Inkjet Mall Digital Negative Blog

Discussion in 'Digital Negatives' started by donbga, Jan 6, 2011.

  1. donbga

    donbga Member

    Messages:
    3,054
    Joined:
    Nov 7, 2003
    Shooter:
    Large Format Pan
  2. sanking

    sanking Member

    Messages:
    5,437
    Joined:
    Mar 26, 2003
    Location:
    Greenville,
    Shooter:
    Large Format
    Sounds pretty interesting. I believe that what he has done is create a perfectly linear profile for Pictorico with a Dmax of around 3.0. If you want to use the profile for a specific process I suspect you will have to create an .acv curve to linearize your process since editing the Cone profiles would be very difficult. I don't believe you coud even insert a curve in the profile, or do plus and minus ink adjustments as we do when editing with the QTR profiles.

    Sandy
     
  3. piticu

    piticu Member

    Messages:
    100
    Joined:
    Nov 30, 2005
    Location:
    Slobozia, Ro
    Shooter:
    4x5 Format
  4. OP
    OP
    donbga

    donbga Member

    Messages:
    3,054
    Joined:
    Nov 7, 2003
    Shooter:
    Large Format Pan
  5. gmikol

    gmikol Member

    Messages:
    618
    Joined:
    Mar 2, 2009
    Location:
    Vancouver, W
    Shooter:
    35mm
    Interesting enough, but I'm a little wary of this one-size-fits-all approach. He's basically shoe-horning a very interesting application of using piezo inks to make backlits into a digital negative application. I think it fails, and here's why:

    Using a PS curve that aggressively (taking 100%K down to like 60%K) in a PS curve, and also raising the image white (paper black) point to something like 10%K like Jon has done in that blog post throws away fully 1/2 of the available tonal values. Seems like a big risk for posterization. And the black point of a paper should be set by exposure time (in this case, reducing it a bit) Otherwise, it's just wasted ink and sacrificing more tonal values. I'm amused by how he crows about how his profile can represent all 256 shades in a ramp, and then throws half of them away.

    I know that the Piezo curves can't be edited like regular QTR curves, but they *do* they respond to the "Advanced Adjustments" controls in QTR (at least in QTRgui), Ink Limit and Gamma. This seems like it would be a better choice to get your negative black point (paper white point) set, and to get the curve (mostly) linear, before you straighten it out completely (or add any preferred tonal curve) with a PS curve.

    Those are my thoughts...they are worth what you paid for them...

    --Greg
     
  6. Ben Altman

    Ben Altman Member

    Messages:
    205
    Joined:
    Jan 19, 2007
    Location:
    Ithaca, NY a
    Shooter:
    Large Format
    I agree with Greg.

    As Sandy King and I have been figuring out, some Cone inks can be used for digital negs, and we will be doing more to figure out which ones are best. One big advantage these monochrome inks have over colored inks is that they are all... much the same color - and thus presumably each have similar spectral response to UV (and visible light for that matter). This makes it easier to do rough assessments of a profile with transmission measurements of the negative, saving some test strip printing, and also means profiles are more likely to transfer from one exposure unit, and even process, to another with reasonably small differences. In other words my profile developed with BLB bulbs may be close with your Amergraph, etc.

    In general, I suspect that a perfect inkset for digital negs is not a good one for positive inkjet prints. Jon Cone refers to this when he says that a narrow range of inks is best for negs; this is my experience also - no need for super-light or super-dense inks. So while one can make very good negatives with standard inksets, whether Epson, Canon, HP, or Piezography, the best results are probably to be gained with a dedicated machine that allows all the ink channels to be useful.

    One could do this in an Epson inkset by moves like, for instance, running an extra LC cartridge in the LM slot (with a chip transplant on the cartridge - what warrranty? :smile:). And maybe both PK and MK would be useful. I'd have to think it through more.

    Anyway, I don't see a big gain in what Jon is suggesting right now.

