i1Studio Spectrophotometer logistical question

[NAwlins_Contrarian]

I want to rent an X-Rite i1Studio Spectrophotometer to make a bunch of ICC profiles for my printer + paper combinations, but I'm hoping somebody can help me with a potential logistical problem. Basically, I have heard something to the effect that the X-Rite software prints the first set of patches, then sits there waiting for you to measure them before it will do anything else. This could be a problem because my eyes suggest to me that there are visible dry-down changes until relatively long times after printing, and what I want to do is print the first set of patches, let them dry over night, then come back and measure them, then print the second set of patches, and let them dry for some hours, then measure them. But if I rent the device for a week and want to make profiles for several different papers on each of two different printers, that will only be practicable if, say, I can print on five or ten different paper types the first set of patches, let them sit overnight, then the next day resume where I left off and measure those patches, and print the corresponding second sets of patches for each of the five or ten different paper types, and so on. Does the X-Rite software let me do this? Does it let me save a 'profile in progress' file for each paper, move on to another paper, then later return to the first paper?

I understand that some people measure patches after 15 minutes. Maybe with some printers and some papers this is fine. Maybe my eyes are wrong, and in fact 15 minutes would be plenty enough waiting. Many of y'all are experts, so I'd welcome your opinions on that too.

In case it matters, combinations for which I want to create custom profiles include at least (all using OEM dye inks):
* Epson R280 with Canon Pro Premium Matte PM-101
* Epson R280 with Canson Platine Fibre Rag
* Epson R280 with Hahnemühle Photo Gloss Baryta
* Epson R280 with Hahnemühle Photo Silk Baryta
* Epson R280 with Inkpress Metallic Gloss
* Epson R280 with Inkpress Metallic Satin
* Epson R280 with Mitsubishi Pictorico Pro Hi-Gloss White Film
* Epson R280 with Red River Palo Duro Softgloss Rag
* Epson R280 with Red River Pecos River Gloss (60 lb)
* Epson R280 with Red River Polar Matte (60 lb)
* Canon Pro-100 with Inkpress Metallic Gloss
* Canon Pro-100 with Inkpress Metallic Satin
* Canon Pro-100 with Mitsubishi Pictorico Pro Hi-Gloss White Film

(Yes, I'm aware that Red River and some others recommend Epson 1400 profiles for the R280, but IME sometimes those work well and sometimes they don't. Yes, I'm aware that Inkpress recommends certain Canon paper profiles for its metallic papers on the Pro-100, but likewise I question whether custom profiles might not do better.)

Thanks!

[Fernando García]

Basically, I have heard something to the effect that the X-Rite software prints the first set of patches, then sits there waiting for you to measure them before it will do anything else. This could be a problem because my eyes suggest to me that there are visible dry-down changes until relatively long times after printing, and what I want to do is print the first set of patches, let them dry over night, then come back and measure them, then print the second set of patches, and let them dry for some hours, then measure them.

That used to be a problem with the Colormunki software, but with the i1Studio a new workflow was introduced in which you can save your progress at several steps of the profile creation process, so now you can print the target images and measure the patches later, after letting the print dry as long as you want.

[Mark D Segal]

Of course your basic problem here, and that of many others who have this software is that X-Rite never wrote a manual for it, and no-one else has. So there is a big void which X-Rite has never felt enough shame to rectify.

Anyhow, the answer to your question is two-fold:

(1) You don't need to use i1Profiler to print targets. You can save the targets out as TIFF files to your hard drive and print them correctly with the Adobe Color Print Utility downloadable for free from the Adobe labs website. Then you can measure the whole batch of them with your rented i1Profiler software.

(2) Within i1Profiler, you can do it all there, and save the Workflows within i1Profiler, which will save everything you've done up to the last step you've completed. Then you can recall those workflows from where you left-off and do whatever else remains to be done to complete the process. There is a Save Workflow command in the lower right corner of the interface. Make sure to give each paper you work on a separate name. Then all the individual workflows you've created will show-up in the "Assets" tab you access in the lower left corner of the i1Profiler interface.

[NAwlins_Contrarian]

Fernando, thanks, that reassures me greatly that renting the device will work for me.

Mark, maybe I have misunderstood the printer profiling process with this device. I understand that with many devices, and as is standard when you print targets and send them off to a service to have custom profiles built for you, you only print once, typically with a substantial number of patches (> 1000). However, I was / am under the impression that the i1Studio Spectrophotometer, like the ColorMunki Photo it (sort-of?) replaced, is designed for an iterative process: print a relatively smaller number of patches, read those with the spectrophotometer, and based on the reading of the first set, the software creates a customized second set of patches to print, which are then read in a second use of the device, the results of which the software then uses to create the profile. So the system relies on two sequential uses of the spectrophotometer with the printing of a customized target (based on the reading of the general target) in between. Is that incorrect? Thanks!

