r1900 Quadtone RIP help

[5d-creative]

Like any other newbie I will make my first post to the forum with a request for help. Hopefully in the future I will be able to return the favor.

I have a epson r1900 printer with the standard UC ink-set. I am trying to follow Ron Reeder's digital negative .pdf. Ron, if you are listening thanks for the .pdf, I'm sure once I get a little help it will be well worth the cost.

I am having trouble making changes to the .txt profile files. Specifically I am trying to map Clay's 2200 profile to the uc_matte profile provided with the Quadtone driver. When I try to install my changes I get an error in the terminal window that says:

invalid ink name lc used in copy-curve_lc

I suspect I have the incorrect ink names but honestly don't have a clue as to what the correct names are.

Could anyone advise a course of action? Maybe I need a different profile than Clay's 2200 profile but I couldn't locate another one specific to the r1900. If you know of one a link would be greatly appreciated.

I am eager to get this rolling as I have made some good prints with traditional negatives but I have an entire series made in digital that is begging to be printed in palladium.

A kings ransom to any one who can help.

Cheers,
Reathel
www.5d-creative.com

 

[gmikol]

There is no light cyan (LC) on the R1900. Any curve set that uses C and LC will need to be re-calibrated to use only cyan.

The correct ink names are: MK C M Y R OR PK GL
(mate black, cyan, magenta, yellow, red, orange, photo black and gloss optimizer).

Furthermore, I don't know if this was an accidental typo, but the syntax should be "COPY_CURVE_LC" (or whatever ink color), and not "COPY-CURVE_LC".

Hope this helps you out a bit.

--Greg

 

[5d-creative]

@gmikol - Thanks. That helped. I successfully printed a negative that looks visually correct. I'll do a test print tomorrow and see where I'm at.

Cheers,
Reathel

 

[gmikol]

No problem...glad to help out. QTR is an incredibly powerful piece of software, but it has a steep learning curve. Don't fret too much about getting your negative "perfect"...if there is separation *in your final print* between the steps in a digital stepwedge (18 or 21 or more steps, evenly spaced), and your paper white and dmax look OK, then you can let the linearization sort the rest out. But there's a little bit of mathematical gymnastics required, if you haven't already read about it. QTR is actually designed for making B&W inkjet prints, and the linearization process doesn't work right for negatives without a little tweaking. There are some tutorials out there on exactly how to accomplish that, but I can't name any at the moment.

One more thing to keep in mind, though...the R1900 inks seem to be somewhat different than both the R1800, as well as the K3 inks found in the other x800 and x880 printers. I used to use an R1900 for diginegs, and I found that only black (PK) and yellow were really useful, and that cyan contributed very little. On the K3 inks (and presumably, the original Ultrachrome inks that the 2200 uses), cyan contributes more UV blocking. My testing was done for carbon transfer, however, which is most sensitive in the sub-380nm range. I think Pt/Pd has more sensitivity above 400nm (someone please correct me if I'm wrong), so it may be a little different for you, but at least try out a curve with no cyan to see if it makes much of a difference for you. I know that the absorption of the K3 inks varies greatly in the region between 350nm and 425nm.

Good luck--

Greg

 

[5d-creative]

Thanks...

Greg,

I printed out a negative and managed to get a result that was less than impressive, i.e. lots of room for improvement.

After your comments I see part of the error may have been choosing mk as my primary ink instead of your suggested pk. Regardless I am still feeling a bit in the dark on how to navigate the curve profile. Below you'll find the curve profile I created and the results for input/output per Ron Reeder's suggested measurement process. Any help you or another can provide would be appreciated.

Also of note is that the exposure time is quite long (1hr) because of the light source I am using (T5 florescent grow bulbs).

curve profile...