    Ben
     
    Last edited by a moderator: Jan 8, 2011
  7. Ron-san

    Ron-san Member

    Messages:
    154
    Joined:
    Oct 28, 2006
    Location:
    Seattle, WA
    Shooter:
    4x5 Format
    Hello friends--

    I have read through Jon Cone's blogs. I have no comment on the part pertaining to the use of K7 inks to make standard positive inkjet prints. But the part pertaining to digital negatives was disappointing.

    First, some background. In the normal use of QTR you only have to plug a small set of numbers into a textfile ( limit for each ink, crossover point where each lighter ink gives way and is supplanted by the next darker ink, and so forth) to completely specify how all the inks are used. With a little practice this textfile (.txt file ending) format is easy for human eyeballs to comprehend and work with. Then, when you install the textfile, a bunch of Roy Harrington's little subroutines take over, and based on the numbers you have plugged, compiles this information into specific instructions to be sent to the printer and which take over the job of smoothly blending one ink into the next.

    This compiled information is in something called a .quad file and these files are found in a folder on your harddrive where QTR stores them. If you open one of these .quad files (drag its icon on top of the textfile icon on your toolbar) you will find something wondrous. Near the top you will see the letter K (for the dark black ink) with number after number ranging down below it. At the beginning these numbers are all zero, but if you scroll down you will begin to see increasingly large numbers until they stop. There are 100 of these numbers and each one specifies exactly how much K ink will be laid down at each percentage step of tonal value. Keep scrolling down. There will be 100 positions and values for the Y ink, the M ink and so forth for all of the printer's inks. If you have the mind of a Dalek, or like to immerse yourself in homemade Xcel spreadsheets (like Ben Altman -- hi Ben) you can skip the textfile part of QTR and write the .quad file yourself by laboriously typing in each number. But most of us find it far easier to work with the .txt files and let Roy's subroutines compile things into the .quad version.

    Back to the Inkjet Mall inks. When Cone first started using QTR to print his inks he supplied profiles written in the human-friendly .txt format. In these files the limit for each of the seven inks was specified as well as the crossover number for each ink (ie, he wrote the profile as a seven part system). And, at the end, he applied one of Roy's linearization functions to fine tune and straighten everything out. But, at some point, he stopped supplying users with the human readable .txt file and instead only supplied the Dalek-friendly .quad version. Which means, you and I would find it extremely difficult to take one of his profiles and tweak it or rewrite it to make digital negatives. I have no idea why he did this.

    Onward to K7 inks and digital negatives. If I were using QTR and the K7 inks to make digital negs, I would do pretty much what I do with the Epson UC inkset. I would move the dark ink limit up and down to set negative contrast, then fiddle with the limits of the lighter inks to get the midtones on my final print to be roughly linear (the tones on the negative would be definitely non-linear), and then I would apply a Photoshop correction curve to the Ink settings (not to the image file) to finish and fine tune linearization (I think and hope this is more or less the approach that Altman and King are taking). Cone does not do this. First, he fiddles his profile so that the inks print linearly on Pictorico (this is not a terribly useful step since the final negative will be, for most alt processes, decidedly non-linear). Then everything else is accomplished by applying a Photoshop correction curve to the image file. This is retrograde and ignores everything we have learned about making digital negatives in the last ten years. What is most sad is that this approach makes no use at all of the powerful tools available in QTR to set neg contrast and non-destructively linearize print midtones.

    Is there any reason to mess with the K7 inkset for digital negatives? Sandy King seems to be hoping that the K7 inks will print to higher resolution than the Epson inks. That may be an issue with carbon printing, but it is definitely a non-issue with palladium and other matte prints. In principle, 7 overlapping inks should be able to print a very smooth and grain free negative (certainly everyone seems to agree that they make very nice positive prints). But, the way I use the Epson inks, I also have seven inks overlapping to various degrees (by having all the dark inks follow the K ink, and all the lighter inks follow the distribution of the LK ink). The Epson inks and QTR make very good negs and prints. But someday I would like to see a rigorous side by side comparison.