[Mark D Segal]

Ah - possibly. I overlooked which exact set-up you are using and had in mind the process with an i1Pro2 and i1Profiler. With that combination one prints targets with large numbers of patches just once. So Fernando's reply would be more relevant to your situation.

[unesco]

In case of your printers with dye inks, ColorMunki/i1Studio will give moderate results. Because of measurements on small number o patches it is not ideal for dye inks, since dye printers are not as linear as pigment ones. Iterative processing of Prism engine in that tool gives huge power with such a small number of patches but it assumes a lot of interpolation which is not perfect with dye inks which are far from ideal.
I have done number of profiles with it for Epson 1400 - they were quite ok, but incomparable to i1Photo Pro(2) results. This states to original Claria inks. Using e.g. L800 inks gave ugly profiles with ColorMunki/i1Studio. I have no experience with Canon dyes.

(1) You don't need to use i1Profiler to print targets. You can save the targets out as TIFF files to your hard drive and print them correctly with the Adobe Color Print Utility downloadable for free from the Adobe labs website. Then you can measure the whole batch of them with your rented i1Profiler software.

Mark, are you sure Adobe Color Print Utility works well with Canon printers? As far as I remember it is ok with Epsons, but gives bad results with Canons and Canon software is recommended instead. Just to be double-checked.

[Fernando García]

I was / am under the impression that the i1Studio Spectrophotometer, like the ColorMunki Photo it (sort-of?) replaced, is designed for an iterative process: print a relatively smaller number of patches, read those with the spectrophotometer, and based on the reading of the first set, the software creates a customized second set of patches to print, which are then read in a second use of the device, the results of which the software then uses to create the profile. So the system relies on two sequential uses of the spectrophotometer with the printing of a customized target (based on the reading of the general target) in between. Is that incorrect?

No, that´s correct.

[Mark D Segal]

Mark, are you sure Adobe Color Print Utility works well with Canon printers? As far as I remember it is ok with Epsons, but gives bad results with Canons and Canon software is recommended instead. Just to be double-checked.

As I've written in various of my reviews, ACPU works fine for Epson printers. For Canon printers it depends on the model. I had good experience using it for the Pro-1000, but not for the Pro-2000. Never figured out or received clarity on why that should be the case, but the safest approach with Canon printers is to print the targets either from i1Profiler or from Canon'r Print Studio Pro with Color Controls turned off.

[GWGill]

Because of measurements on small number o patches it is not ideal for dye inks, since dye printers are not as linear as pigment ones.

That's not what I would expect. Dye inks should be more linear, because the different dyes combine in a more straightforward way, much like colored filters. Pigment inks on the other hand have a more complex character, partly reflective, partly filtering, leading to more complex ink interactions.

Iterative processing of Prism engine in that tool gives huge power with such a small number of patches but it assumes a lot of interpolation which is not perfect with dye inks which are far from ideal. I have done number of profiles with it for Epson 1400 - they were quite ok, but incomparable to i1Photo Pro(2) results.

Note that this is a limitation of the software, not the instrument. Comparable quality profiles can be made with a ColorMunki/i1Studio to those using the i1Pro, it's just not as convenient to read the required number of patches.

[digitaldog]

I’ve seen no issues with dye inks and the 50 + 50 set of patches with the product but if I were the OP I’d rent a different solution FWIW.

[Mark D Segal]

I’ve seen no issues with dye inks and the 50 + 50 set of patches with the product but if I were the OP I’d rent a different solution FWIW.

This could make sense, but what would a viable alternative be? For example, would X-Rite's licensing allow an i1Pro2-i1Profiler solution be available for rent, and if so, would the O/P have a good time using it given the non-existence of a manual?

[digitaldog]

This could make sense, but what would a viable alternative be? For example, would X-Rite's licensing allow an i1Pro2-i1Profiler solution be available for rent, and if so, would the O/P have a good time using it given the non-existence of a manual?

Yes.

[Mark D Segal]

Interesting that one can rent this equipment from another license-holder for making one's own profiles, but if one owns the equipment, without a commercial license X-Rite allows us to make profiles for others only under very restrictive conditions. But I can see the distinction.

And I agree - looking back the O/P's obvious familiarity with these processes he'll probably be OK working through with the i1Pro2/i1Profiler set-up, though there is a learning curve using it in Advanced mode, for which the lack of a manual doesn't help matters.

Your suggestion would allow him to do all the printing in one operation and the readings in another, probably simplifying everything in his context.

