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

[RogerB]

Though I got successful after re-read each row many times, both profiles are problematic as you can see in the picture above. Can we enhance the .icm profile beside re-read?

I think it is worth saying to anyone using Argyll (or any other profiling application!) that if you have to keep re-reading, it means that the overall quality of your readings is poor. Re-reading does not make a significant improvement in the average errors of the readings. If you have to keep re-reading, then try changing your reading technique until you can eliminate the re-reads. As for enhancing the profile, the answer, unfortunately, is no. If the measurements are bad the profile will be bad.

 

[Emulator]

I have uploaded new versions of the two profile files included in post #196, to correct some naming discrepancies.

I have also added a CAP957 version of a Sihl glossy profile.

 

[sampheap]

I have uploaded new versions of the two profile files included in post #196, to correct some naming discrepancies.

I have also added a CAP957 version of a Sihl glossy profile.

Thank you for the update.
I noticed the one done with CAP480 is giving darker colors than one by CAP957? I also experience the same darker color CAP480 with some paper, I will try CAP957 if that will improve indeed.

 

[Emulator]

I see that there are now 15,037 viewings on this thread and 903 in the last 18 days, or an average of 50 a day on a thread that had drifted to page 8 of "recent posts". See my other posting re "robots".

There have been no added posts in that time, although the number of viewings or downloads of the batch files at the bottom of post number one of this thread, have been increasing slowly.

Viewing other websites with a Google search shows that robots do look at uploaded files on threads and in one case displayed a broken up and disjointed version of one of these uploaded batch files.

It would be helpful if downloaders of the batch files could provide some feedback, regarding the success of the batch file process of generating profiles, good or bad.

Please resist humorous or off topic replies. This is intended to be a serious query and unrelated replies can bury a posting.

 

[Emulator]

Between Friday 19/09/14 and Sunday 21/09/14 (similar time of day) there have been 145 viewings in two days (72/day) and not a single posting. I think that fairly conclusively indicates that viewing numbers are largely influenced by "robots"!!

 

[The Hat]

Please resist humorous or off topic replies. This is intended to be a serious query and unrelated replies can bury a posting.

[

Hi Ian
I had resisted placing any humorous comments (Just) till this bunch of nuts and bolts appeared ..

 

[RogerB]

It would be helpful if downloaders of the batch files could provide some feedback, regarding the success of the batch file process of generating profiles, good or bad.

The batch file that I posted (ArgProf#2) was an attempt to make printer profiling a "turnkey" application and it has worked very well for me. Argyll has some quirks in file naming and I aimed to iron these out under the hood to give the user complete control over the .ti3 and profile names.

There is currently a thread on the Luminous Landscape forum started by a potential Argyll user who has berated Graeme Gill for not providing a fancy GUI for Argyll. He has been put firmly in his place by actual users!

For me, the lack of a GUI was initially a barrier, but encouraged by members of this forum I generated my own TUI (Text User Interface), which is in ArgProf#2. For anyone who hasn't tried this batch file the interface for reading a chart and generating a profile looks like this:-

The user input is inside the yellow box. No further user intervention is required - other than reading the chart of course. The profile is simply saved in the folder specified in the batch file. For me this is just as easy as any fancy GUI and much quicker to load!

As far as results go I have generally found the profiles to be very accurate. However, on occasions I have run into the dreaded dog-leg in the greyscale response. I have used Gretag Macbeth i1Match software (nice GUI) for many years and have been pretty happy with the profiles. Out of interest I have generated a .ti2 file that allows me to read the i1Match TC9.18 charts in Argyll to see how the profiles compare. The Argyll profile is almost always better - better colour accuracy, and more neutral greys. On the rare occasions when it isn't better, it's because of the greyscale dog-leg. In those cases I have accepted the slightly reduced accuracy of the i1Match profile.

I have no idea why Argyll should sometimes give this dog-leg, but I do know that it can have a nasty impact on shadow rendering on B&W prints. In spite of that, Argyll, with the TUI, is my go-to printer profiling application. However, I always use Gamutvision to check the profiles that I generate, so few, if any, bad profiles get through the net.

 

[lowell374]

The batch file that I posted (ArgProf#2) was an attempt to make printer profiling a "turnkey" application and it has worked very well for me. Argyll has some quirks in file naming and I aimed to iron these out under the hood to give the user complete control over the .ti3 and profile names.


I have no idea why Argyll should sometimes give this dog-leg, but I do know that it can have a nasty impact on shadow rendering on B&W prints. In spite of that, Argyll, with the TUI, is my go-to printer profiling application. However, I always use Gamutvision to check the profiles that I generate, so few, if any, bad profiles get through the net.

Care to share the details of how you use Gamutvision to check the profile?

 

[cls]

I would finally a real GUI implemented in W7 or sth. for ArgyllCMS, maybe even a "printerknowledge" exclusive

 

[RogerB]

Care to share the details of how you use Gamutvision to check the profile?

Gamutvision has lots of options that allow you to "explore" profiles, but the day-to-day process that I use is outlined in another thread http://www.printerknowledge.com/threads/profile-assessment.8249/page-4#post-64553 There are also quite a lot of posts in the current thread showing screen-shots from Gamutvision and giving some interpretation of their significance.

Of course, this is all just mathematics, and the ultimate test of a profile is how it prints. If the test charts were not printed correctly then the profile may look good in Gamutvision but produce mediocre prints. Even if the test charts are good and the reading of the chart is good there's no guarantee that the profile will be good. The cases that I cite above show this. Argyll gives poor B&W shadow rendering whereas i1Match using the same data gives very good rendering, although the colour accuracy is not as good as that given by Argyll.

For anyone generating their own profiles without the benefit of a tool like Gamutvision it's important to print one or two good evaluation images and to examine them very critically. Profiling applications can be fallible!