QTR-Quadtone RIP

Hi -- Can anyone explain the basic process of linearization using a .txt file on a Mac? Thanks! Scott

The process is hard to explain without a little more info on what you mean by linearizing with a text file. 

In a sense, you don't linearize anything with the text file. You simply enter the measurements from the printed target into a text file. The QTR curve creation program does the linearizing

Measurement files are basically tab delimited text files. The can contain either Density measurements on separate lines, or L*a*b* measurements on separate lines. You can make them by hand from entering measurements from a densitometer or generate them with an spectrometer and measurement software. 

When you "drop" a measurement file onto the linearize-data app you will get a text based graph of the measurements and they will be concatenated into a string that begins LINEARIZE="measurement measurement measurement etc..."

The linearization is done when you install the profile and the curve creation program does its thing. After all the partitioning and temp curve creation the curve creation program takes those measurements in the LINEARIZE= line and makes a straight line between the first value and the last value. It then and generates a reverse correction curve for each of of the in-between values so they will produce the correct density on that straight line when printed. 

There is another option of linearing the quad file with the linearize-quad app but that doesn't seem to be what you're asking about.

Hope that helps,
Richard Boutwell

www.RichardBoutwell.com
www.BWMastery.com

Great, Thanks! 


Sent from Yahoo Mail. Get the app


    On Tuesday, September 13, 2016 3:57 PM, "sw3457@... [QuadtoneRIP]" <QuadtoneRIP@yahoogroups.com> wrote:



