100% LK point down below 45% of K. I think it is actually higher than that.

Any comments?

Greg <gregnixon@...> wrote:

> **
>
>
> I am attempting to create a profile with my 1400 using UTFS inks.
>
My memory of this setup is that the M and C are the same density, and the LC
and LM may be the same or very close.  Keep in mind that, in the QTR
partitioning, the inks in a progression must be different densities.  So,
you'll have to have the grays be, for example, Y, LC, C, K, and then set up
a toner channel with, for example, Y, LM, M, K (I'm forgetting if the K
needs to be there).

Is this matte paper with K = Eboni?

At any rate, first try Y, LC, C and K.  have the others not used.

If you are using PK and have it in the K position, recall that the C and M
might have a dmax that is greater than the PK, but they'll do that with huge
loads that will probably result in pizza wheel marks.  So part of what you
need to do is reduce the ink loads of the lighter inks below their dmax
points.

>
> Looking at a printed 21 step wedge, the lighter end seems to rise smoothly
> to about 60% where it flattens out, and from there on there is not much
> difference between the darker steps.
>

What profile where you using?  Was this the pre-linarized QTR output from an
initial profile?

As long as the differences in the 21-step print are enough that the results
will linearize with QTR, this might not matter.  You're often better off
with compression of the dark end than not because this gives the program
real data to use as opposed to interpolation.

>
> The prints look as you would expect, not too bad until the part-shadow and
> shadow.
>
> I have an idea that the dark grey ink is being set as lighter than it
> really is.
> When I print the first calibration print, the K ink limit comes up at 45%
>

If the K is fairly flat between 35 and 45, consider putting the ink limit at
35 and the boost to 45.

Consider using 35 as the ink limit for the M and C -- in fact for all inks.
 Use less light ink.  It's expensive water and the 1400's 1.5 pl drops don't
need it.  Particularly with glossy papers & inks, the higher the load, and
wetter the paper, the more problems with pizza wheel marks.

... in the calibration scan, the 100% patch of the K row is the same density
> right down to 45%.
>
That is rather normal.  You want to set the final K ink load at the left
edge of the plateau.  But, again, I prefer to use boost for the final 10%
because that area usually has a very low slope.

That means that I am attempting to find the 100% LK point down below 45% of
> K. I think it is actually higher than that.
>
There is no LK in that inkset.  But, again, see what the density of the
lighter inks are at 35% and consider that as the default ink limit.

Later when the toner channel is set up, split the ink limit between the gray
and toner channels.

I hope this helps.

Paul
www.PaulRoark.com




>
> Any comments?
>
>  
>


[Non-text portions of this message have been removed]

Thanks again Paul for your reply.

Yes M&C are the same density, LM and LC are very close to each other.
K is MK not PK. The paper is Epson Archival Matte.

"Keep in mind that, in the QTR partitioning, the inks in a progression must be different densities."
The progression is the grey step wedge? which would mean that I can't use both M & C in the printer, as they are the same density.

I used a profile .qidf from the 1290 set UT2 Enhanced matte Carbon (I think) parsed the text and then renamed, filled in the values etc.

I will try the setting that you have suggested and see what it looks like. 

Cheers
Greg


--- In QuadtoneRIP@yahoogroups.com, Paul Roark <roark.paul@...> wrote:

