A basic guide (see post #1) to setting up ARGYLL CMS profiling on your computer

[RogerB]

The file size generated by the batch file in question is down to the -T360 parameter. This gives a 16-bit, 360dpi TIFF. I agree with Grandad35 that 72dpi is sufficient for printing targets. although I personally use 180dpi for my Argyll targets simply because it makes the text look nicer! Even so, a 528-patch A4 chart generated by Argyll is only a little over 0.5MB.

As far as bit depth goes, it would be very hard to convince me that 16-bit files give any improvement in profile accuracy over 8-bit, given the measurement capability of the CM or even the i1Pro. And that's even assuming that you have a 16-bit print pipeline!

 

[sampheap]

@Emulator, "Argyll_V1.6.3 Profiles (The folder you work from and contains the created profiles, .ti and .tif files)"
So I will need to create one more folder in the bin 'Profiles'? Are .ti 1 and 2 created first or after executing the CAP480?

And about the .tiff file, is each .tiff generated from one computer different and cannot be shared or used by another then how is it possible we handed the .tif file to a few friends to print their patch? Are the .ti files to be re-generated each time we do the read?

I am sorry to ask so many questions, but I am really a bit lost here.

 

[Emulator]

bin is a folder which contains binary files and is located in the C: \Argyll_V1.6.3 folder, with many other folders and files. You create the Argyll_V1.6.3 folder when you install Argyll and it is usually placed in the C: HD drive. The Batch files like CAP480.bat are placed in the C: \Argyll_V1.6.3 \bin folder. The PATH command which you edit (see the Argyll documentation referred to in post#1), allows the computer to find the .bat files, from the command line input.

C: \Argyll_V1.6.3 Profiles is a folder you create to hold the profiles, the .tif files and the .ti files. You may choose to use other names when you create these storage areas. The .tif files and the .ti files are created by Argyll in response to the commands contained in the CAP480 (or other .bat) batch file, each time you run the .bat file. You set up the command line prompt to run from this folder (see post#1) and it will place the files in the folder, keeping the process tidy.

Each set of .ti files and each.tif file generated by running the .bat file will be different. They are generated by the use of a randomising process in Argyll. The .ti3 file is generated after you do the read and is used by Argyll to create the profile. They have to be available to Argyll and processed together.

The fact that you can give a .tif file to someone else and they can print it and return the print to you, allows you to create a profile for someone else. There are strict copyright laws associated with this process on most profiling software. However, I believe Argyll does allow use for this purpose. But the object of this thread is to encourage those people interested in creating profiles to install Argyll on their own computer and create their own profiles.

 

[The Hat]

@Emulator, "Argyll_V1.6.3 Profiles (The folder you work from and contains the created profiles, .ti and .tif files)"
So I will need to create one more folder in the bin 'Profiles'? Are .ti 1 and 2 created first or after executing the CAP480?

And about the .tiff file, is each .tiff generated from one computer different and cannot be shared or used by another then how is it possible we handed the .tif file to a few friends to print their patch? Are the .ti files to be re-generated each time we do the read?

I am sorry to ask so many questions, but I am really a bit lost here.

I think you’re getting the wrong end of the stick here, a profile that is generated on one printer can’t be successfully shared with another printer, the profiling process doesn’t work that way.

Firstly you need your monitor calibrated properly, then the use of good quality inks from the one supplier and then be sure that your printer is working properly and correctly.

Lastly you need to know exactly which paper type you’re going to be printing on that needs this specific profile, only then can you properly begin to make a start on producing a worthwhile genuine profile.

If at any time you change your print settings, ink, paper or even the brightness in your computer room then you’ll have to start the whole process over once again.

Now once you have gotten your first successful profile then you can exchange tip and ways to improve you technique and do it faster, this process only works successfully on photos however it next to useless for printing with graphic files.