[NAwlins_Contrarian]

Thanks everyone for the updates. To clarify / update some points:

I’ve seen no issues with dye inks and the 50 + 50 set of patches with the product but if I were the OP I’d rent a different solution FWIW.

I can rent an i1Studio Spectrophotometer from LensRentals for a week for $47 + shipping. If there is a better device available for rent in this price range, please suggest a device and rental house, I'm very interested in reasonable alternatives. For that matter, if the software for the i1Studio Spectrophotometer will let you print and measure more than 50 + 50 patches, I'm open to doing that too.

[L]ooking back the O/P's obvious familiarity with these processes he'll probably be OK working through with the i1Pro2/i1Profiler set-up, though there is a learning curve using it in Advanced mode, for which the lack of a manual doesn't help matters.

I've never made a printing profile before, but part of this is that I want to learn / experiment, and I've been reading a bit; and part is that I want profiles for several papers, but can't justify the expense of commercially-made custom profiles. FWIW I consider myself fairly technically-minded, have some background in engineering, and have been using a ColorMunki Display to calibrate and profile my monitors, in its software's 'advanced' mode. But I realize that these issues can get pretty deep, and I'm definitely a newbie in the area.

[W]ould X-Rite's licensing allow an i1Pro2-i1Profiler solution be available for rent?

Per LensRentals, X-Rite does not allow the rental of any of their software, but they allow anyone in lawful possession of one of their devices to download, install, and use the software. LensRentals would not send me a disc; I'd download it directly from X-Rite.

In the past I've had custom profiles for my little R280 and two papers made by Profiles by Rick, and they were an obvious step up from what I had used before: Epson 1400 profiles for the same papers. My plan to rent the i1 device and make a bunch of profiles is both much cheaper and presumably a learning experience. It will also give me the opportunity to compare the two single-step, 1728 patch custom profiles I have to ones I can make with this device, two steps, and 50 + 50 patches (or whatever the software will allow).

[Ethan Hansen]

I can't speak to how well the i1Studio will cope with your inkjets. Output from the Canon is better behaved than the R280; I'd expect this leads to better results from the tiny sample size the i1Studio software supports. Our experience with the older ColorMunki version showed profile behavior at or near dMax was more of a guess than reality.

A suggestion is to pair ArgyllCMS with the i1Studio instead of the bundled X-Rite software. As noted on the Argyll site, the i1Studio has an integrated UV filter. Any of your papers that contain optical brighteners may show undesirable behavior if viewed in an environment with moderate UV light (note that this isn't a software issue, it's due to hardware limitations). The ability to use a reasonable patch size and rebuild targets based on your actual printer behavior should tip the scales in Argyll's favor.

LensRentals found a creative loophole in X-Rite's licensing scheme. There is nothing in the EULA preventing one from renting out hardware. You cannot share software nor sell or distribute profiles. They are on better legal ground than the Profiles by Rick outfit who certainly appear to be in violation of X-Rite's licensing terms.

[Doug Gray]

I can't speak to how well the i1Studio will cope with your inkjets. Output from the Canon is better behaved than the R280; I'd expect this leads to better results from the tiny sample size the i1Studio software supports. Our experience with the older ColorMunki version showed profile behavior at or near dMax was more of a guess than reality.

A suggestion is to pair ArgyllCMS with the i1Studio instead of the bundled X-Rite software. As noted on the Argyll site, the i1Studio has an integrated UV filter. Any of your papers that contain optical brighteners may show undesirable behavior if viewed in an environment with moderate UV light (note that this isn't a software issue, it's due to hardware limitations). The ability to use a reasonable patch size and rebuild targets based on your actual printer behavior should tip the scales in Argyll's favor.

LensRentals found a creative loophole in X-Rite's licensing scheme. There is nothing in the EULA preventing one from renting out hardware. You cannot share software nor sell or distribute profiles. They are on better legal ground than the Profiles by Rick outfit who certainly appear to be in violation of X-Rite's licensing terms.

I1Studio only does M2 profiles but not because of an integrated uV filter. They both use "white" LEDs with no significant intrinsic uV.

M2 profiled prints on OBA papers exhibit strong color shifts when printed using Abs. Col. viewed in daylight or under significant uV. Same is true in the other direction when M0 or M1 profiled prints are viewed indoors under light containing little or no uV. This is increasingly a problem with white LEDs becoming so common. Normal consumer white LEDs don't have uV and most high CRI LEDs don't either. There are a few that produce significant uV below 400nm that are made for specialized purposes such as light booths meeting D50 uV level requirements.