    And, for myself, since I cannot afford the space to dedicate one printer just to negatives, using the Epson inks means I can print Xmas cards, my wife's bird prints, and other stuff as well as making good negs.

    Sorry for the long post, but I just had to get this off my chest. Cheers, Ron Reeder
     
  8. Ben Altman

    Ben Altman Member

    Messages:
    205
    Joined:
    Jan 19, 2007
    Location:
    Ithaca, NY a
    Shooter:
    Large Format
    Hi Ron...

    To clarify, in suggesting a dedicated printer I was only talking about looking for the nerdy ultimate... Sandy's desire is in fact to have a dual-purpose Piezography printer that he can use for B&W inkjet prints as well as digital negs. I think Jon has suggested some other inks to try, which we'll get to when Sandy has time to work on this again. But we've gotten good negs and Carbon prints from the standard K7 Selenium inks, and if it's good enough for Sandy...

    Regular inksets will do a fine job - particularly, as you say, for the more matte surfaces. I think you'd agree that a couple or three of the Epson inks don't block much UV, though.

    QTR is what makes the difference, as you say, however one controls it. And believe me, I don't type numbers into a .quad file... Getting my numbers to load semi-automatically through the .txt file was one of the first things I figured out!

    Best, Ben
     
  9. pschwart

    pschwart Member

    Messages:
    1,131
    Joined:
    Jul 15, 2005
    Location:
    San Francisco, CA
    Shooter:
    Multi Format
    and another approach is to take a black ink that is known to be a good UV blocker, make your own dilutions for the additional channels, and use refillable cartridges or a CIS. Sounds like a pain but it only took about an hour -- ink is mostly distilled water with a tiny bit of glycerin and a wetting agent. I did this using MIS Eboni so I could make prints using 100% carbon pigment. The prints were beautiful but a bit too warm for my taste. I never tried this with digital negatives.
     
  10. sanking

    sanking Member

    Messages:
    5,437
    Joined:
    Mar 26, 2003
    Location:
    Greenville,
    Shooter:
    Large Format
    I would not make any assumptions about the Inkjet Mall digital negative profile. In theory it has the potential to provide greater smoothness and sharpness, especially in the upper mid-tones and highlight areas. Whether or not that will turn out to be true or not only time will tell, but from my perspective it seems worth trying, and I already have the mixed shades that are required. So when I get the time what I propose to do is following a dual track, continuing my work with Ben with the ink set proposed by Jon Cone for digital negatives, including a change to PK, and simultaneously printing a special step wedge for Jone Cone to see if his profile for digital negatives can be adjusted to work for the 7600.

    Although my original intention was to use the 7600 with the K7 set for both digital negativs and inkjet printing I am now trending toward dedicating it to digital negatives *if* I can see and advantage over the profile that Ben an I already worked out with the regular K7 set with the MK installed.

    So far what I know for fact is that the QTR drivers gives higher resolution than the Epson driver, and this is true for both the Epson inks and the K7 inks.

    I agree with Ron that if you are working with pt/pd or some other metal salt process printed on art papers the issue of sharpness is mostly moot since even the Epson drivers gives higher resolution on Pictorico than you will get on paper with these processes.

    Sandy
     
  11. sanking

    sanking Member

    Messages:
    5,437
    Joined:
    Mar 26, 2003
    Location:
    Greenville,
    Shooter:
    Large Format
    Well, I can definitely vouch for one point that Jon Cone made in his article about using the mixed K7 shades in two slots to make digital negatives. He said that you would not be able to print regular digital inkjet prints with the mixed shades and I just confirmed this. I changed over the shades a couple of days ago to see if I would be able to use the Cone profile but instead of purging the inks, which is very wasteful, I decided to just print through the inks in the lines until the old shades were cleared. So between yesterday and today I printed fourteen 24X36" prints to clear the lines and they all looked great, until the new shades kicked in on the last print, and there is definitely posterization as Cone promised. The place on the print where the change over took place was sudden, and quite dramatic.

