QTR-Quadtone RIP

I'm coming back to Eboni6 after an attempt to restart my darkroom career. Not quite a failure, but certainly not a smashing success either!

I have new carts for the 2400 as well as the inks, just trying to make sure I have my head straight before I start filling and replacing.

After reading Paul's various documents on the K3 printers with the Eboni/HP combo, I think I want to set up my 2400 with a straight Eboni6 approach and no HP inks.

1) Would there be any problem with simply filling the unused C and LC carts with the yellow dilution just to have something in there, or does that matter?

2) I plan on making a profile using Paul's "EbHP13-HPR-Eb6.qidf" as a starting point, removing the linearization numbers, printing a new 21 step test and reading that with a densitometer to relinearize. Does that sound like a good strategy?

Parker Smith <parkersmithphoto@...> wrote:

> **
>
>
> I'm coming back to Eboni6 ... for the 2400 ...
>
> After reading Paul's various documents on the K3 printers with the
> Eboni/HP combo, I think I want to set up my 2400 with a straight Eboni6
> approach and no HP inks.
>
1) Would there be any problem with simply filling the unused C and LC carts
> with the yellow dilution just to have something in there, or does that
> matter?
>
I recommend the LM and M inks for your LC and C positions, respectively.
 These will duplicate what is already in your setup, but they would have
several useful functions.  First, they are similar in density to the HP
inks that were in those  positions in my 7800-EbHP setup <
http://www.paulroark.com/BW-Info/7800-EbHP-2013.pdf>.  That should allow
profiles from the 7800 setup to be very close to correct in terms of their
printing densities without any basic changes needed, aside from the
recommended re-linearizations.

Second, having the LM and M inks installed there would allow the Epson
driver, preferably with an ICC made with QTR's Create ICC-RGB, as an
alternative workflow to the rip.  From some applications, it's  handy to
have the Epson driver produce decent images.

Additionally, the two additional inks can be used for greater smoothness
(though they are not normally needed).


>
> 2) I plan on making a profile using Paul's "EbHP13-HPR-Eb6.qidf" as a
> starting point, removing the linearization numbers, printing a new 21 step
> test and reading that with a densitometer to relinearize. Does that sound
> like a good strategy?
>

The basic settings are probably close between the 2400 and 7800.

For the LC and C (duplicate ink) positions, you can set them up as "copy
curve" inks or "toners."   You cannot have duplicate density inks in the
same partition (like "gray" partition).   The EbHP profiles are set up with
"toners."  Note  that when there is a second "channel" of an ink like  a
toner or "copy curve" position  the ink limits are reduced.  For example
with a "copy curve" duplicate, you can simply cut the ink limits for the 2
inks in half, since  there will now be 2 inks firing instead of 1.  The
existing EbHP profiles have done this when there is an HP ink in there as a
"toner."  So, you can probably just leave them in the profile  as is.  The
LM and M inks, as  mentioned above, are probably similar enough to the HP
inks in density  that the profile will work fine as is, with
re-linearization recommended.

Good luck with the transition back to carbon from silver.

If you want a neutral appearing, non-OBA print, be sure to try Epson Hot
Press Natural.

Paul
www.PaulRoark.com <http://www.paulroark.com/>


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

Thank you again, Paul. One last question: should I remove the existing linearization data from the profile before I print the 21 step test?

Parker Smith <parkersmithphoto@...> wrote:

> **
>
>
> Thank you again, Paul. One last question: should I remove the existing
> linearization data from the profile before I print the 21 step test?
>

Yes.  To re-linearize a profile you remove  (clear) and linearization data,
print a test strip, read that test strip, and enter that data into the
profile.

Just out of curiosity  it would be interesting to see what happens if you
print one of the profiles as is, complete with linearization data still in
it.  Then see how linear the test strip  is.  I'd be curious how close the
systems are.

Paul
www.PaulRoark.com





>
>  
>


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

Paul,

Can you suggest one of the existing profiles for the 7800 to use? 

I have the printer set per your recommendations, but it seems most of the profiles you provided have channels turned off, either to do a straight E6, or create a cool tone. I tried to find one where the cyan channels were set as toners; quite possibly I either missed it or don't fully grasp what I'm looking at in curve creator!

Thanks,

Parker

Parker,


> Can you suggest one of the existing profiles for the 7800 to use?
>

I'm not sure what you have.  I have the profiles posted at
http://www.paulroark.com/BW-Info/7800-EbHP-2013-Profiles.zip .

I have the printer set per your recommendations, but it seems most of the
> profiles you provided have channels turned off, either to do a straight E6,
> or create a cool tone.
>
You're correct.  I had not reviewed them recently.  What I do a lot is use
a 100% Eb6 profile plus an HP (cool) profile, mixing them as needed with
the QTR sliders.  You can do the same thing to get all the inks firing.
 I'd overweight the Eb6 profile, since that is where most of the inks are.
 That is, have the Eb6-only profile be about 75% and the HP/cool profile be
25%.  This is probably the easiest way to use all the inks.

There is an ICC that is a blend of both.

I tried to find one where the cyan channels were set as toners; ...
>
There may not be any QTR profiles set up that way.  To do so, just activate
the second LM and M, in the LC and C positions, call them toners, have
their densities be the same as the other LM and M, and assign 1/4 of the
dilute ink limit to them and 3/4 to the gray channel dilute inks.  These
will need re-linearizing, I assume.

For at least some of the profiles I have overweighted the Y (2%) ink.  For
these, to keep that approach (which holds down the highlight warmth) I
would just have the second LM and M copy the curves from the original LM
and M and have the ink limit for each half of what is currently assigned in
the profile.

