Luminous Landscape Forum
Raw & Post Processing, Printing => Colour Management => Topic started by: Onslow on February 11, 2019, 11:47:41 pm
-
I have the i1Pro2 and i1Publisher software 1.8.3 as well as Colorthink Pro 3.0.4. I was asked the following question the other and was stumped as I have never done it. I don't see why you couldn't. Thoughts anyone?
"Can you measure a swatch and get statistics on grayscale bars (similar to one used from the darkroom printing days) and evaluate the values to a B/W print and trend if the inks (prints) are fading? I’m my case I am look to evaluate a single swatch (middle gray) over an extended period of time."
-
Chromix has software that can do this. Or at a more elementary level, you can set-up an Excel spreadsheet to keep track of the value of the same patches measured at intervals.
-
using the i1P 'Measure Chart' or 'Measure Reference Chart' allows you to define your own chart (for you, a 1 swatch chart)
and measure and save.
As Mark said, you probably will need to do a little manual work to save the L*a*b* #s in a spreadsheet and graph that over time.
***
A chart definition describes the quantity and layout of a series of color patches.
When using the i1Pro, the i1Pro2 or the i1iO, you can specify the number of patches, in rows and columns, that you intend to measure.
Use this option when the patch set for a printed chart is unknown, but you wish to record the color values of the printed patches for the purposes of comparison and analysis.
***
This appnote is mostly applicable (although you're obviously not doing this to customize a CC24 for scanning reference)
https://www.xrite.com/-/media/xrite/files/apps_engineering_techdocuments/c/custom_reference_data_en.pdf
Also, I've seen 21 and 51 step BW charts on both Keith Cooper's site or Scott Martin's site that are primarily used for QTR linearization but I'd imagine you can re-purpose those a little to suit your needs (and I think it's probably worthwhile to measure > 1 specific density patch (esp. if you have a kinda automated method)).
-
using the i1P 'Measure Chart' or 'Measure Reference Chart' allows you to define your own chart (for you, a 1 swatch chart)
and measure and save.
As Mark said, you probably will need to do a little manual work to save the L*a*b* #s in a spreadsheet and graph that over time.
***
A chart definition describes the quantity and layout of a series of color patches.
When using the i1Pro, the i1Pro2 or the i1iO, you can specify the number of patches, in rows and columns, that you intend to measure.
Use this option when the patch set for a printed chart is unknown, but you wish to record the color values of the printed patches for the purposes of comparison and analysis.
***
This appnote is mostly applicable (although you're obviously not doing this to customize a CC24 for scanning reference)
https://www.xrite.com/-/media/xrite/files/apps_engineering_techdocuments/c/custom_reference_data_en.pdf
Also, I've seen 21 and 51 step BW charts on both Keith Cooper's site or Scott Martin's site that are primarily used for QTR linearization but I'd imagine you can re-purpose those a little to suit your needs (and I think it's probably worthwhile to measure > 1 specific density patch (esp. if you have a kinda automated method)).
Yes, I agree, it's better to measure a gray-scale consisting of a range of shades of gray rather than one alone. This way one can assess the hue constancy over the range as well as the accuracy of Luminance rendition from one tonal step to the next.
-
It'd be interesting to see/understand when/how the printer decides to start mixing (or un-mixing) some of the various blacks as it walks the density ramp (which would probably result in some a*/b* deviations)
-
This is super easy to do in Argyll. You can print out a patch chart and the do your readings over time. Everything can be imported into Excel for 'time-lapse' comparison.
-
It'd be interesting to see/understand when/how the printer decides to start mixing (or un-mixing) some of the various blacks as it walks the density ramp (which would probably result in some a*/b* deviations)
My understanding is that in a normal application-based ICC profile controlled printing pipeline, the CMS and the driver treat grays like they would treat any other colour. The quality of the profiles in helping to map adjustments from source to destination would be most important, and in this fundamental respect X-Rite's basic technical functionality is pretty good provided the profile creator knows what he/she is doing when making the profiles.
-
This is super easy to do in Argyll. You can print out a patch chart and the do your readings over time. Everything can be imported into Excel for 'time-lapse' comparison.
Likewise with i1Profiler, by saving the data from the patch readings in i1Profiler CGATS Custom format and importing this *.txt file into Excel.
-
Or pull it into QTR and it will dump out the .txt file w/ all the Lab/XYZ numbers and the ASCII 'graph' like you would if doing a linearization.
In Argyll, you could use 'spotread'.
-
Another way to do this is to check Aardenburg archives and see if they have tested the paper/printer combination. If so, it would be a much better data set.
-
Another way to do this is to check Aardenburg archives and see if they have tested the paper/printer combination. If so, it would be a much better data set.
In respect of what he tests for, sure, but if the purpose is to monitor drift of one's own printing set-up, a custom approach is inevitable.
-
In respect of what he tests for, sure, but if the purpose is to monitor drift of one's own printing set-up, a custom approach is inevitable.
Unless he has the equipment to do accelerated light fast testing, it is highly unlikely that there will be any observed change with respect to a middle grey (unless it is a dye printer) under normal viewing conditions. One approach would be to put the print in position where it gets direct sunlight exposure or close enough to a light source with enough UV. I reported last year about going back and looking at some of my prints that have been hanging in an office location for 10 years with at least 12 hours of office lighting per work day. I wish that I had a color checker print hanging as that would have permitted me to do some direct measurements. Even so, the print colors (Epson K3 inks) looked very good. I also was able to measure the lighting intensity.