# QuadToneRIP curve descriptor file
#
# for R1900 with ultrachrome inks

CALIBRATION=NO
GRAPH_CURVE=yes

#
# number of inks must be 4, 6, or 7
# the ink limits are percentages
# usually they are all the same but they can be individually set
#
N_OF_INKS=7
DEFAULT_INK_LIMIT=60

LIMIT_MK=30
BOOST_K=40
LIMIT_C=4
LIMIT_M=3
LIMIT_Y=10
LIMIT_r=3
LIMIT_or=3
LIMIT_pk=40
LIMIT_gl=0
#
# Describe Usage of each Ink: K,C,M,Y,LC,LM,LK
# All Inks of Printer must be listed
#

#
# Gray Partitioning Information
#
N_OF_GRAY_PARTS=2
GRAY_INK_1=MK
GRAY_VAL_1=100

GRAY_INK_2=pk
GRAY_VAL_2=35

GRAY_INK_3=
GRAY_VAL_3=

GRAY_INK_4=
GRAY_VAL_4=

GRAY_INK_5=
GRAY_VAL_5=

GRAY_INK_6=
GRAY_VAL_6=

GRAY_INK_7=
GRAY_VAL_7=

GRAY_HIGHLIGHT=0
GRAY_SHADOW=0

GRAY_GAMMA=1
GRAY_CURVE="0;2 4;7 11;13 20;19 30;24 40;29 50;36 60;42 70;50 80;60 85;68 90;75 96;90 97;93 98;95.5 99;97 100;98"

#
# Toner Partition Information
#
N_OF_TONER_PARTS=0
TONER_INK_1=
TONER_VAL_1=
TONER_INK_2=
TONER_VAL_2=

TONER_HIGHLIGHT=10
TONER_SHADOW=10

TONER_GAMMA=1
TONER_CURVE=

#
# Toner 2 Partition Information
#
N_OF_TONER_2_PARTS=0
TONER_2_INK_1=
TONER_2_VAL_1=
TONER_2_INK_2=
TONER_2_VAL_2=

TONER_2_HIGHLIGHT=10
TONER_2_SHADOW=10

TONER_2_GAMMA=1
TONER_2_CURVE=

#
# Unused Inks
#
N_OF_UNUSED=0
UNUSED_INK_1=
UNUSED_INK_2=
UNUSED_INK_3=
UNUSED_INK_4=
UNUSED_INK_5=

#
# the following is a special UltraChrome ink flag it copies the
# light-black curves to the light-cyan, light-magenta and yellow inks.
# this insures that the LC and LM inks are used to directly neutralize
# the color of the light-black.
# the color balance of inks is controlled by the LIMIT_... values above.
#

and here are the input outputs...
input output
0 59
10 69
20 71
30 74
40 78
50 82
60 86
70 88
80 93
90 96
100 99

So as you can imagine the print step tablet was quite dark. I've created another profile but would be curious to here any advice before I print with it.

btw, I live down in Eugene, OR.

Cheers,
Reathel

 

[gmikol]

Are you on a Mac? I'm not used to reading the files directly, as I use QTRGui on Windows.

You don't really want to use MK ink at all for diginegs, from what I understand. It doesn't adhere well to the OHP material, and the neg can easily get smudged. That, and because the MK is so dense, you can end up with images looking very "grainy". So there's only 1 gray ink, PK, which should have a gray value of 100. Ink Limit? You'll have to figure that out for yourself, I couldn't guess where to start, having never printed Pt.

Here's a procedure that I used to try to figure out your maximum ink density, for a given paper/sensitizer/exposure combination:

1) Create a square step wedge (white to black) in Photoshop (or other imaging program of your choice). I use 18 steps, because that's every 15 units of RGB from 0 to 255. Make sure it's at least 3" square, and 4 or 5 will make it easier to read.
2) If you want to add a "safe edge" or labels or anything around the image, make sure the step wedge itself stays in the center of the image.
3) Save the image, then rotate it 90 degrees, and save it with a new name. Now you have 2 copies, the original and the rotated.
4) Create a completely linear profile using PK only. Gamma=1, limit 100, highlight=0, shadow=0, no gray curve.
5) Print the step wedge.
6) Copy the above profile, but make it for yellow only, and print the rotated step wedge in the same position on the same page.

