i1Studio with ArgyllCMS

[TonyW]

Years ago I tried a colleague's ColorMunki for printer profiling.  It seemed capable of producing acceptable profiles but not particularly better than the manufacturer's canned ones, but came into its own with profiling non-OEM inks and papers.

I thought some time ago that I may have the chance to purchase a GretagMacbeth Spectrolino, but so far the current owner wants to keep hold of it.   So being impatient and wanting to experiment (keep the brain active during lockdown  ) I have ordered an X -Rite i1Studio and would appreciate some thoughts as I have made some assumptions which I hope will be at least somewhere near the mark, or corrected here by more experienced users:


⦁   The i1Studio appears to be a much more capable device than the X -Rite software allows it to be, although it will never be a real match for the professional X-Rite products such as the i1 Profilers it may be close enough for me for now?

⦁   As much as I now steer away from CLI environments hearing good things  I decided to look at ArgyllCMS.  What an impressive piece of work (already made a donation!) , with so many options to understand and explore!   I feel confident that with a little effort to understand the finer points that this software should push the i1Studio to its limits and mine 

⦁   Having read quite a lot of views about the number of test patch values you can print on a single A4 sheet I have played with creating a few 210, 480, 660, and 750 patches.  As I do not have the instrument yet I have not printed but I know I will need to make up some form of a ruler to guide the instrument. 

Obviously, there has got to be a limit on how small the patches can be for accurate and reliable measurements, any thoughts on what that needs to be?

I guess that at some stage with this particular device there comes a point of diminishing returns with more and smaller patches not adding any value?

Is it likely to be worthwhile to make a target with slightly larger patches and print more pages?

As I also have the X-Rite i1Display Pro my intention is to continue using that for monitor profiling with the Eizo Colour Navigator software as I understand it should be more accurate for screen display calibration than a spectrophotometer.

Appreciate your thoughts and experiences

[HarveyM43]

Anders Torger did a pretty great tutorial on the Colormunki (i1Studio)/Argyll here-
https://torger.se/anders/photography/argyll-print.html#toc18

I did try varied patch sizes- using the i1Pro patch style you get 462 patches per sheet. Even with a guide it can be painful; a misread and you need to reread the set or you’re left figuring out which column/patch is bad & using a text editor to change the results file.

There's also a GUI here: https://sourceforge.net/projects/cprofiler2020/  but I have no experience with it.

[TonyW]

Thank you, I have just finished reading Anders tutorial very useful

As to the patch size and potential for misreads, I can see the issues of keeping to the straight and narrow when reading.  It is my intention to fabricate a simple dual guide system to sit on either side of the i1Studio to keep it anchored in the 'x' direction over the patch columns with free movement in the 'y' direction up to stop points where paperwhite can be read at the top and bottom of the target. Inside this guide, will sit a black mask sized appropriately for the i1Studio lens aperture.  A further refinement might be to have a plate to hold the paper straight and parallel to the guide with stops fitted to the guide to enable it to slide into position in the 'y' direction.

Thanks for the link to the GUI, I have downloaded and will have a play - it looks quite comprehensive although I have to say I quite enjoyed playing with CLI and batch files after such a long break from the early days of DOS commands 

[HarveyM43]

Glad you had a chance to look through the links. I'll just add a quick note that besides controlling the X & Y directions the Z direction (height) can adversely your readings. The thicker your guide base is the greater the cross-talk from adjacent column patches.   

[GWGill]

I'll just add a quick note that besides controlling the X & Y directions the Z direction (height) can adversely your readings. The thicker your guide base is the greater the cross-talk from adjacent column patches.

The Munki/i1Studio isn't designed to be spaced away from the paper - the geometry of the illuminant doesn't allow for that.
So it really should be sliding on the paper itself. A "gotcha" is that the instrument has two small rubber pads that
are designed to stop it slipping when taking individual measurements, and these impede smooth sliding. If
you are mainly doing strip reading you could remove these, or if you leave them they will wear down in time anyway :-)

[TonyW]

Thanks, guys for the warning about spacing from the paper to the i1Studio.

I made a quick mock up (attached), no i1Studio yet, of what I was thinking about, using materials at hand i.e. white foam core for the guide rails and some black card to sit over the A4 target sheet. 

The black card is relatively thin at 0.33mm and the aperture I cut approx 9mm wide.  Obviously I would expect this to wear and need occasional replacement.

Do you think that the card thickness is enough to cause issues without removing the feet? 

[HarveyM43]

I can't give you a definitive answer; I can tell you that my 840 patch set without guide had a max dE2000 1.00 lower (better) than my 924 patch set done on a guide similar to yours.
I think if I were to do a guide again I would make colormunki ride between the guide rails without the base paper slit.

[TonyW]

Thanks for the added info.  i1 now delivered so will give it a try 

[NAwlins_Contrarian]

I have a ColorMunki Photo, which is the next-older but very similar (some say identical) device to the i1Studio. Like you, I have explored using Argyll CMS as an alternative to the X-Rite software. I had a thread about that at:
https://forum.luminous-landscape.com/index.php?topic=134572.0

FWIW, I still have not tried it. Frankly, in part Argyll intimidates me--not so much the command line (I was five years into using and programming computers before I used one with a GUI), but the numerous options, switches, and permutations. Some thoughts, which you may or may not find helpful:

* The i1Studio software (currently v. 1.5.1) seems to me somewhat more accurate and flexible than the older ColorMunki Photo software (last version 1.2.4). That's not to say that you can't do better with Argyll--you probably can.