-
OK, two kinds of drift - colour fading of the print, and drifting of the printer over time. My latest comment was on the latter. On the former - yes looking at Aardenburg makes a lot of sense; but he can also take interval measurements of known gray values from his own print under his own illumination conditions and plunk them into a spreadsheet to compare. Seems simple enough. But nothing "accelerated" of course.
-
Thanks everyone for some great answers. They've raised more questions. :)
My understanding of what he wanted was to be able to make a test chart i1Profiler with say 12 or 24 patches of middle grey. Print that chart as one makes a print profile, let it dry down overnight, and measure the chart. Save that data, and then after some period of time or light exposure, remeasure the data points and do a comparison.
I have been to the xrite site for the scanner reference one but I am not sure how that can strictly apply. I would need to read the print somehow and seem unable to generate to patch set or test chart to the required patches. It also only gives me an rgb option, not b+w.
-
Thanks everyone for some great answers. They've raised more questions. :)
My understanding of what he wanted was to be able to make a test chart i1Profiler with say 12 or 24 patches of middle grey. Print that chart as one makes a print profile, let it dry down overnight, and measure the chart. Save that data, and then after some period of time or light exposure, remeasure the data points and do a comparison.
I have been to the xrite site for the scanner reference one but I am not sure how that can strictly apply. I would need to read the print somehow and seem unable to generate to patch set or test chart to the required patches. It also only gives me an rgb option, not b+w.
This has nothing to do with scanners. Do you have a Spectrophotometer? What version of i1Profiler are you using? Is it licensed for using Measure Chart?
-
This has nothing to do with scanners. Do you have a Spectrophotometer? What version of i1Profiler are you using? Is it licensed for using Measure Chart?
Mark, I have i1Pro2 and i1Publisher software 1.8.3 and Colorthink Pro 3.0.4
My licensing information says it is licensed for all except CMYK printers.
Under Printer Workflow selection, I can see Measure chart, select it, Define a chart with Device setup is i1Pro2 etc as shown in the attachments. The issue is that I am unable to generate a chart based on that.
Hold on, after writing the above, I stumbled on Patch Set Generator at the bottom and the 4th attachment shows that. I should be able to input a few mid grey values and then save that as a patch set, use that to generate a target and then measure the target a few times and see what occurs say under exposure to sunlight over say 6 months.
Does that sound feasible?
-
OK, you've defined the chart as having 24 patches, which is fine. Now change the Measurement Mode to "Spot" and you will take a spot reading (that's one shot) of each of his 24 patches. You must read exactly 24 patches for this to work. Measure them in the same order every time. Then in the little Page Data bit at the bottom left of the GUI, click on Save; when your computer file structure comes up, migrate to the folder where you want to keep these measurements, for file type select i1Profiler CGATS Custom.txt, click save. Remember that location. Now go to Excel, open a new spreadsheet and go to File>Import. When the file system comes up, migrate to the file you just saved out of i1Profiler and click Open or Import or whatever it says in your version of Excel. That will import into Excel the data you measured as a list in CGATS format. This is what you want to have. You'll repeat this exercise periodically and compare numbers between episodes.
-
OK, you've defined the chart as having 24 patches, which is fine. Now change the Measurement Mode to "Spot" and you will take a spot reading (that's one shot) of each of his 24 patches. You must read exactly 24 patches for this to work. Measure them in the same order every time. Then in the little Page Data bit at the bottom left of the GUI, click on Save; when your computer file structure comes up, migrate to the folder where you want to keep these measurements, for file type select i1Profiler CGATS Custom.txt, click save. Remember that location. Now go to Excel, open a new spreadsheet and go to File>Import. When the file system comes up, migrate to the file you just saved out of i1Profiler and click Open or Import or whatever it says in your version of Excel. That will import into Excel the data you measured as a list in CGATS format. This is what you want to have. You'll repeat this exercise periodically and compare numbers between episodes.
Mark, thank you for that. That really helps.
Sent from my SM-G950F using Tapatalk
-
I stumbled on......
There's the problem you need to ask x-Rite to solve; poor documentation.
A lost cause, but you can but ask again.
see what occurs say under exposure to sunlight over say 6 months
If you're using a pigment based printer on photo printing paper, I doubt you'll be able to measure any deterioration in that time scale.
There's likely to be a change immediately after printing as the solvents dry out. Then the changes are likely to be smaller than the accuracy and repeatability of the instrument.
-
If you're using a pigment based printer on photo printing paper, I doubt you'll be able to measure any deterioration in that time scale.
There's likely to be a change immediately after printing as the solvents dry out. Then the changes are likely to be smaller than the accuracy and repeatability of the instrument.
For me I use a 3880 with OEM inks but don't know what the person who asked me is using.. :(
-
But you have the answer to your original question.
Yes, you/someone could make a target, print it and repeatedly measure it over time and graph the results.
Depending on the ink/paper combination, a slight shift *may* be observed depending on the time span involved.
Going through Aardberg's database might yield more information depending on the exact paper/ink combo.
Most of the work is already done (esp. if you have a Mac (but ultimately agnostic, you just want the CGATS data))................
http://www.northlight-images.co.uk/black-and-white-print-linearisation-with-i1profiler-and-qtr/