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

[Emulator]

Many thanks @pharmacist, it took 13.5 minutes on my ancient Intel dual core. I must say it was nice to see a computer working hard!

The only trouble I had after your last posted advice, was appreciating the correct directory protocol. Although you had stated so, I had not taken on board the significance that you should start in the Profile destination directory, where you had placed the group of ti files, rather than the Argyll program directory. The PATH setting obviously took care of access to the program files. The fact that the program could not find the A4profile.ti3 file drove the point home.

I will now have a look at the profile in Gamutvision.

The next target will be to get @The Hat & @rodbam making Argyll profiles.

 

[The Hat]

Hi Ian,
You got a better chance of these winds stopping all together than me using any colour profile gadget.
I just never seen the need to use them in my line of work, if I ever get a wrong colour I just tweak the printer a little and that works for me, usually a one off thing..

 

[PeterBJ]

Many thanks @pharmacist,
The next target will be to get @The Hat & @rodbam making Argyll profiles.

The cost of the necessary hard- and software?

 

[Emulator]

What is the view on display profile settings? I have created D50 and D65 at 110cd/sqm. not sure which I prefer.

The new profile has produced a very pleasing Evaluation Image print.

 

[pharmacist]

I think I have chosen D50. Nice to hear the evaluation image print is OK with your first ArgyllCMS created printer profile.

 

[Emulator]

@pharmacist, I noticed that PhotoShop Elements uses the internal ICC file name in its profile list, where as other software seems to use the external name. It looks as though I need to use the same name in both.
I have added an image to post#34, although it is difficult to do the print justice, with all the scanning and conversion.

The puzzling thing is that Gamutvision, ProfileDump and Profile Inspector results are not too good, I'll have to investigate further.

These are good apart from the spikes, but they are minor.

 

[pharmacist]

The most important thing is consistency of reading and reading not too fast to get consistent results. You can check with soft proofing the image with the created profile and check for strange color transitions with both relative and perceptual content. Any errors can be observed in strange banding or sudden strange color transitions, which are not expected. Then rereading the target is compulsory. Also check the printed target for strange and subtile banding in same patches, which influences the reading and the results of the profile.

Concerning naming: it is up to you if you want both same external and internal names. I sometimes use _ in the external names (to prevent breaking the name), but use spaces in the internal names.

 

[Emulator]

Yes, you can see some strange glitches in the top right corner display of the images above in post #36.

 

[Emulator]

Hi @pharmacist, I re-read the same patch sheet, with the benefit of a little experience, without any errors. After I pressed the CM switch in the start position, I waited for three cursor flashes and then measured a row in approximately 6 seconds. The profile creation ran smoothly, in about 14 minutes. The results look much better, see Gamutvision round trips for Colorimetric and Perceptual intents, Delta E*ab values and spectrum, no glitches this time. As we found before Colorimetric has very good Delta E values.

 

[pharmacist]

Another tip reading profiles: please put at least 1 or better 2 sheets of the same paper underneath the target, to determine the paper white. This improves the quality of your profile.