Using ArgyllCMS + Colormunki to produce excellent printer profiles

[nrdlnd]

I think the i1 Pro can also read 750 Patches, if the Colormunki can read it,

Yes I think so also, maybe more patches especially with the good ruler that comes with the instrument. I didn't care or want to crop and resize the chart but used the commands in Argyll. What I could try is to to change the size of the patches and then print borderless to get them bigger (I don't want to crop and resize as I want an easily repeatable process). I have all my commands in a text editor to edit them and then copy to the command prompt. I think it's also important to have good readings to get a good result but I do not know what happens if the meter partly reads outside of the patches.
EDIT: I've made 760 patches by using a0.86, 0 margins in the Adobe Printer Utility and then in the printer driver using "Borderless" and "Auto expansion = min". The resulting patches are still 7mm wide that the i1 Pro can read. Of course with crop and resizing there is possible to squeeze even more patches on one A4 paper but I see no meaning with that as the quality difference is minimal (see my above edit).

Per

 

[supergrobi]

I built a ruler with a slot with the width of the patches (8mm). Without resizing and cropping it would be to narrow, a simpler process would be nice, but it is ok, next time i will make a action in Photoshop to crop and resize them for the next patches.
I also believe there is not a big difference to squeeze even more patches on one page, maybe it´s better to print more patches on a second page, if a higher quality is needed.
Can you see differences in your prints? Do you use the patches of one single page or more?

Matthias

 

[nrdlnd]

Can you see differences in your prints? Do you use the patches of one single page or more?

I have edited my earlier posts. The differencies are very small and the profiles made with ArgyllCMS are very similar. Maybe the "preconditioned" profile and the one with double amount of patches on two papers are a little better but it's difficult to see with bare eyes. I haven't analyzed the profiles.

Per

 

[Emulator]

I quite happily use 957 patches on A4 with a ColorMunki Photo and a mask made of 'artist's matt black card' and balsa guides, with a slot of 7mm width. I think the CM would work with a narrower slot than that, but I have not tried it yet.

It is important to prevent adjacent patches colour, either from the sides or the strip you are measuring, infiltrating the reading. Ambient light can also influence the quality if you are not careful, although the clearance is only about 12 thousandth of an inch.

I suggest you also try at least 30% grey scale patches to improve the overall quality of the profile.

Alternatively you could always resize your A4 patch sheet to A3 and make it easier to read.

 

[RogerB]

I quite happily use 957 patches on A4 with a ColorMunki Photo and a mask made of 'artist's matt black card' and balsa guides, with a slot of 7mm width. I think the CM would work with a narrower slot than that, but I have not tried it yet.

It is important to prevent adjacent patches colour, either from the sides or the strip you are measuring, infiltrating the reading. Ambient light can also influence the quality if you are not careful, although the clearance is only about 12 thousandth of an inch.

I suggest you also try at least 30% grey scale patches to improve the overall quality of the profile.

Alternatively you could always resize your A4 patch sheet to A3 and make it easier to read.

At the risk of repeating myself, I would urge anyone attempting to fit the maximum number of patches on a given size sheet to verify that their measurements are sufficiently repeatable. All you need to do is measure the same chart twice (or more) and compare the two .ti3 files. You can do this very easily in Argyll using colverify. The command that I use for this is

colverify -v -c -s -N -D -L ProPhoto.icm Reference.ti3 Sample.ti3

where Reference.ti3 and Sample.ti3 are the files you want to compare. Make sure that you have ProPhoto.icm in your relevanat Argyll directory.

Argyll's colverify will compare the two files and give a summary of the differences. You should expect an average error (CIE94) less than 0.5 and a peak error less than 1.0, even for the "Worst 10%" as reported by colverify. It's surprising how a few errors can result in a poor profile.

 

[nrdlnd]

At last is something happening in this thread!

I quite happily use 957 patches on A4 with a ColorMunki Photo and a mask made of 'artist's matt black card' and balsa guides, with a slot of 7mm width.

I will try do do this with the ruler to i1 Pro. Put black card under the opening of the reading slot to get better readings.

I would urge anyone attempting to fit the maximum number of patches on a given size sheet to verify that their measurements are sufficiently repeatable.

Great! I will try this also!

Per

 

[RogerB]

Here's something for impatient Argyll users (like me).

My computer takes about 5 minutes to generate a high quality profile with my 1058-patch target using Argyll CMS. Yesterday I had four profiles to generate and not much time. When the first profile was being processed it occurred to me that I could be measuring the next pair of charts, so I opened chartread again. Sure enough it opened up as normal and I was able to start measuring while the first profile was being processed. Measuring a 1058-patch target takes me about seven minutes, so the first profile was finished before I had measured the second targets. Then, while the second profile was being generated I was measuring the third pair of targets .... and so on. Worked like a charm for me, as someone who hates sitting doing nothing.

FWIW, this is on Win7 64-bit.

 

[RogerB]

I will try do do this with the ruler to i1 Pro. Put black card under the opening of the reading slot to get better readings.
Per

Be careful! The i1Pro measures a spot about 4mm in diameter, but the accuracy of the reading for dark tones is reduced (a lot) if the measuring head is raised too far above the target. If you do add a mask make sure that it is very thin. My experience is that it is not necessary, but I'm not trying to get the absolute maximum number of patches on a sheet .....

 

[supergrobi]

I quite happily use 957 patches on A4 with a ColorMunki Photo and a mask made of 'artist's matt black card' and balsa guides, with a slot of 7mm width.

I suggest you also try at least 30% grey scale patches to improve the overall quality of the profile.

It would be nice if you could give me an example for the command.
The "colverify " command sounds really great, very useful, thanks a lot!

 

[nrdlnd]

Be careful! The i1Pro measures a spot about 4mm in diameter, but the accuracy of the reading for dark tones is reduced (a lot) if the measuring head is raised too far above the target. If you do add a mask make sure that it is very thin. My experience is that it is not necessary, but I'm not trying to get the absolute maximum number of patches on a sheet .....

My afterthought is that this black mask may not be completely "black" and may reflect some light and disturb the reading. I'm trying to get the strips at least 7mm wide.