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

[Emulator]

I have re-read the 660 chart (after a coffee) and it has produced a virtually identical result with either batch file so the D50M2 seems to make no visible difference in this case. I don't think the omission of -H in chartread is valid either. If you don't try these web suggestions you don't make progress. Still haven't made the beep work in chartread.

 

[The Hat]

Hi Ian,

Did you try changing your Coffee…?

 

[Emulator]

Hi Ian,

Did you try changing your Coffee…?

We were given a box of nine different varieties of Whittard Coffee for christmas, varying in strength from 2 to 5, we are working down from 5 to 2, but most of it is being drunk by visitors.
(Edited to thumbnails to save space)



But the coffee I like best is Lavazza:


I think I'll try hot milk.

 

[Emulator]

@pharmacist,
Argyll comments about profile size for inkjet printers:

The number of target patches needs to be chosen, depending on the media size, the type of device, and the quality of profile required. For an inkjet device, something like 3000 test points or more is desirable for high quality profiles, while 500-1000 will probably suffice for a medium quality profile. A few hundred may be sufficient for a preliminary profile. Well behaved printing devices (such as a chemical proof, or a high quality printing press) may produce good profiles with 1000 to 2000 test points. Well behaved RGB devices such as CRT monitors may need only a few hundred points, if a shaper/matrix type profile is to be produced, while pseudo RGB printers, or other RGB devices that a CLUT type profile may be used with, should probably choose somewhere between 500 and 3000 patches. For 'N' color profile creation, 3000 or more test points should probably be used.

 

[pharmacist]

Hi Emulator,

note the sentence: Well behaved RGB devices such as CRT monitors may need only a few hundred points, if a shaper/matrix type profile is to be produced, while pseudo RGB printers, or other RGB devices that a CLUT type profile may be used with, should probably choose somewhere between 500 and 3000 patches

99% of most consumer printers are pseudoRGB, so 750 is well within the range and Colormunki uses even less patches (2x50), but I think with a a lot of presumptions. The expert target of the Spyder4print only uses 729 patches.

As I have stated above: the commercial target I uploaded (Marc-en-ciel target) uses less than 600 patches and still produces excellent profiles.

 

[Emulator]

but it does say 500 - 3000, so my 2250 is in range!

Looking at the sizes of the Argyll and ColorMunki lut table entries, where as Argyll creates the profile, I think the CM must have a pre-created, more detailed ideal profile and then pull it around with the few patches it reads. (Argyll first 2048, CM second 4096)

 

[The Hat]

I went a bit upmarket because I’m a Tea man myself..

 

[RogerB]

but it does say 500 - 3000, so my 2250 is in range!

I can feel a case of Post-Argyll-Stress-Disorder coming on here!

My experience is that more patches can produce better profiles but that the "better" profiles don't necessarily give better prints. The vast majority of the profiles I have made have been from 918 patches and for a well-behaved printer they are usually very good. Just out of curiosity I made a 1768-patch Argyll target (4x A4 sheets) and used the "Ultimate" quality setting. It took Argyll about six hours to generate the profile on my quad-core 3.2GHz machine. The analysis in Gamutvision looks truly excellent, with errors for the ColorChecker less than 0.2dE

However, I recently did a profile using 918 patches and the Gamutvision predicted errors are less than 1.0 dE as shown below. Now, predictions are all well and good but the real test is in the printing. So, I printed my version of the ColorChecker using Absolute Colorimetric and measured the Lab values for each patch. The numerical comparison shows an average dE for the best 90% of the patches to be 0.69 - pretty close to the Gamutvision prediction. (The worst 10% are the very light greys where the target L-value is higher than the paper can manage!)

From the Lab values I generated an image file (TIFF) to compare with the original file. The result is given below.



The point here is that the comparison of the actual printed vales with the target values shows that youcan't see any difference. So, if the print were even more accurate it wouldn't look any different, even under these super-critical conditions.

 

[Emulator]

I went a bit upmarket because I’m a Tea man myself..

Is that a mahogany box or plastic?

 

[Emulator]

@RogerB I think I read somewhere in Argyll notes that the "u" option should not be used in profiles and was only intended for testing.
6 hours! Have a coffee. I think it is a case of decreasing returns.

But it's all good fun.