Hope this helps.

Paul
www.PaulRoark.com


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

Paul,
For the R2400, I used your existing EbHP13-EpsHPn-HighY profile and
added the C and LC as copy curves, splitting the ink limits in half
between the two similar inks. After printing the 21 step test, I used a
Spyder to read the following values:
0.000.040.100.170.230.320.410.480.550.630.700.780.850.941.011.081.181.27\
1.371.441.62
Adding these to the linearization panel and printing the new profile
with some circular test targets shows that it's very good, but not quite
perfect.If I wanted to re-relinearize, would I leave these numbers in,
print a new test strip, read those numbers and then replace the existing
numbers? Or is it better to manually tweak the numbers to do the final
smoothing? Or, some third option that hasn't occurred to me?  [;)] 
Thanks again!


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

Parker Smith <parkersmithphoto@...> wrote:

> **
>
>
> Paul,
> For the R2400, I used your existing EbHP13-EpsHPn-HighY profile and
> added the C and LC as copy curves, splitting the ink limits in half
> between the two similar inks. After printing the 21 step test, I used a
> Spyder to read the following values:
> 0.000.040.100.170.230.320.410.480.550.630.700.780.850.941.011.081.181.27\
> 1.371.441.62
> Adding these to the linearization panel and printing the new profile
> with some circular test targets shows that it's very good, but not quite
> perfect.If I wanted to re-relinearize, would I leave these numbers in,
> print a new test strip, read those numbers and then replace the existing
> numbers?
>
Print the profile with the Linearization tab totally cleared.  Then use the
resulting numbers to plug into the boxes.

Paul
www.PaulRoark.com


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

> Print the profile with the Linearization tab totally cleared.  Then use the
> resulting numbers to plug into the boxes.
> 
> Paul
> www.PaulRoark.com
> 
> 
> [Non-text portions of this message have been removed]
>

What I meant is that I did print it out without the Lin #s, read with the spectro, plugged in the resulting #s into the linearization panel, and then printed out a new chart which looked like it might still have some gradient issues.

Can I read the new numbers and then replace the existing numbers?

Parker Smith <parkersmithphoto@...> wrote:

> **
>
>
>
> ... I did print it out without the Lin #s, read with the spectro, plugged
> in the resulting #s into the linearization panel, and then printed out a
> new chart which looked like it might still have some gradient issues.
>
> Can I read the new numbers and then replace the existing numbers?
>
No, it needs the numbers that are produced by the profile with the
linearization boxes empty.

If there is a problem it's usually due to a pre-linearization curve that
has a sharp turn in it.  I always graph the pre-linearization Lab L values
in Excel to see if the curve is reasonably smooth, even if not linear.  If
I see a sharp kink, I try to remedy it by changing some of the numbers on
the Ink Setup page or, better, taking manual control of part of the profile.

I'm not sure how you have the system set up, but what I do sometimes is
have 2 channels or partitions of the carbon ink.  Then with the one of them
I turn it into a set of manually drawn curves.  This is much easier that
you'd think.  You simply increase the ink limits of the channel you want to
draw to 100%, hit "show curve," and do a screen grab of those curves.
 Print the  graph, and enter the coordinates into the Point List for the
curve you want to take manual.  The ink limits will be what you had in the
original curve, at least initially.

Once you have one (and only one) of the channels manual, you can alter the
points to correct for the glitch in the overall curve.

In general, that pre-linearization curve should be relatively smooth.
 Don't Expect miracles from a smoothing algorithm that only has 21 of 256
input points.

Paul
www.PaulRoark.com


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

To follow up on this topic, I was struggling to get the QTR curves fully linear, and (finally) tried Paul's advice about using the Epson driver with an ICC.

I made test strips with both the color controls and the Advanced BW mode, and then made ICCs from both of these. After printing the Northlight BW test file, the color driver looked good and the ABW mode with ICC was just astonishing. There was detail from the brightest highlights down in to the very deepest shadows. Yes!

If anyone out there is trying an Eboni setup, I highly recommend taking Paul's advice and working first with the Epson driver in ABW mode. His instructions for making an ICC with a flatbed scanner and a gray card gave terrific results.

I guess whatever works for you is best.

Let me make a few comments on this, however.

First, many of the newer Photoshop and OS combinations will not allow ICCs
to be used with Color Controls or ABW.  They see that as combining two
profiles.  (It's too bad they have "dumbed down" the systems.)

Second, using ABW with Eboni-6 does not make very efficient use of the
inks.  I would think the LK and LLK inks are being use relatively more than
the others.  But, again, if it works for you, use it.

Paul
www.PaulRoark.com

On Tue, Apr 2, 2013 at 8:21 AM, Parker Smith <parkersmithphoto@...>wrote:

> **
>
>
> To follow up on this topic, I was struggling to get the QTR curves fully
> linear, and (finally) tried Paul's advice about using the Epson driver with
> an ICC.
>
> I made test strips with both the color controls and the Advanced BW mode,
> and then made ICCs from both of these. After printing the Northlight BW
> test file, the color driver looked good and the ABW mode with ICC was just
> astonishing. There was detail from the brightest highlights down in to the
> very deepest shadows. Yes!
>
> If anyone out there is trying an Eboni setup, I highly recommend taking
> Paul's advice and working first with the Epson driver in ABW mode. His
> instructions for making an ICC with a flatbed scanner and a gray card gave
> terrific results.
>
>  
>


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