    Sandy
     
    Last edited by a moderator: Jan 27, 2011
  12. Garrettsnix

    Garrettsnix Member

    Messages:
    1
    Joined:
    Aug 12, 2016
    Location:
    Bulgaria
    wow. big dif between the 8x10 and the 16x20. It might be one of those digital printing things where the 16x20 has to be run on a larger machine so the price goes bananas once you bump it up.

    Im not looking based on price alone but I wanted to have an idea beforehand. Both so I could make an educated decision about forgoing the inkjet and also so I didnt end up paying 5 times what you are and think its normal. But around 2 a print is good for me.

    Do you know what kind of prints they are? Iris, digital c?
     
  13. John Dean

    John Dean Member

    Messages:
    9
    Joined:
    Apr 1, 2017
    Location:
    Atlanta Ga.
    Shooter:
    4x5 Format
    The newer Epson printers, like the 3880 or 3800 or the7800, 7880, 7890, 7900, etc have much better resolution than the 7600. However, as Sandy has just stated, with Jon's inks you are going to have a much better rendition of resolution on the film than any 19th century alternative process on paper that I've ever used can exhibit. So then you come down to linearity and tonal ramp control, and dmax. All I know is that I tried the Bulkholder curves on an ancient Epson 1280 nearly 18 years ago and the prints on platinum/palladium with platine paper were beautiful, and I didn't even know what I was doing. ( I am a professional digital pigment printmaker for 18 years ). This week I plan to order the set of K7 Piezo inks to put in a 7800 printer. This will allow me to have the full tonal set plus the new darker PK and MK inks which I will be able to use for OHP negs, gloss fiber media, and matte rag media. I'm actually going to let inkjet mall create my platinum and gum negative curves, and then make my own after I test them. I also got an email recently from Mark Nelson telling me that his system works well with Jon's piezo inks. So I'll try that also. A guy named Angel invented a work flow for using the HPZ 12 channel inks for OHP film with a greenish hue to the inks. Here is some of Angels work using that process - http://en.albarrancabrera.com/cosmos/ I am not sure how many channels he is actually using. HP had no interest and quit promoting it.
     
  14. John Dean

    John Dean Member

    Messages:
    9
    Joined:
    Apr 1, 2017
    Location:
    Atlanta Ga.
    Shooter:
    4x5 Format
    I'm not exactly sure of what inks you are referring to and how many slots you are using ( you can use K6 in that printer and eliminate the very light gray K7 to use for something else ) and see no difference in your prints and give yourself an additional ink channel to use for whatever you want - the software curves will determine which inks are turned on and which are not used ). The curves have to be created and selected differently if the shades are moved around for different purposes for different inksets. I do know the newer printers have 8 - 10 slots ( not 7 ) to choose from when combining shades for various purposes. Curves and ink assignments for film and curves for gloss papers vs matt papers are obviously very different animals. I am just in the very beginnings of working with this, but my impression right now is that with an 8 channel printer, using K6 piezo inks plus the film black and the film gloss enhancer on board is no problem. I know one thing, I'm going to find out before I buy any of the inks. IF you are correct I will have to set it up for negatives only, but I hope that is not the case.






     
  15. John Dean

    John Dean Member

    Messages:
    9
    Joined:
    Apr 1, 2017
    Location:
    Atlanta Ga.
    Shooter:
    4x5 Format
    Oh hell I just saw that these posts I was commenting on are from 2011 ! Damn, no discussion since then? Anyway sorry, this system is way more developed and refined now.
     
  16. CMB

    CMB Member

    Messages:
    107
    Joined:
    Mar 17, 2011
    Location:
    Santa Cruz,
    Shooter:
    Multi Format
    Only two inks: Green and (Photo) Black
     
,