Now you have a square with every combination of ink from 0PK/0Y (clear Pictorico), to 100% PK/100% Y which you can print, and then you'll see all the different combinations that will yield a paper white with your sensitizer/paper/exposure. I'd post pictures, but I'm away from my other computer, which has all that stuff on it.

If you want, you could also try to print a step wedge with 100% PK one way, and a profile with 50% Y and 50% C going the other way, as well, to see if that is better. Some people have had good success with using green as a blocking color.

Hope that helps some. Good luck...

--Greg

PS--In case no one warned you in advance, be prepared to burn somewhere between a few sheets and a whole pack of Pictorico before you really get this dialed in. The good part is, once you're dialed in, you're all set. And if you change your paper or sensitizer or whatever, the second time through should be much faster.

 

[R Shaffer]

Really dark sounds like an understatement.

Have you determined your minimum exposure to get to D-max with the OHP your are using. I'm not familiar with your exposure unit, so maybe it does take a hour. If your really overexposing, then your burning through your blocking color.

In my experience, MK ( with 3800 ) works fine.

You have an unusual two part curve in that you using MK & PK. I think it is more typical to have a two part curve with MK & LK or PK & LK. So your 1st part, MK ( or PK ) would be the darkest steps on negative ( highlights on print ) and the 2nd part LK is for your lighter steps ( shadows on print ). Using LK prints smoother.

Also, yellow is a good blocker.

Lastly, I don't see where your using the other inks. My qidf files have statements like these

COPY_CURVE_C=K
COPY_CURVE_M=K
COPY_CURVE_Y=K
COPY_CURVE_LC=LK
COPY_CURVE_LM=LK

HTH

 

[5d-creative]

Thanks...

Greg, and R. SHaffer,

Thanks for the advice. I'll digest this and print some new wedges tonight. R. Shaffer, I don't have an LK ink so from what I have just read the yellow sounds like the best substitution. The MK ink seems to adhere pretty well to the transparency film. I think I'll try the pure linear prints of MK, PK and Yellow and see what that looks like.

The min. for black is unfortunately about an hour, but that's what I have to work with and I have even seen some suggestions that longer exposures can have some benefits.

A calculus professor once told me that just beyond confusion is understanding, I can only hope this is true.

Cheers,
Reathel

 

[R Shaffer]

Greg, and R. SHaffer,

Thanks for the advice. I'll digest this and print some new wedges tonight. R. Shaffer, I don't have an LK ink so from what I have just read the yellow sounds like the best substitution. The MK ink seems to adhere pretty well to the transparency film. I think I'll try the pure linear prints of MK, PK and Yellow and see what that looks like.

The min. for black is unfortunately about an hour, but that's what I have to work with and I have even seen some suggestions that longer exposures can have some benefits.

A calculus professor once told me that just beyond confusion is understanding, I can only hope this is true.

Cheers,
Reathel

If I did not have LK, then I would look to maybe cyan as the second part of the curve. Your probably gonna need lots of black + yellow to get enough blocking.

Have you printed the QTR Calibration page? That will give you an idea on blocking ability.

 

[clay]

I posted a QTR profile for the 1400 (it uses the same Claria ink set as the 1900) a while back. This might be a starting point for you:
Dead Link Removed

 

[5d-creative]

progress

To everyone, thanks for your advice. I printed a Quadtone Rip calibration page to see how the inks blocked the UV light. I'll include the scan of the image with this post. In summery it appears the mk ink produces the strongest blocking starting at around 90% it blocks almost all of the UV light. The highlights are not completely clear but that may have something to do with my clearing baths. I'll place something opaque on the next test to verify this. The MK has by far the most pronounced grain of all the inks.

The PK ink is the next best at blocking out the UV light, has much finer grain characteristics, and has the best delineation between the steps on the print out.

