Digital Negatives with QuadToneRIP

Note: Since early in 2016, I have been making my digital negatives using the PiezoDN system from Jon Cone and Walker Blackwell at InkjetMall. I believe it is in a class by itself, however I realize that a dedicated solution such as PiezoDN which requires a special inkset and special software is not for everyone.

This method of making digital negatives is for those using Epson printers with the standard Epson color inkset. The only special software necessary is QuadToneRIP (Mac and Windows) and PrintTool (Mac only) which you probably already have. If not,  you can find them at www.quadtonerip.com

You will also need a densitometer or spectrophotometer to read the results of the printer step-tablet, and a spreadsheet (Excel, Numbers, &c.) to evaluate the results and produce the Gray Curve.

I prefer Pictorico Ultra Premium OHP film, but others are successful with Pictorico Premium, Aristo, Fixxons, Inkpress, Agfa Copyjet, &c.

This article was updated on 27 Feb 2018.

Introduction

The use of well-made digital negatives for Platinum/Palladium printing has the substantial benefit of standardizing many of the variables involved in contrast control. In other words, digital negatives can and should be tailored to eliminate many of the variables of the process that are concerned with controlling contrast.

The goal of all approaches to making digital negatives is to create a negative in which the density range (DR) matches the exposure scale (ES) of the printing process. Methods that rely on the standard printer driver are limited to using the Media Type selection to control the amount of ink (ink limits) applied to the film. A Photoshop curve is required to do the heavy lifting of pulling and pushing the tonal values to where they need to be in order to produce a print with a linear tonal scale. QuadToneRIP (QTR) allows for much finer control of how the inks are used. A good QTR profile, before the addition of a process specific correction curve (Gray Curve), will produce a negative that prints fairly close to linear. The Gray Curve needed to linearize the profile will be very gentle, especially when compared to the very steep toe and shoulder section and flat mid-ranges of the curves required in other approaches.

Please note that this Curve uses Photo Black. There has been some confusion about this since Reeder and Hinkel used to recommend the use of Matte Black. As of 2016, Reeder has changed his recommendation to Photo Black. Most of us who have been making inkjet digital negatives for a while (I began in the early oughts), are aware that Matte Black has several drawbacks:

  • It is susceptible to rubbing off on every film I’ve tried.
  • It has way more UV blocking ability than we need which makes controlling it more difficult.
  • In every attempt I made to use it, it added noticeable graininess to the highlights.

Photo Black, on the other hand:

  • adheres well to all film substrates
  • has more than enough UV density for all processes (it can achieve a density of at least 3.2)
  • does not add graininess

I still run into someone now and then who insists that MK is better. I’ll just say one more time that it is not – use PK.

The example I am using in this article is for a Palladium (Pd) print on a typical paper, Revere Platinum, with only enough restrainer (“Na2” Platinum in this case) to prevent fog in the highlights in the event that the ferric oxalate sensitizer has gone slightly ferrous. (Some people prefer to use more restrainer, others prefer none at all.) I also add 1 drop of Tween 20 per 1 ml of solution to aid in even absorption. Exposure is for 150 seconds (2½ minutes) under a bank of Hitachi F20T9/BL fluorescent Blacklight (BL) tubes. The print is developed in Potassium Oxalate (KO).

JKS QTR 3880 21step TestTarget
21-step Test Image with Notes

Note: This target is a positive. I do the inversion to negative in PrintTool.

The QTR Curve

More properly called an Ink Descriptor File, the QTR Curve is a set of instructions that determines how the printer applies ink to the substrate when printing with the QTR driver. In the Mac version, QTR works as an alternative to the Epson printer driver. It appears in the printer list as a separate item with a name such as Quad3880, and is accessed through the Print menu from any application such as Photoshop or Print Tool. In the PC version, QTR is a stand-alone program. The PC version also includes a Curve Creator tool, which the Mac version lacks. In the Mac version, QTR Curves are created as text files using TextEdit or another text editor app.

The procedure for designing a curve has 3 major stages. For each stage you will print a 21-step target negative utilizing the relevant curve for that stage. You will then print it in the darkroom using the process that you are designing the curve for. All process variables must be pre-determined and adhered to throughout, since any change will require starting over. (I recommend using an analog step-tablet such as the Stouffer TP-4×5 21-step for preliminary testing since it will teach you a lot about your process and how variables such as paper, developer, metal mix, &c. interact with each other.)