     Thanks Richard!  What I'm really trying to find out is the format for entering densitometer readings into the text file.  In the "LINEARIZE=measurement measurement measurement etc..." line, what are the specific "measurements"?  If I read my test patches with a reflective densitometer, what is the syntax for the correction?  Let's say my 50% patch has the density I want in my 60% patch, do I write LINEARIZE=50,60?  Just not sure of the proper grammar here.  Thanks again for taking the time to answer my question.  Scott   #yiv5453920372 #yiv5453920372 -- #yiv5453920372ygrp-mkp {border:1px solid #d8d8d8;font-family:Arial;margin:10px 0;padding:0 10px;}#yiv5453920372 #yiv5453920372ygrp-mkp hr {border:1px solid #d8d8d8;}#yiv5453920372 #yiv5453920372ygrp-mkp #yiv5453920372hd {color:#628c2a;font-size:85%;font-weight:700;line-height:122%;margin:10px 0;}#yiv5453920372 #yiv5453920372ygrp-mkp #yiv5453920372ads {margin-bottom:10px;}#yiv5453920372 #yiv5453920372ygrp-mkp .yiv5453920372ad {padding:0 0;}#yiv5453920372 #yiv5453920372ygrp-mkp .yiv5453920372ad p {margin:0;}#yiv5453920372 #yiv5453920372ygrp-mkp .yiv5453920372ad a {color:#0000ff;text-decoration:none;}#yiv5453920372 #yiv5453920372ygrp-sponsor #yiv5453920372ygrp-lc {font-family:Arial;}#yiv5453920372 #yiv5453920372ygrp-sponsor #yiv5453920372ygrp-lc #yiv5453920372hd {margin:10px 0px;font-weight:700;font-size:78%;line-height:122%;}#yiv5453920372 #yiv5453920372ygrp-sponsor #yiv5453920372ygrp-lc .yiv5453920372ad {margin-bottom:10px;padding:0 0;}#yiv5453920372 #yiv5453920372actions {font-family:Verdana;font-size:11px;padding:10px 0;}#yiv5453920372 #yiv5453920372activity {background-color:#e0ecee;float:left;font-family:Verdana;font-size:10px;padding:10px;}#yiv5453920372 #yiv5453920372activity span {font-weight:700;}#yiv5453920372 #yiv5453920372activity span:first-child {text-transform:uppercase;}#yiv5453920372 #yiv5453920372activity span a {color:#5085b6;text-decoration:none;}#yiv5453920372 #yiv5453920372activity span span {color:#ff7900;}#yiv5453920372 #yiv5453920372activity span .yiv5453920372underline {text-decoration:underline;}#yiv5453920372 .yiv5453920372attach {clear:both;display:table;font-family:Arial;font-size:12px;padding:10px 0;width:400px;}#yiv5453920372 .yiv5453920372attach div a {text-decoration:none;}#yiv5453920372 .yiv5453920372attach img {border:none;padding-right:5px;}#yiv5453920372 .yiv5453920372attach label {display:block;margin-bottom:5px;}#yiv5453920372 .yiv5453920372attach label a {text-decoration:none;}#yiv5453920372 blockquote {margin:0 0 0 4px;}#yiv5453920372 .yiv5453920372bold {font-family:Arial;font-size:13px;font-weight:700;}#yiv5453920372 .yiv5453920372bold a {text-decoration:none;}#yiv5453920372 dd.yiv5453920372last p a {font-family:Verdana;font-weight:700;}#yiv5453920372 dd.yiv5453920372last p span {margin-right:10px;font-family:Verdana;font-weight:700;}#yiv5453920372 dd.yiv5453920372last p span.yiv5453920372yshortcuts {margin-right:0;}#yiv5453920372 div.yiv5453920372attach-table div div a {text-decoration:none;}#yiv5453920372 div.yiv5453920372attach-table {width:400px;}#yiv5453920372 div.yiv5453920372file-title a, #yiv5453920372 div.yiv5453920372file-title a:active, #yiv5453920372 div.yiv5453920372file-title a:hover, #yiv5453920372 div.yiv5453920372file-title a:visited {text-decoration:none;}#yiv5453920372 div.yiv5453920372photo-title a, #yiv5453920372 div.yiv5453920372photo-title a:active, #yiv5453920372 div.yiv5453920372photo-title a:hover, #yiv5453920372 div.yiv5453920372photo-title a:visited {text-decoration:none;}#yiv5453920372 div#yiv5453920372ygrp-mlmsg #yiv5453920372ygrp-msg p a span.yiv5453920372yshortcuts {font-family:Verdana;font-size:10px;font-weight:normal;}#yiv5453920372 .yiv5453920372green {color:#628c2a;}#yiv5453920372 .yiv5453920372MsoNormal {margin:0 0 0 0;}#yiv5453920372 o {font-size:0;}#yiv5453920372 #yiv5453920372photos div {float:left;width:72px;}#yiv5453920372 #yiv5453920372photos div div {border:1px solid #666666;min-height:62px;overflow:hidden;width:62px;}#yiv5453920372 #yiv5453920372photos div label {color:#666666;font-size:10px;overflow:hidden;text-align:center;white-space:nowrap;width:64px;}#yiv5453920372 #yiv5453920372reco-category {font-size:77%;}#yiv5453920372 #yiv5453920372reco-desc {font-size:77%;}#yiv5453920372 .yiv5453920372replbq {margin:4px;}#yiv5453920372 #yiv5453920372ygrp-actbar div a:first-child {margin-right:2px;padding-right:5px;}#yiv5453920372 #yiv5453920372ygrp-mlmsg {font-size:13px;font-family:Arial, helvetica, clean, sans-serif;}#yiv5453920372 #yiv5453920372ygrp-mlmsg table {font-size:inherit;font:100%;}#yiv5453920372 #yiv5453920372ygrp-mlmsg select, #yiv5453920372 input, #yiv5453920372 textarea {font:99% Arial, Helvetica, clean, sans-serif;}#yiv5453920372 #yiv5453920372ygrp-mlmsg pre, #yiv5453920372 code {font:115% monospace;}#yiv5453920372 #yiv5453920372ygrp-mlmsg * {line-height:1.22em;}#yiv5453920372 #yiv5453920372ygrp-mlmsg #yiv5453920372logo {padding-bottom:10px;}#yiv5453920372 #yiv5453920372ygrp-msg p a {font-family:Verdana;}#yiv5453920372 #yiv5453920372ygrp-msg p#yiv5453920372attach-count span {color:#1E66AE;font-weight:700;}#yiv5453920372 #yiv5453920372ygrp-reco #yiv5453920372reco-head {color:#ff7900;font-weight:700;}#yiv5453920372 #yiv5453920372ygrp-reco {margin-bottom:20px;padding:0px;}#yiv5453920372 #yiv5453920372ygrp-sponsor #yiv5453920372ov li a {font-size:130%;text-decoration:none;}#yiv5453920372 #yiv5453920372ygrp-sponsor #yiv5453920372ov li {font-size:77%;list-style-type:square;padding:6px 0;}#yiv5453920372 #yiv5453920372ygrp-sponsor #yiv5453920372ov ul {margin:0;padding:0 0 0 8px;}#yiv5453920372 #yiv5453920372ygrp-text {font-family:Georgia;}#yiv5453920372 #yiv5453920372ygrp-text p {margin:0 0 1em 0;}#yiv5453920372 #yiv5453920372ygrp-text tt {font-size:120%;}#yiv5453920372 #yiv5453920372ygrp-vital ul li:last-child {border-right:none !important;}#yiv5453920372