The Y is not as good as the PK in blocking out the UV but has a smoother transition between the steps on the print out.

So I'm thinking that using the MK for the shadows and the PK for the mid tones and highlights makes sense. I'm not certain exactly how to do this in the profile curve and any advice would be appreciated.

Clay, thanks for the profile of the 1400. The r1900 uses UC Hi-Gloss 2 Inks so the 1400s profile I guess won't be helpful. But I appreciate the thought.

Cheers,
Reathel

 

[gmikol]

I posted a QTR profile for the 1400 (it uses the same Claria ink set as the 1900) a while back. This might be a starting point for you:
Dead Link Removed

Clay...the R1900 does not use a dye (Claria) inkset. It uses an "Ultrachrome Hi-Gloss 2" pigment inkset. AFAIK it is unique in the Epson world, differing even from the original Ultrachrome Hi-Gloss in the R1800.

--Greg

 

[gmikol]

IMO a smooth transition in the raw ramps isn't *that* valuable, since most people will be using a correction curve in the RIP or using QTR's own linearization. So as long as the transitions are not grotesque (or worse, inverted), the linearization will do the rest.

Also, I may differ in opinion from many QTR/digineg practicioners on this point: I don't see the point (or the benefit) from partitioning different color inks to print highlights vs. shadows, as Reathel (5d-creative) is planning on doing. In a negative process, it's difficult to partition differently-colored inks, except by eye, so getting the transition smooth might be difficult. It also means laying down more total ink, in most cases, vs. using a better-blocking color for the shadow (low ink density) tones. This, especially on the R1900, has the potential to lead to pizza wheel marks, since the platen gap is not adjustable, like it is on, say, the 3880.

I'm not trying to start a which-is-better debate (war?). It just seems like 5d-creative is fairly new to this, and I wanted to present an alternate method, which I am happy with for my personal work.

As always...good luck, and be sure to keep us up to date...

--Greg

 

[clay]

I think I would just stick with the pK, Y, and O in combination to get your pure white and just fix the curve with the black boost and gamma with a final tweak using the gray curve. That grittiness with the mK may be a pain in the butt to deal with in the midtones.

 

[pschwart]

The 1400 uses Claria (dye) ink; the R1900 uses Ultrachrome (pigment) ink.
I wouldn't expect these to behave at all alike.

I posted a QTR profile for the 1400 (it uses the same Claria ink set as the 1900) a while back. This might be a starting point for you:
Dead Link Removed

 

[clay]

Thanks! I was mistaken about its relationship to the 1400. The dye printers and the pigment printers do behave very differently. That said, I still think that I would try to avoid using the mK ink because of the graininess issue.

The 1400 uses Claria (dye) ink; the R1900 uses Ultrachrome (pigment) ink.
I wouldn't expect these to behave at all alike.

 

[5d-creative]

Thanks, I am currently printing out a new step chart. I think I have a long way to go yet. The step chart looks a bit thin but I'll give it a go anyhow. I'll report back results.

FYI I had the MK and PK mixed up, the PK (photo black is the grainy ink).

 

[pschwart]

My experience has been that matte black (MK) can be grainy; photo black (PK) has always produced smoother negatives for me.

FYI I had the MK and PK mixed up, the PK (photo black is the grainy ink).

 

[sanking]

My experience has been that matte black (MK) can be grainy; photo black (PK) has always produced smoother negatives for me.

My experience on the Epsons 2200 was that the MK and PK both produced very smooth results in terms of grain. My take on why the MK produces grainer results is because it is so much stronger a UV blocker than any other color and if you don't compensate for this in the profile with a lower ink limit you will find that all of your printing is with the MK ink. To get best grain, either with the MK or PK, you need to adjust the ink output so that you have two or three inks that give about equal blocking power. This generally involves setting the ink output for the Y and M higher than for the K.

I don't know a thing about the R1900 so can not offer any advice on that subject.

Sandy