  1. The Base Curve or Ink Descriptor File
    • This is your starting point
  2. The Gray Curve
    • This does what I call the “heavy lifting” part of getting the tones close to linear
    • Richard Boutwell’s BWMASTERY-21-step-QTR-Correction-Curve-Tool-for Digital Negatives is the best tool for this step that I have found
  3. Linearization
    • This step is for “fine tuning” the results of step 2
    • It utilizes the QTR-Linearize-Data applet that comes with QuadToneRIP

This is a screenshot of the text file of the finished curve we will make here:QTR Curve 3

The “#” at the beginning of a line indicates that this line is not part of the code. It is a note where I have specified the purpose of this file. This corresponds to the Notes tab in the PC Curve Creator tool. Each of the other sections also corresponds to tabbed sections of the PC Curve Creator tool.

  • Ink limits are determined in part by testing with the calibration target to find the maximum load that the substrate can hold, and in part by trial and error to find what is needed for a particular printing process. The Default Ink Limit can be anything or nothing. Always set a limit for each ink.
  • Number of Gray Parts and the other variables associated with it is an area with many possibilities for experimentation.
  • After using default settings for many years, I recently did some trial-and-error experiments with the Gray Highlight, Gray Shadow, and Gray Overlap settings and ended up with something very different than I had used previously.
  • Gray Gamma can also be useful in some situations. It is similar to moving the middle slider in a Photoshop Levels adjustment.
  • The GRAY_CURVE field can contain either
    • a set of input output numbers as in the example above
    • a path to a Photoshop curve saved on your computer (though this is limited to 15 points)
    • Note that you do not apply the correction curve to the image in Photoshop. Rather, it is built in to the QTR Curve.
  • The beauty of this method of making digital negatives over any of the methods using the Epson driver, is that if you have done a good job with the ink limits and the gray variables the actual curve will be very gentle and smooth unlike the curves needed to linearize other methods.
  • In Stage 3, we will be adding a LINEARIZE line at the very end of the file.

Please see the last section Printing The Negative for specific instructions on setting up PrintTool and QTR for printing negatives.

Stage 1 – The Base Curve

Feel free to use this as a starting point if you like. Other processes may need adjustments to the ink limits. I recommend leaving the all the color inks as they are and raising or lowering only the K, LK, and LLK limits as necessary.

QTR Curve 1
QTR Base Curve
  1. Make a 21-step digital test negative using your QTR Curve with GRAY_CURVE=”0;0 100;100″ as a placeholder. This ensures a straight line GRAY_CURVE.
  2. Print that negative using your preferred coating mix for the process you are profiling at your pre-determined exposure time (minimum time to achieve maximum black) and develop using your standard procedure.
  3. Measure the 21-step results. If you want to see a graphic representation of the results, plot them in a spreadsheet and create a graph like this:

QTR Graph 1

Note: The raw data is not as smooth as what I have shown here. I am using the QTR Data-Tool to smooth the measurements then plotting the L* values from that. 

Stage 2 – The Gray Curve

Once you are certain that your exposure time is correct (the minimum exposure required to achieve maximum black), and that you have a good base curve (all steps of your target are distinguishable even though they are not yet evenly spaced), it’s time to begin the process of adjusting ink densities in a way that results in a linear distribution of tones in the print. For the first part of that we are going to use QTR’s Gray Curve function. Basically what we are doing is seeing where there is too much ink and where there is too little, and giving QTR the instructions it needs to correct that situation.

Note: It is possible to eliminate the Gray Curve and just use Linearize, however, Linearize can’t be used iteratively.

I’ve used a variety of spreadsheet based methods of creating the correction curve, starting with a simple mirror image, and most recently with this method which uses some quite sophisticated math to convert density readings to LAB. The latter, developed by Richard Boutwell and available for free download from his website, is what I am using here. (Richard also has other interesting tools for making digital negatives, but you must buy them.)

A note about Density versus LAB: Density is a logarithmic scale and as such a curve made using Density readings without converting to LAB will be mathematically linear but not visually linear. Therefore the resulting print will appear dark. The LAB system was designed to compensate for this and to produce a visually linear print. I only learned about LAB about 2 years ago. At that point I finally understood why I had been having to manually and through much trial-and-error make a curve that bent towards the lighter side of the graph to get a result that looked right to my eyes.