* The i1Studio software process remains a bit unconventional: print 50 standard patches, measure them, and then the software calculates a customized set of additional patches, 50 for color profiles and 100 for B&W-oriented profiles, which you then print and measure, from which the software creates the ICC profile.

* The i1Studio software process seems to produce pretty good profiles. The software seems to me pretty smart--prints made using these profiles seem (subjectively) to me far better than you'd expect from only 100 or 150 patches. I don't know how they'd compare to e.g. profiles made using conventional 729-patch (9x9x9 matrix) or 1728-patch (12x12x12 matrix) sets.

* There are reports of driver-related bridge-burning of sorts between the X-Rite software and Argyll. IOW if you install Argyll, you may well have issues trying to use the X-Rite software without changing the drivers back.

* If you use Argyll, the process will require far more patches, and therefore presumably far smaller patches. The i1Studio software produces patches somewhere around 1.0 x 1.25 in / 25 x 32mm. I seem to recall concluding that the smallest practicable patch for the hardware is about 0.35 x 0.35 in / 9 x 9 mm. AFAIK, all else being equal, smaller patches mean fewer samples per scan, and therefore at least somewhat lower accuracy. I think I'd settled on the assumption that I could use about 400 patches on a U.S. letter-size sheet--remember the device needs both white space before and after the patches during its run, and you need the paper thickness left and right to make it track correctly. But this was just my impression, and certainly YMMV.

* As already stated, the i1Studio is designed to slide directly on the paper. I don't think you could or should try to use in not on the paper. You might use a ruler or build something to guide that slide to facilitate using smaller patches.

By all means, if you go this route, please report back with your experiences. I remain interested in this approach.

[TonyW]

^^Excellent information thanks for the link and write up of your experience.

Due to having to do some household repairs and family business I have only had chance to use the i1Studio to play with the X Rite software and produce the standard 50+50 target.  Trying with some unfamiliar paper Pinnacle Lustre it seemed to produce visually a similar look to the manufacturers profile even though the paper is fairly high in OBA’s.  Purely subjective of course but an encouraging start.

What I was not too keen on (although pre warned after reading Keith reviews) was having to produce the first target then wait for drying before reading and having to repeat for the second.  To be safe I let dry overnight, perhaps overkill?  ArgyllCMS should solve the need to produce a second sheet for a modest profile and even multiple sheets can all be printed at the same time and read together after drying.

My own computer experience similar to yours prior to experiencing the first GUI in Windows.  I agree the number of switches and permutations initially quite intimidating.  Having now played a little with the command line to produce targets and print a few I feel a lot more confident and TBH quite enjoyed going retro.  BTW the GUI linked to in an earlier post looks to be very comprehensive and IMHO worth investigating.  There are still loads of switches and options that, frankly I do not have a clue what they are for so I guess I will learn as and when needed.  Like eating an elephant one bite at a time 😉

I am fully expecting possible hiccups needing to switch between drivers but I suspect/hope it may turn out to be a non issue as I believe that I will stay with Argyllcms for the duration

I am guessing that you may be right with 400 - 600 patches A4 being a reasonable max per sheet A4.  With multiples for greater accuracy, bearing in mind the potential for the laws if diminishing returns kicking in at some point. 

I will be using a guide ruler and need to look at a way to consistently place the ruler in the optimal position for reading small patches.  I might give the paper mask a try at some point as a comparison but have taken Graeme Gills suggestions on board as strong recommendations.

[huluvu]

Anders Torger did a pretty great tutorial on the Colormunki (i1Studio)/Argyll here-
https://torger.se/anders/photography/argyll-print.html#toc18

I did try varied patch sizes- using the i1Pro patch style you get 462 patches per sheet. Even with a guide it can be painful; a misread and you need to reread the set or you’re left figuring out which column/patch is bad & using a text editor to change the results file.

There's also a GUI here: https://sourceforge.net/projects/cprofiler2020/  but I have no experience with it.

I know this thread is three years old already but I gained so much from Anders Torger's Argyll tutorial that I wanted to bump this up again. Thank you, HarveyM43 for linking the tutorial and thank you Anders Torger for your effort! As a "command line averse", I was finally able to use Argyll (to great results) using this guide.

[kers]

+1 thank you for bringing it up

I use torges Profile designer for reproductions- well made and accurate.
He always explains a lot about how and why.

[HarveyM43]

I know this thread is three years old already but I gained so much from Anders Torger's Argyll tutorial that I wanted to bump this up again. Thank you, HarveyM43 for linking the tutorial and thank you Anders Torger for your effort! As a "command line averse", I was finally able to use Argyll (to great results) using this guide.

You're very welcome!

I spent a couple months experimenting with ArgyllCMS during lockdown. Mr Gill has created an amazingly broad & deep program and Anders Torger's guide is a great introduction to it.