##Linearizing

Sandy basically covered it. The only point I would like to clarify is that Density from a densitometer can be used to arrive the Lab values QTR needs in the LINEARIZE= line. The Linearize-Data app will actually do a Density to XYZ\_Y to LAB\_L to produce the Linearize="" string. I have never tried just entering Density values into the Linearize= line. Roy might need to correct me, but I THINK there is something in the curve creation script that will check to see if the values there are L\* or Density, and if it is Density, it will convert it to L\*


If you want to make your own Lab to Density and back conversion there is a formula to do that can be found a few different places online. This is one of my favorite posts someone had written on the subject: http://stonerosephotos.com/blog/2012/01/ink-density-reflectance-and-l/

I made this tool that does these conversions a while back: https://www.dropbox.com/s/nrcxze38fmbruiu/BWMASTERY%20Lab%20to%20Density%20Conversion%20Tool.xlsx?dl=0

You can't use Density without translating to L\* for the QTR Linearization because Density is a logarithmic scale, and linearizing to create evenly spaced Density patches will end up with a print that appears too dark compared to what QTR does with linear Lab L\*. It actually can be done if you are using the gray curve line but will mean that you can't use Linearize line (it will recorrect the linear Density to linear L\*

I actually made a version of my gray curve correction tool for digital negatives for someone last year who wanted to create linear Density rather than linear L\*. He was attempting to compensate for the lower dmax and overal lower density of the PTPD print compared to what a normal inkjet print was producing and what he was seing on the screen. It worked to some degree, but was't compatible with all the other stuff out there so I never posted more about it. It was just another option/take on the linear/compressed shadow issue when dealing with processes with a lower Dmax.

This has gotten pretty far off the original subject but I did say it was hard to talk in gerneal terms about Linearization with a text file...

Richard Boutwell

Thank you 

Sent from Yahoo Mail. Get the app 

    On Wednesday, September 14, 2016 11:54 PM, "richard@... [QuadtoneRIP]" <QuadtoneRIP@yahoogroups.com> wrote:
 

     Scott, 
Take a look at this post I made last about how the QTR Linearize function and Gray Curve inputs can differ:In Search of the Perfect QTR Profile Part 3: QTR Correction Curve Tool 
|    |
|    |        |    |       In Search of the Perfect QTR Profile Part 3: QTR Correct...  I don't mean for this to become a site dedicated to printing with QuadToneRIP, and I have a few new exciting things waiting in the wings that will be comin...    |    |
|      View on www.bwmastery.com      |    Preview by Yahoo    |
|    |

   
The correction curve I make with the spreadsheet tool is basically a series of lookups and linear interpolations, which I assume is about all the QTR Linearization is doing. The difference is how the QTR linearization generates the splines for the intermediate quad values between each of the control points. The Gray Curve seems to be doing something in a different order and can be smoother in some instances. I've included examples of cases where it will create a correction curve for something that fails to linearize with the QTR curve creation linearization (using the same measurement data file). Take a look at the screenshots in the results at the end of the post to see what i mean. 
This isn't to say that one is better than the other, but that this can be a useful tool that can get you closer than more generalized gray curve settings. 
Hope that helps, Richard Boutwell
http://www.richardboutwell.com/http://www.bwmastery.com/

Hello 


Sent from Yahoo Mail. Get the app


    On Thursday, September 15, 2016 12:18 AM, "richard@... [QuadtoneRIP]" <QuadtoneRIP@yahoogroups.com> wrote:



     Scott, I forgot to mention that the input;output values in the gray_curve line are not measurements from your densitometer. Those are the ink%s that are being corrected. You only use the densitometer readings to see what the native response from the profile is and the measurements lookup the reverse correction curve.
The whole "I need an ink level of 60% to produce the correct tone at the 50% patch is a very general way of looking at it". The spreadsheet I linked to does all that for you much more precisely by creating the linear target values and then doing an interpolation for each of the points between the Dmin and Dmax. This is the same kind of interpolation mentioned in one of the calibration PDFs for doing a proper cross over point calculation. But instead of needing to do 60+ different measurements and 19 interpolations you just need to enter the density readings—either manually or by copy and pasting—and the template does the rest for you. 
Hope that helps, Richard Boutwell
http://www.richardboutwell.com/http://www.bwmastery.com/

Hi folks,

I just tried using QTR-Linearize-Data for the first time and found that for linearizing digital negatives it works, but not quite as Roy described it here, unless I misunderstood something. I read the part quoted below to mean the input data in the text file for digital negs should be from black to white. 