Begin by entering the 21-step Density readings from the print of your base negative. The spreadsheet converts the Density Readings to LAB (specifically the L* component which stands for Luminosity or Lightness) and generates a set of 21 number pairs in QTR Gray Curve format that you will copy and paste into the Gray Curve field in your Ink Descriptor file. Note that the function that compiles the string treats the numbers as text which results in many more decimal places than necessary. Don’t worry about this.

  1. When the print from Stage 1 is dry, use a densitometer or spectrophotometer to read the 21-steps. Enter those values into the spreadsheet.
    • Until recently, I used an X-rite 400 reflection densitometer to make Density readings.
    • Now I use DataColor Spyder3 Print spectrophotometer which can read LAB values as well as Density, and has the benefit of saving the readings as text files.
    • I’ve recently picked up a used X-rite i1 Pro which is said to be a superior instrument compared to the SpyderPrint, but I’m finding the software to be less user friendly, at least for this purpose.
    • Some people do this step using a scanner. I feel this introduces an opportunity for error since you are reading a scan of the test print rather than the print itself.
  2. Copy the Gray Curve string from the Correction Curve Tool and paste it into the Gray Curve field in your text file.
  3. Save the text file with a new name and run the install script.

    QTR Curve 2
    QTR Curve with Gray Curve

Make a negative using this curve, print it, and measure the results same as in the previous stage.

If you are using a spreadsheet graph to track your progress, enter the results of this step to it to see how close you got.

QTR Graph 2

Stage 3 – Linearize

In the final step, we take the text file from of the LAB or Density readings from the print made in the previous step and run it through the QTR-Linearize-Data applet. If you use Density readings here QTR_Linearize-Data will automatically convert them to LAB.

  1. When the print from Stage 2 is dry, use a densitometer or spectrophotometer to read the 21-steps.
  2. Save this file where you can easily find it.
  3. Then follow one of the procedures below.
Method 1:
  1. Navigate in Finder to the data file you want to Linearize.
  2. Right-click (or Control-click) on the file.
  3. Select Open With > Other… > Applications > QuadToneRIP > QTR-Linearize-Data.app
  4. It will create a new text file using the same filename with “-out.txt” at the end.
  5. At the bottom of that file you will find two sets of Linearize numbers – use the one labeled “For Digital Negatives”.
  6. Copy the entire string including the word LINEARIZE.
  7. Go to your previous QTR text file (the one you just used to print the negative) and paste the LINEARIZE line at the end.
  8. Save with a new name – I usually just add “-lin” to the end of the name.
Method 2:
  1. Open 2 finder windows and resize them so both are fully visible side-by-side.
  2. In one window open the Applications > QuadToneRIP folder so that QTR-Linearize-Data.app is visible; in the other navigate to the folder with your data file.
  3. Drag & drop the data file onto the app
  4. It will create a new text file using the same filename with “-out.txt” at the end.
  5. At the bottom of that file you will find two sets of Linearize numbers – use the one labeled “For Digital Negatives”.
  6. Copy the entire string including the word LINEARIZE.
  7. Go to your previous QTR text file (the one you just used to print the negative) and paste the LINEARIZE line at the end.
  8. Save with a new name – I usually just add “-lin” to the end of the name.

The -out.txt file will look something like this:QTR Linearize DataCopy the “For Digital Negative” LINEARIZE= string and paste it at the end of the curve as shown below. Then go to File > Save As … and save the curve with a new name. I simply add -lin to the end of the previous iteration.

QTR Curve 3
QTR Curve with Gray Curve and Linearize

Print your 21-step target file one more time using this curve, measure the results, and add to the graph to verify that the linearization process has achieved the desired result.

QTR Graph 3

Don’t try too hard to make it perfect. As long as it’s reasonably close to straight you are in good shape. There will be some degree of “wobble” in all hand-coated processes.

File Preparation

Prepare the image file in the same way you would if your goal was to produce a high-quality black and white digital print. I work with scanned black and white film in 16-bit grayscale mode. If your originals are color film or files, they can be prepared in any way you like but will need to be converted to grayscale for printing. I use Gray Gamma 2.2 for the document profile. Whatever profile you choose, be consistent.

It’s not a bad idea to proof your file on paper with QTR before printing your negative. When you are satisfied with the digital print, you are ready to make the negative.

Printing The Negative

Recent versions of Photoshop, CS6 and CC, lack the ability to print with No Color Management which was an option at least through CS3. This unfortunately introduces the possibility of unexpected interference with the operation of QTR from the operating system or from Photoshop. It is possible to print negatives using QTR from within Photoshop but not without using some form of color management which is exactly what we want to avoid. Print Tool (by Roy Harrington, the author of QTR) is a simple stand-alone printing utility that restores the ability to print with No Color Management.

