QTR-Quadtone RIP

Good Evening-

I'll preface this with a statement I've always believed: Color
Management is (fairly) easy.  Today I'm learning it's not.  I'm a former
Kodak Professional employee who lived and breathed it- and now I'm
choking on it :)

That said:
My worksheet is at http://www.gotSheep.com/ICC/21target_QuadRip_BW.xls
<http://www.gotSheep.com/ICC/21target_QuadRip_BW.xls>   .  For those
concerned about XLS files, the graph I created is here:
http://www.gotSheep.com/ICC/DensityChart.pdf
<http://www.gotSheep.com/ICC/DensityChart.pdf>

I purchased a set of UT2 inks.  After finally getting them loaded into
the R280 printer (auto-reset chips aren't working right) I printed off
the calibration chart in QTR.  Eyeballing the chart I modified the ink
slider to 40%, then printed the chart again.  With that new chart in
hand I created a spectrolino calibration target and read the values in. 
Said data is present on sheet one in the XLS file- all spectral data,
XYZ, LAB, and Density is present.  Paper white is 0 (1A).

You'll note that the data rounds off on the black at 1.71, and continues
to build on the greys up to the max the printed put down at 40%
(1.34/1.07).

That said:  I should have all the data I need to create the printing
curve in QTR, but when I do so it's ... well, black cat, coal mine,
midnight, 25 speed film.  The 21 step wedge prints 1 step- 0- as white,
and then becomes quite dark (eyeball: 1,0 density) at the 2nd wedge.

My curve I created is here at:
http://www.gotSheep.com/ICC/R280_RedRiver_PolarMatte_UT2_BW_mod2.qidf
<http://www.gotSheep.com/ICC/R280_RedRiver_PolarMatte_UT2_BW_mod2.qidf>

In it I go through the calculations to map LK to B for Cyan/Magenta
(copy the curve from) (seen here:
http://www.gotSheep.com/ICC/CurveCreator.png 
<http://www.gotSheep.com/ICC/CurveCreator.png> ), whereas the resultant
curve looks like this:  <http://www.gotSheep.com/ICC/RRcurves.png>
http://www.gotSheep.com/ICC/RRcuves.png
<http://www.gotSheep.com/ICC/RRcuves.png>

Now, believing myself to be a fairly intelligent individual that used to
*design* color management systems for sites... having read all the
available tutorials on the subject... and lastly really wanting to see a
single print look decent, could I trouble one member to point me in a
direction that will stop me from pounding my head into the fresh drywall
in the nursery?

Thanks in advance-

Jason

(The image I'm dying to print is here-
http://purduephotog.deviantart.com/art/IR-Tahoe-Meadow-92211698
<http://purduephotog.deviantart.com/art/IR-Tahoe-Meadow-92211698>   - a
35 frame IR panorama taken in Reno with a home-made modified 300D Rebel)



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

Hi Jason. Given the info you presented, it looks to me as if you may
have erred on the 3 lighter inks. Their limits ought to be about 13.3
on each and they are about 7.5% of the K density where you have only
3.xxx for them. I believe the curve creator thinks it is blending much
lighter inks than they actually are as w3ell as putting down more of
them(eg:45% limit total).
Remember that all ink densities are listed in the boxes in relation 
to the 100% or K ink. You appear to have done that with the C & M but
not with the Y,LC,LM.

Hope this is helpful.