QTR-Linearize-Data will not take numbers from black to white; whether Den or LAB, it gives the error message "The Lab values are not in order. Cannot be linearized." 

However, when a text file with numbers from white to black is dropped onto QTR-Linearize-Data, it produces both a standard white to black LINEARIZE= string, AND an inverted one, black to white, for digital negatives.

LINEARIZE="88.18 83.49 75.44 69.33 66.12 63.64 61.83 60.05 57.75 55.53 51.78 48.73 45.82 42.59 38.68 34.25 30.54 27.43 24.22 22.41 20.68"

For Digital Negative:
LINEARIZE="20.68 22.41 24.22 27.43 30.54 34.25 38.68 42.59 45.82 48.73 51.78 55.53 57.75 60.05 61.83 63.64 66.12 69.33 75.44 83.49 88.18"


This may not be news to anyone but me, but I thought I should mention it just in case. I've never used QTR-Linearize-Data before, though I use QTR-Linearize-Quad frequently with PiezoDN.

One question though: Can both GRAY_CURVE and LINEARIZE be used in the same QTR Curve? I know at least one person who does that with digital negatives, and successfully I think, but I've never seen it discussed or written about anywhere.

Cheers,
Keith

Keith Schreiber
jkschreiber.com

> On Sep 13, 2016, at 11:02 PM, Roy Harrington roy@... [QuadtoneRIP] <QuadtoneRIP@yahoogroups.com> wrote:
> 
> For ordinary printing the
> values must be from White to Black, but for digital negatives use Black to White.
> LINEARIZE will create a correction curve such that the new curve set will produce
> a step with linear Lab values -- i.e. it linearizes for Lab values.

I began using the LINEARIZE command three or four years ago for my QTR profiles for digital negatives and did not have any issues with the basic work flow, which is as follows.

1. Open a 21 or 51 step target, convert it to a negative, and print on OHP. Use the same print settings you plan to use in printing your negatives, including of course No Color Management.

2. Using the step wedge negative make a positive print in your process, using you standard process controls.

3. Scan the positive step wedge print with a spectrophotometer or colorimeter, beginning at the lightest step and ending with the darkest one. Save the .txt file.

4. Drop the .txt file on the QTR-Linearize-Data application, and the Linearize numbers will be created. Two sets of numbers are generated, one for direct printing of positives, a second reversed set for printing digital negatives.

5. Insert the Linearize sequence in your QTR profile. For example.
LINEARIZE="19.5 20.76 22.12 23.32 23.98 26.61 27.8 28.77 30.31 31.95 33.04 34.01 36.41 37.57 39.03 41.4 42.63 43.75 46.79 48.73 50.27 51.43 52.4 53.36 54.67 55.67 56.41 58.57 60.34 61.61 63.59 65.74 67.37 70.14 71.12 72.49 73.85 75.2 76.76 79.49 80.81 82.85 85.17 86.87 89.44 91.06 93.03 93.63 94.01 94.66 95.3"

(Alternatively, if you use a densitometer you can just read the log densities of the step wedge print, from lightest to darkest, and insert them directly into the QTR Profile.)

6. Run the Install Command for your Quad printer, then check the Terminal script to make sure that the profile was installed. If the values are not increasing as the program expects you will get a command "Invalid Linearize Command -- Not Constantly Increasing." This can be confusing because sometimes the numbers are constantly increasing, but apparently not enough to satisfy the program. To verify that the profile was installed go (for MAC) Library>Printers>QTR>Quadtone>QuadYourPrinter. Click on the specific .quad file to see it in graph form.

Finally, both LINEARIZE and GRAY_CURVE= can be used in the same profile. This can be very useful because you can have one profile that is linearize for a process with LINEARIZE, then linearized for specific variations of the process with GRAY_CURVE=.

If you have been using GRAY_CURVE in the past for digital negatives for alternative printing and attempt to switch to LINEARIZE the biggest issue you will have is not the digital work flow, but process control. You can make GRAY_CURVE= work by fudging the curve, which can be way off and still work. But for LINEARIZE to work with Lab values you must have 21 discrete values ranging from light to dark, and the values must be constantly increasing enough to satisfy the requirements of the program.

Sandy

Roy,

Related to Keith’s mentioned use of QTR-Linearize-Quad and Dig-negs:  PiezoDN utilizes QTR-Linearize-Quad. The .quads themselves are built (differently than normal Piezo print quads) to enable calibration of the printed negative from the darkroom print using QTR-Linearize-Quad and 129 error-corrected measurements from light to dark.