PrintToolSettings-1

  1. Select correct Quad Printer
  2. Page Setup (Select page size and orientation)
  3. Open image file in Print Tool (The image should be a positive – inverting and flipping is done here in step 6)
  4. Set desired Position and Scaling (usually 100% and centered)
  5. Set Printer Color Management to No Color Management
  6. Check Negative — Invert & Flip
    • Click Prefs… to set your border options and Flip to print emulsion down PrintToolSettings-2
  7. Click Run Print…
  8. Open QuadToneRIP menu, make the following selections, then save as preset
    1. Mode:  16-bit or 8-bit
    2. Curve 1:  [name of curve]
    3. Paper Feed:  Sheet Feed (this is up to you)
    4. Resolution:  2880 dpi
    5. Speed:  Uni-directional
    6. Black Ink:  PhotoPrintToolSettings-3
  9. Insert transparency film in printer
  10. Click Print
  11. Let the negative cure overnight or use a gentle flow of warm air to speed up the drying/curing process if you must. This is important!

 

10 thoughts on “Digital Negatives with QuadToneRIP”

  1. Hi Keith
    I would be very grateful for your help again to finalise my gray curve to add to the profile you kindly sent me.
    I have taken LAB readings with my Spyder3Print of a Palladium printed 21 step tablet using your profile. They range from Step 0%=94.7, 40%=60.33, 70%=42.08 and 100%=24.15. I put these into an Excel spreadsheet and get a line from top left down to bottom right (roughly).

    What I don’t understand is how I create a gray curve from these figures. When I used Ron’s method and scanned the print and then read the steps in Photoshop, I got values that were close to those of the steps i.e maybe 28 for step 25%. I was then able to do the input/output table and put the figures into the curve.

    In the case of the 40% reading, do I take the 60.33 from 100 (39.67) and that is the number I put into the gray curve? Also I presume you make step 0%=0 and 100%=100 regardless of the readings? Not sure how the spreadsheet helps.

    Obviously I am missing something here so please could you enlighten me.

    Kind regards

    Charlie

  2. Yikes! I see why you’re confused. I seem to have left out a rather important part – the spreadsheet itself – in the Making the Gray Curve section (steps 3 and 4) above. The purpose of the spreadsheet is to analyze the print values and use them to produce a correction curve. Long ago I did this by hand (without a spreadsheet) and lots of trial and error. I mean lots! I’ve made my own simple spreadsheet and tried a few other similar ones made by others and shared, but by far the best is one I came upon a couple years ago by Richard Boutwell. You can find it here: https://www.dropbox.com/s/qvz5rbc4zi868yi/BWMASTERY-21-step-QTR-Correction-Curve-Tool-for%20Digital%20Negatives.xlsx?dl=0
    It takes Density values for input, ordered from low to high, so you’ll need to reread your printed target with the SpyderPrint in Density mode. Then you will copy the resulting Gray Curve and paste it into the Gray_Curve field in your QTR text or qidf file and save with a new name. Then continue with step 5.
    Since writing this article several years ago, I’ve also found a better way to fine tune the QTR curve after creating and adding the Gray Curve if it needs further adjustment. I really need to update this article.
    It might be better to do this via email for now, since I can then send screenshots if necessary. If you want to do that my email is keith@jkschreiber.com.

  3. Hi Keith,

    I am doing digital negatives on a p800, and am not very happy with the results with using the default epson drivers. As a result I am interested in trying to use QTR, and came across your helpful writeup above. I understand the process you describe there in general, and intend to try to use it with the base file for the p800 file you posted on the QTR yahoo group a while back, Now, the part that is unclear to me is how to tune the base file (before worrying on the gray curve) for a given process to get a step tablet result where you are in the ballpark (the step tablet varies from pure black to pure white, and most steps have a unique tone). This is where the epson normal driver screws me up—even when varying color density up and down I end up with a step tablet where a little bit less than half the tablet is black. Assuming that the base file when printed through QTR does not give me in the ball park, my understanding is that you tweak K, LK and LLK limits up and down. Could you elaborate on how do this?