Regards,
Duane




--- In QuadtoneRIP@yahoogroups.com, "Jason" <hirsch@...> wrote:

>
> Good Evening-
> 
> I'll preface this with a statement I've always believed: Color
> Management is (fairly) easy.  Today I'm learning it's not.  I'm a former
> Kodak Professional employee who lived and breathed it- and now I'm
> choking on it :)
> 
> That said:
> My worksheet is at http://www.gotSheep.com/ICC/21target_QuadRip_BW.xls
> <http://www.gotSheep.com/ICC/21target_QuadRip_BW.xls>   .  For those
> concerned about XLS files, the graph I created is here:
> http://www.gotSheep.com/ICC/DensityChart.pdf
> <http://www.gotSheep.com/ICC/DensityChart.pdf>
> 
> I purchased a set of UT2 inks.  After finally getting them loaded into
> the R280 printer (auto-reset chips aren't working right) I printed off
> the calibration chart in QTR.  Eyeballing the chart I modified the ink
> slider to 40%, then printed the chart again.  With that new chart in
> hand I created a spectrolino calibration target and read the values in. 
> Said data is present on sheet one in the XLS file- all spectral data,
> XYZ, LAB, and Density is present.  Paper white is 0 (1A).
> 
> You'll note that the data rounds off on the black at 1.71, and continues
> to build on the greys up to the max the printed put down at 40%
> (1.34/1.07).
> 
> That said:  I should have all the data I need to create the printing
> curve in QTR, but when I do so it's ... well, black cat, coal mine,
> midnight, 25 speed film.  The 21 step wedge prints 1 step- 0- as white,
> and then becomes quite dark (eyeball: 1,0 density) at the 2nd wedge.
> 
> My curve I created is here at:
> http://www.gotSheep.com/ICC/R280_RedRiver_PolarMatte_UT2_BW_mod2.qidf
> <http://www.gotSheep.com/ICC/R280_RedRiver_PolarMatte_UT2_BW_mod2.qidf>
> 
> In it I go through the calculations to map LK to B for Cyan/Magenta
> (copy the curve from) (seen here:
> http://www.gotSheep.com/ICC/CurveCreator.png 
> <http://www.gotSheep.com/ICC/CurveCreator.png> ), whereas the resultant
> curve looks like this:  <http://www.gotSheep.com/ICC/RRcurves.png>
> http://www.gotSheep.com/ICC/RRcuves.png
> <http://www.gotSheep.com/ICC/RRcuves.png>
> 
> Now, believing myself to be a fairly intelligent individual that used to
> *design* color management systems for sites... having read all the
> available tutorials on the subject... and lastly really wanting to see a
> single print look decent, could I trouble one member to point me in a
> direction that will stop me from pounding my head into the fresh drywall
> in the nursery?
> 
> Thanks in advance-
> 
> Jason
> 
> (The image I'm dying to print is here-
> http://purduephotog.deviantart.com/art/IR-Tahoe-Meadow-92211698
> <http://purduephotog.deviantart.com/art/IR-Tahoe-Meadow-92211698>   - a
> 35 frame IR panorama taken in Reno with a home-made modified 300D Rebel)
> 
> 
> 
> [Non-text portions of this message have been removed]
>

Thanks very much Duane-

I took your advice and remapped the curves.  For the first time I
actually have a 'grey' value, but unfortunately immediately after the
10 patch we're back into the blacks.

I'll keep digging- I had hoped to use some of the BO only curves as
supplied with QTR but none of them help me with the more refined
portions of the greys.  Right now I have white, dark medium grey, and
black within 20% on the 21 step wedge.