We utilize a similar thing (with a different error correction formula) in Piezography Professional Edition. Basically we’ve applied our algorithm for fixing measurement errors (either in brush-strokes from hand-coated prints or flicker in spectros) prior to QTR-Linearize-Quad in order to properly calibrate Piezography with minimal errors and high bit-depth and smoothness. This error correction algo. for PPE is the algorithm that we’ve used for a decade + to profile Piezo .quads (at 256 steps). I recently updated it a bit for our building of Pro curves as well . . .

best,
Walker

> On Dec 9, 2016, at 3:55 PM, Roy Harrington roy@... [QuadtoneRIP] <QuadtoneRIP@yahoogroups.com> wrote:
> 
> 
> Hi Keith
> 
> I'm a little confused about your comments regarding QTR-Linearize-Data vs QTR-Linearize-Quad.
> The Linearize-Data script has been around a long time and its mainly to analyze measurement data files to extract
> the L-values needed for the LINEARIZE capability in Profile Creation program.  
> I added dealing with Digital Negatives a while ago by adding the reverse order output from Linearize--Data
> and changing the LINEARIZE function to accept reverse order data.  (this is the only way because QTRgui is unmodifable)
> LINEARIZE is different for negs because the correction is opposite.
> 
> Later than the above I added a Linearize-Quad script that analyzes measurement data like above but just immediately
> applies this to an existing .quad file.  (mostly this is useful for Piezo .quad's that don't go thru Profile Create)
> The Linearize-Quad was not part of the digital negative linearizing since that's in Profile Create.
> So I don't get the "I use QTR-Linearize-Quad frequently with PiezoDN" comment.
> 
> --
> Lastly -- yes you can use both GRAY_CURVE and LINEARIZE be used in the same QTR Curve.
> 
> Roy
> 
> On Thu, Dec 8, 2016 at 10:30 PM, Keith Schreiber keith@... <mailto:keith@...> [QuadtoneRIP] <QuadtoneRIP@yahoogroups.com <mailto:QuadtoneRIP@yahoogroups.com>> wrote:
> 
> 
> Hi folks,
> 
> I just tried using QTR-Linearize-Data for the first time and found that for linearizing digital negatives it works, but not quite as Roy described it here, unless I misunderstood something. I read the part quoted below to mean the input data in the text file for digital negs should be from black to white. 
> 
> QTR-Linearize-Data will not take numbers from black to white; whether Den or LAB, it gives the error message "The Lab values are not in order. Cannot be linearized." 
> 
> However, when a text file with numbers from white to black is dropped onto QTR-Linearize-Data, it produces both a standard white to black LINEARIZE= string, AND an inverted one, black to white, for digital negatives.
> 
> LINEARIZE="88.18 83.49 75.44 69.33 66.12 63.64 61.83 60.05 57.75 55.53 51.78 48.73 45.82 42.59 38.68 34.25 30.54 27.43 24.22 22.41 20.68"
> 
> For Digital Negative:
> LINEARIZE="20.68 22.41 24.22 27.43 30.54 34.25 38.68 42.59 45.82 48.73 51.78 55.53 57.75 60.05 61.83 63.64 66.12 69.33 75.44 83.49 88.18"
> 
> 
> This may not be news to anyone but me, but I thought I should mention it just in case. I've never used QTR-Linearize-Data before, though I use QTR-Linearize-Quad frequently with PiezoDN.
> 
> One question though: Can both GRAY_CURVE and LINEARIZE be used in the same QTR Curve? I know at least one person who does that with digital negatives, and successfully I think, but I've never seen it discussed or written about anywhere.
> 
> Cheers,
> Keith
> 
> Keith Schreiber
> jkschreiber.com <http://jkschreiber.com/>
> 
> 
>> On Sep 13, 2016, at 11:02 PM, Roy Harrington roy@... <mailto:roy@...> [QuadtoneRIP] <QuadtoneRIP@yahoogroups.com <mailto:QuadtoneRIP@yahoogroups.com>> wrote:
>> 
>> For ordinary printing the
>> values must be from White to Black, but for digital negatives use Black to White.
>> LINEARIZE will create a correction curve such that the new curve set will produce
>> a step with linear Lab values -- i.e. it linearizes for Lab values.
> 
> 
> 
> 
> 
> 
> -- 
> Roy Harrington
> roy@... <mailto:roy@...>
> www.harrington.com <http://www.harrington.com/>
> 
>