    Regards,
    -Per

    1. Hi Per,
      I haven’t used the Epson driver for digital negatives for at least 15 years, so I can’t help with that. But even though I don’t have a P800 myself, I have made QTR curves for a friend or 2 for that printer. The base curve for Pt/Pd is very similar to my 3880. (Note: To avoid confusion with ICC profiles, I am using the term Curve for the QTR Ink Descriptor File, even though profile is more apt in my opinion. It is also called a Curve in the QTR dialog.)

      In your email, you mentioned starting with making negs for standard silver paper. This will require a K limit quite a bit lower that the one I use for Pt/Pd since the Exposure Scale (ES) of silver paper (assuming Grade 2 paper or filter) is quite a bit shorter at ~1.10 compared to ~1.90 for the straight palladium that my example is made for.

      To find the K limit for for any process, first figure out all of the printing details and conditions that you will be using. You really need to do this first because any changes may effect the ES of the print. Examples for silver are things like contrast grade or filter, and developer. Toning may be an important consideration too especially if it’s a method that changes print density and or contrast.
      Print the QTR 8-channel ink separation target (QuadToneRIP>CurveDesign>Images inkseparation8.tif in the Mac version, not sure in Windows) using QTR Calibration Mode which is found in the QuadToneRIP>Mode dropdown list in the QTR setup dialog. Leave Calibration Ink Limit at 100; set Resolution to the highest setting available; Speed to uni-directional; and be sure Photo Black is selected. I strongly recommend using Pictorico Ultra Premium transparency film. Don’t worry if there is a little bleeding or pooling of ink. Allow the negative to dry for at least 24 hours. This is important since density decreases a little as the ink cures. You can speed it up a little by using hot air from a hair drier, but I’d still let it cure for at least 12 hours.

      Print this negative using the exact same materials and methods you will be using for real prints. When it is fully processed and thoroughly dry, find the first step in the Black strip that is pure paper white. This is your K limit. I expect it will be around 15% for grade 2 silver paper. I usually set K Boost about 20% higher than the K Limit.

      The limits for the LK and LLK inks are a bit more of an intuitive thing. As long as there is no flooding/pooling/smearing on the Calibration image, they can be as high as 100%, at least in theory. In my QTR curve for Pd on Platine, Revere, HPR, etc., the limits are K=50 (Boost=60), LK=60, LLK=80, and Y=10. In my QTR curve for New Cyanotype (Mike Ware’s formula), which has an ES of about 1.40 the way I measure it, which is about 1 stop more than grade 2 silver paper, I use limits of K=20 (Boost=25), LK=40, LLK=60, and Y=10. Yours should be no higher than those. I don’t have a formula for this. My LK and LLK limits have been determined by trial and error.

      I’ve never made digital neg curves for silver paper. In fact I haven’t used silver paper in nearly 20 years, so I’m extrapolating from experience with other processes here. I hope it works for you.

      Cheers,
      Keith

  4. Thanks for your guidance, Keith. I will give this a try and see how it goes. I am very excited to see if I do better this way compared to with the Epson driver. Question though: you say “Lk and llk limits can be 100%” but then “lk=40 and llk=60, don’t go higher”. What am I missing?

  5. Thanks for all your help through the process.I am chugging through the procedure, and I can already tell that QTR is giving me better results than trying to use the epson driver. I am now at the stage where I am trying to linearize the curve. I am using manual densiometer readings to read the 21 step wedge. What is the format of the txt file that I am supposed to feed to the linearize program? Can I use density values or do I need to figure out a way to convert the density values to some other format?

    1. Take those manual density readings and put them in a text file, one measurement per line, light to dark. Be sure there are no reversals. Save it. (You might even run it through the QTR Data-Tool to smooth the measurements. This is optional and can be done on the readings for each stage if desired.) Then use the QTR-Linearize-Data.app to create the LINEARIZE string. The app converts to Density to L* on the fly. Copy/paste the digital negative version from the resulting -out.txt file at the end of your ink descriptor file as shown above. Save with new name. Don’t forget to run the install command.

  6. Hi Keith,

    Just to let you know: With some false starts and stops, I finally got a workable profile going and it is looking really good even without pizeography inks (I wish I had a printer where that was an option, but it is what it is). Thanks for all your help, I could not have done it without your information. The time you spent writing the post really did help someone else do something cool.

    Kind regards,
    -Per

  7. Thank you Keith for writing such a clear page.
    I’ll follow your directions to print negatives for silver paper.

    Per, have you succeeded in building a good file for silver printing? Could you share the values you have found for the different ink limits (even the color ones)?

    Regards,
    Al

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s