Jason

--- In QuadtoneRIP@yahoogroups.com, "dlruckus" <dlruckus@...> wrote:
>
> Hi Jason. Given the info you presented, it looks to me as if you may
> have erred on the 3 lighter inks. Their limits ought to be about 13.3
> on each and they are about 7.5% of the K density where you have only
> 3.xxx for them. I believe the curve creator thinks it is blending much
> lighter inks than they actually are as w3ell as putting down more of
> them(eg:45% limit total).
> Remember that all ink densities are listed in the boxes in relation 
> to the 100% or K ink. You appear to have done that with the C & M but
> not with the Y,LC,LM.
> 
> Hope this is helpful.
> 
> Regards,
> Duane
> 
> 
> 
> 
> --- In QuadtoneRIP@yahoogroups.com, "Jason" <hirsch@> wrote:
> >
> > Good Evening-
> > 
> > I'll preface this with a statement I've always believed: Color
> > Management is (fairly) easy.  Today I'm learning it's not.  I'm a
former
> > Kodak Professional employee who lived and breathed it- and now I'm
> > choking on it :)
> > 
> > That said:
> > My worksheet is at http://www.gotSheep.com/ICC/21target_QuadRip_BW.xls
> > <http://www.gotSheep.com/ICC/21target_QuadRip_BW.xls>   .  For those
> > concerned about XLS files, the graph I created is here:
> > http://www.gotSheep.com/ICC/DensityChart.pdf
> > <http://www.gotSheep.com/ICC/DensityChart.pdf>
> > 
> > I purchased a set of UT2 inks.  After finally getting them loaded into
> > the R280 printer (auto-reset chips aren't working right) I printed off
> > the calibration chart in QTR.  Eyeballing the chart I modified the ink
> > slider to 40%, then printed the chart again.  With that new chart in
> > hand I created a spectrolino calibration target and read the
values in. 
> > Said data is present on sheet one in the XLS file- all spectral data,
> > XYZ, LAB, and Density is present.  Paper white is 0 (1A).
> > 
> > You'll note that the data rounds off on the black at 1.71, and
continues
> > to build on the greys up to the max the printed put down at 40%
> > (1.34/1.07).
> > 
> > That said:  I should have all the data I need to create the printing
> > curve in QTR, but when I do so it's ... well, black cat, coal mine,
> > midnight, 25 speed film.  The 21 step wedge prints 1 step- 0- as
white,
> > and then becomes quite dark (eyeball: 1,0 density) at the 2nd wedge.
> > 
> > My curve I created is here at:
> > http://www.gotSheep.com/ICC/R280_RedRiver_PolarMatte_UT2_BW_mod2.qidf
> >
<http://www.gotSheep.com/ICC/R280_RedRiver_PolarMatte_UT2_BW_mod2.qidf>
> > 
> > In it I go through the calculations to map LK to B for Cyan/Magenta
> > (copy the curve from) (seen here:
> > http://www.gotSheep.com/ICC/CurveCreator.png 
> > <http://www.gotSheep.com/ICC/CurveCreator.png> ), whereas the
resultant
> > curve looks like this:  <http://www.gotSheep.com/ICC/RRcurves.png>
> > http://www.gotSheep.com/ICC/RRcuves.png
> > <http://www.gotSheep.com/ICC/RRcuves.png>
> > 
> > Now, believing myself to be a fairly intelligent individual that
used to
> > *design* color management systems for sites... having read all the
> > available tutorials on the subject... and lastly really wanting to
see a
> > single print look decent, could I trouble one member to point me in a
> > direction that will stop me from pounding my head into the fresh
drywall
> > in the nursery?
> > 
> > Thanks in advance-
> > 
> > Jason
> > 
> > (The image I'm dying to print is here-
> > http://purduephotog.deviantart.com/art/IR-Tahoe-Meadow-92211698
> > <http://purduephotog.deviantart.com/art/IR-Tahoe-Meadow-92211698>
  - a
> > 35 frame IR panorama taken in Reno with a home-made modified 300D
Rebel)

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

Jason, You might try lowering or 0'ing the boost value as well. If you
are already using a K limit at the maximum the paper can take without
reducing density( the turnover you mentioned), putting the boost at a
higher level might put more ink down than you want in the lower
values. Ostensibly, it isn't supposed to affect anything other than
the 100% patch but my experiences have given some question to that.

Another thing you might look at is to try the overlap value at O as
well. Since you are duplicating the C&M curves and both are then
overlapped with K, it's putting a lot of dark ink down. Likewise, you
may have your limits for the medium and light inks at too high a
level. Try moving down on your curves to a point just before they
start to nearly level out and set your levels and densities from those
points. There is no reason why you shouldn't be successful with your
inkset.