>
> Greg <gregnixon@...> wrote:
> 
> > **
> >
> >
> > I am attempting to create a profile with my 1400 using UTFS inks.
> >
> My memory of this setup is that the M and C are the same density, and the LC
> and LM may be the same or very close.  "Keep in mind that, in the QTR
> partitioning, the inks in a progression must be different densities."  So,
> you'll have to have the grays be, for example, Y, LC, C, K, and then set up
> a toner channel with, for example, Y, LM, M, K (I'm forgetting if the K
> needs to be there).
> 
> Is this matte paper with K = Eboni?
> 
> At any rate, first try Y, LC, C and K.  have the others not used.
> 
> If you are using PK and have it in the K position, recall that the C and M
> might have a dmax that is greater than the PK, but they'll do that with huge
> loads that will probably result in pizza wheel marks.  So part of what you
> need to do is reduce the ink loads of the lighter inks below their dmax
> points.
> 
> >
> > Looking at a printed 21 step wedge, the lighter end seems to rise smoothly
> > to about 60% where it flattens out, and from there on there is not much
> > difference between the darker steps.
> >
> 
> What profile where you using?  Was this the pre-linarized QTR output from an
> initial profile?
> 
> As long as the differences in the 21-step print are enough that the results
> will linearize with QTR, this might not matter.  You're often better off
> with compression of the dark end than not because this gives the program
> real data to use as opposed to interpolation.
> 
> >
> > The prints look as you would expect, not too bad until the part-shadow and
> > shadow.
> >
> > I have an idea that the dark grey ink is being set as lighter than it
> > really is.
> > When I print the first calibration print, the K ink limit comes up at 45%
> >
> 
> If the K is fairly flat between 35 and 45, consider putting the ink limit at
> 35 and the boost to 45.
> 
> Consider using 35 as the ink limit for the M and C -- in fact for all inks.
>  Use less light ink.  It's expensive water and the 1400's 1.5 pl drops don't
> need it.  Particularly with glossy papers & inks, the higher the load, and
> wetter the paper, the more problems with pizza wheel marks.
> 
> ... in the calibration scan, the 100% patch of the K row is the same density
> > right down to 45%.
> >
> That is rather normal.  You want to set the final K ink load at the left
> edge of the plateau.  But, again, I prefer to use boost for the final 10%
> because that area usually has a very low slope.
> 
> That means that I am attempting to find the 100% LK point down below 45% of
> > K. I think it is actually higher than that.
> >
> There is no LK in that inkset.  But, again, see what the density of the
> lighter inks are at 35% and consider that as the default ink limit.
> 
> Later when the toner channel is set up, split the ink limit between the gray
> and toner channels.
> 
> I hope this helps.
> 
> Paul
> www.PaulRoark.com
> 
> 
> 
> 
> >
> > Any comments?
> >
> >  
> >
> 
> 
> [Non-text portions of this message have been removed]
>

Greg <gregnixon@...> wrote:

> **
>
>
> ...
> K is MK not PK. The paper is Epson Archival Matte.
>
>
> "Keep in mind that, in the QTR partitioning, the inks in a progression must
> be different densities."
> The progression is the grey step wedge? which would mean that I can't use
> both M & C in the printer, as they are the same density.
>

If you are making a QTR profile, the partitioning phase will put the inks in
order of their density.  If 2 are the same, the system does not work.  You
can use the inkset, but you have to partition one "channel" at a time.  So,
for example, work up the Y, LC, C, and K channel as the "gray" inks.  Have
the M and LM turned off, that is "not used" in the Curve Creator.  Then add
the M and LM in a "toner" channel that uses the full Y, LM, M, and K, can
probably have the exact same curves (slightly different helps spread the
cross-overs), and splits the ink limit between the two channels -- manually
done.  (I'm sorry if this gets a bit complicated for someone just starting
out.  Note that the Epson driver and Create ICC can probably get you a good
print much faster.)

(Are you using Windows or Mac?  I use Windows and Curve Creator GUI.  How
 what I do is done in the Mac world will have to be handled by someone
else.)



> I used a profile .qidf from the 1290 set UT2 Enhanced matte Carbon (I
> think) parsed the text and then renamed, filled in the values etc.
>

Using a profile for a different inkset and different printer is chancy.
 Nonetheless, have you tried just linearizing this to see if it's within the
range of what the program can handle?  The fact that the pre-linearized (you
cleared the linearization tab, I assume) 21-step is dark might be totally
cured by the linearization step.  What it requires is that there is some
minimum separation between the steps.  It cannot handle posterization.
 However, smoothness of the curve is a better predictor of success with
linearization than the mere fact that the pre-linearized print is too dark.

Paul
www.PaulRoark.com


[Non-text portions of this message have been removed]