Posting exactly the same thread again won’t get you answers any quicker, and I would suggest you read this whole post from the very first thread again because you’ll actually find the answers you’re looking for are right there !

 

[Emulator]

Found this and other test targets can be generated by Argyll timage test tool.

 

[Emulator]

@RogerB comments at post #165 raised the issue of comparing two .ti3 files generated from two separate readings of a single target patch A4 sheet, in colverify, although he did not provide a printout showing the detailed results.

The printout below, shows the results of one such test, using the ColorMunki Photo with the 6.5mm width reading slot in the reader illustrated in post#166 . There were no functional read errors during either of the two patch read sequences.

C: \Argyll_V1.6.3 Profiles>colverify -v targen480-24-05-2.ti3 targen480-24-05-1.ti3
No of test patches = 480
No of test patches in worst 10% are = 48
No of test patches in best 90% are = 432
Verify results:
Total errors: peak = 2.119612, avg = 0.157544
Worst 10% errors: peak = 2.119612, avg = 0.461349
Best 90% errors: peak = 0.273141, avg = 0.124420
avg err X 0.000434, Y 0.000424, Z 0.000401
avg err L* 0.051569, a* 0.075651, b* 0.104819

 

[RogerB]

@RogerB comments at post #165 raised the issue of comparing two .ti3 files generated from two separate readings of a single target patch A4 sheet, in colverify, although he did not provide a printout showing the detailed results.

A very god result from the CM, apart from the peak error which looks as if it could be an outlier. Here's a result for a 529-patch chart (23 x 23) with a patch size of 7.3 x 9.7 mm. I think we are both at the point where the accuracy is more than adequate!

 

[Emulator]

A very god result from the CM, apart from the peak error which looks as if it could be an outlier. Here's a result for a 529-patch chart (23 x 23) with a patch size of 7.3 x 9.7 mm. I think we are both at the point where the accuracy is more than adequate!

Have you ever considered using a conventional scanner (as opposed to an Xrite scanner) to read the targets and create a printer profile? Is there software around to convert the scanned image to a .ti3 file? If it proved accurate enough it would be a great time saver!

Added later: It seems that Argyll scanin, which will "Convert a TIFF image of a test chart into .ti3 device values", would do the job.

It will be interesting to compare .ti3 files of the same chart produced through a scanner and by CM.

 

[Emulator]

I was curious to see just how far the ColorMunki Photo would go with a large number of patches on A4.

I created a 957 patches target on a single sheet of A4. This resulted in 7mm wide by 7.5mm high patches when cropped and resized to maximum, to fit the A4 sheet.
targen -v -d2 -G -e2 -B2 -g24 -f957 %pcn1%
printtarg -v -ii1 -a0.75 -A0.5 -T300 -M2 -P -p210x297 %pcn2%

I read the patches twice with the CM Photo (using my standard measuring frame) and as per RogerB's suggestion, compared them in Colverify, see below

C: \Argyll_V1.6.3 Profiles>colverify targen957-2.ti3 targen957-1.ti3
Verify results:
Total errors: peak = 3.031422, avg = 0.208956
Worst 10% errors: peak = 3.031422, avg = 0.647500
Best 90% errors: peak = 0.366032, avg = 0.160485
avg err X 0.000529, Y 0.000616, Z 0.000576
avg err L* 0.072308, a* 0.107413, b* 0.128842

I think the ColorMunki Photo will take a lot of beating!!

This is the chart image on Sihl glossy, which has matured over the last 4 weeks or so.

 

[3dogs]

I reckon if I had to get into this profiling thing seriously I’d have to give up drink altogether..

Follow ME,

copious quantities of water, followed by an Aspirin (no hangover) and try again to fathom whats being concluded..........If my colormunki started spewing out data like that I'd sent it back to where it came from and take myself to the Zoo, for aged residential care and respite!

I have an uneasy feeling tho, somewhere lower down that I am missing something important, really important but WHAT?

Andrew