As a last resort, you might start with using just 3 of the inks(K,C,Y)
to make your curves and when that is OK add the others in one at a
time while modifying the limits etc of the equal density ones until
the curve again is balanced. It might also be that you are seeing
visual effects caused by dot size variation algorithms  programmed
into QTR. Your inks are all fairly high densities so perhaps are
impacting the change points for dot size in the curve creator in some way.

Unless I'm mistaken, the use of multiple inks of the same density is
relatively new to QTR's use, so making the automatic curve creator
work well might require a bit more than the general tutorials
recommendations provide. P. Roark does this in a different way by
using Create Icc  and embedding curves. You might check out his site
for tutorials on his methods.

Regards,
Duane
 


--- In QuadtoneRIP@yahoogroups.com, "Jason" <hirsch@...> wrote:

>
> Thanks very much Duane-
> 
> I took your advice and remapped the curves.  For the first time I
> actually have a 'grey' value, but unfortunately immediately after the
> 10 patch we're back into the blacks.
> 
> I'll keep digging- I had hoped to use some of the BO only curves as
> supplied with QTR but none of them help me with the more refined
> portions of the greys.  Right now I have white, dark medium grey, and
> black within 20% on the 21 step wedge.
> 
> Jason
>

Duane-

That did it.  Well, at least, I now have discernible density up to 50%
(not a correct tone scale, but I can see into the blacks up to that
level).

It looks like I need approximately another 75% reduction in levels to
create a completely graduated tone scale.

So it would appear if I'm going to want to mix my inks to keep
consistent usage I'm going to have a heck of a time using such small
numbers (division by 2x and 3x for each ink).

The next step will be to further refine how much ink is being put
down.  *sigh* This was never so difficult at work.  Print the patch,
read it in, spit out the profile.  Repeat.  I must have been spoiled ;)

Jason

--- In QuadtoneRIP@yahoogroups.com, "dlruckus" <dlruckus@...> wrote:
>
> Jason, You might try lowering or 0'ing the boost value as well. If you
> are already using a K limit at the maximum the paper can take without
> reducing density( the turnover you mentioned), putting the boost at a
> higher level might put more ink down than you want in the lower
> values. Ostensibly, it isn't supposed to affect anything other than
> the 100% patch but my experiences have given some question to that.
> 
> Another thing you might look at is to try the overlap value at O as
> well. Since you are duplicating the C&M curves and both are then
> overlapped with K, it's putting a lot of dark ink down. Likewise, you
> may have your limits for the medium and light inks at too high a
> level. Try moving down on your curves to a point just before they
> start to nearly level out and set your levels and densities from those
> points. There is no reason why you shouldn't be successful with your
> inkset.
> 
> As a last resort, you might start with using just 3 of the inks(K,C,Y)
> to make your curves and when that is OK add the others in one at a
> time while modifying the limits etc of the equal density ones until
> the curve again is balanced. It might also be that you are seeing
> visual effects caused by dot size variation algorithms  programmed
> into QTR. Your inks are all fairly high densities so perhaps are
> impacting the change points for dot size in the curve creator in
some way.

> 
> Unless I'm mistaken, the use of multiple inks of the same density is
> relatively new to QTR's use, so making the automatic curve creator
> work well might require a bit more than the general tutorials
> recommendations provide. P. Roark does this in a different way by
> using Create Icc  and embedding curves. You might check out his site
> for tutorials on his methods.
> 
> Regards,
> Duane
>  
> 
> 
> --- In QuadtoneRIP@yahoogroups.com, "Jason" <hirsch@> wrote:
> >
> > Thanks very much Duane-
> > 
> > I took your advice and remapped the curves.  For the first time I
> > actually have a 'grey' value, but unfortunately immediately after the
> > 10 patch we're back into the blacks.
> > 
> > I'll keep digging- I had hoped to use some of the BO only curves as
> > supplied with QTR but none of them help me with the more refined
> > portions of the greys.  Right now I have white, dark medium grey, and
> > black within 20% on the 21 step wedge.
> > 
> > Jason
> >
>