However, this shift is far less and often completely negligible when printing using Rel. Col., Sat., or Perc. because they all map white to unprinted paper regardless of M0, M1, or M2.

How much color distortion occurs when viewing OBA laden prints made with M2 profiles under daylight depends on the way the inking interacts with uV. My pigment based printers produce nearly identical prints with M2 or M1 profiles when printed using Perc. or Rel. Col. using paper with a b* shift of -14.  OTOH, Abs. prints are grossly different.

[NAwlins_Contrarian]

Doug, if I understand you correctly, what you see as the main problems with the i1 device I proposed renting are that:
(1) it would build profiles that are bad for use with absolute colorimetric rendering intent; and
(2) it would build profiles that are bad for use under LED lighting.
Are those your main concerns?

If so, then I should be okay because I use relative colorimetric (about 75%) and perceptual (about 25%), and at home, to supplement regular window light I have stuck almost entirely with halogen (close to incandescent) bulbs for their variety of more pleasant characteristics. At work we have a mix of daylight (large windows but with tint) and florescent. I realize what is readily available on the market may change, but if I get reasonable use out of the profiles and some experience building them, it would be worthwhile to me.

Thanks!

[Doug Gray]

Doug, if I understand you correctly, what you see as the main problems with the i1 device I proposed renting are that:
(1) it would build profiles that are bad for use with absolute colorimetric rendering intent; and
(2) it would build profiles that are bad for use under LED lighting.
Are those your main concerns?

Au contraire. M2 profiles, which are generated by ColorMunki and Studio are, in fact, best for prints viewing with the kinds of illuminants typical of indoors. Even with high OBA papers and even with Abs. Col. where you are trying to reproduce things like a ColorChecker that will match reasonably closely an actual ColorChecker. The only problem is that if you view such a print outdoors or with a viewing station with standard uV levels of D50 then they will be way off.  But one has to choose. I rather like the better match with indoor illuminants that M2 provides.

If so, then I should be okay because I use relative colorimetric (about 75%) and perceptual (about 25%), and at home, to supplement regular window light I have stuck almost entirely with halogen (close to incandescent) bulbs for their variety of more pleasant characteristics. At work we have a mix of daylight (large windows but with tint) and florescent. I realize what is readily available on the market may change, but if I get reasonable use out of the profiles and some experience building them, it would be worthwhile to me.

Thanks!

But yes, there is typically little difference in prints made with Perc. or Rel. Col. between M0, M1, and M2 profiles. Even with very high levels of OBAs. There are some differences but minimal. And M2 profiles match the illuminants common indoors these days.

[Ethan Hansen]

Doug: Thanks for the correction on the i1Display. My brain obviously lacked sufficient caffeine.

If so, then I should be okay because I use relative colorimetric (about 75%) and perceptual (about 25%), and at home, to supplement regular window light I have stuck almost entirely with halogen (close to incandescent) bulbs for their variety of more pleasant characteristics. At work we have a mix of daylight (large windows but with tint) and florescent.

You should be fine at home and at work when using papers without significant amounts of brighteners. Fluorescent light at work may not play well with some of the glossy films you listed. At least with i1Profiler, there are some changes in how relative and perceptual render neutrals in UV-included vs. UV-excluded profiles on papers with high levels of brighteners. I'd suggest giving it a try and checking the same photo at home vs. at work.

If you feel ambitious, you should be able to hook the i1Studio to your laptop at work and measure the ambient light spectrum using Argyll. Those measurements can then be used to generate profiles tuned for that illumination rather than generic D50. Only go that route if prints look obviously off at work.

[Doug Gray]

Doug: Thanks for the correction on the i1Display. My brain obviously lacked sufficient caffeine.


You should be fine at home and at work when using papers without significant amounts of brighteners. Fluorescent light at work may not play well with some of the glossy films you listed. At least with i1Profiler, there are some changes in how relative and perceptual render neutrals in UV-included vs. UV-excluded profiles on papers with high levels of brighteners. I'd suggest giving it a try and checking the same photo at home vs. at work.

Likely that prints can appear quite off. The typical home LEDs and workplace fluorescents have very high and spikey spectra. I highly recommend making special profiles for these for any displayed prints you want to get the max effect from. Argyll and I1Profiler's full version lets you do this. Can't beat the price with Argyll though.

As for the tone curve issues in I1Profiler, there is a slider that controls this (Perceptual, Rel. Col. is well defined). IIRC, The tone curve adjustments determine how much the neutrals, relative to paper white, are tweeked with differences below/above L*=50. It's a function of the variance between the media white point and D50 white so there is a much greater effect on papers with OBAs between M0/1 and M2.

As for the general effect of Perceptual and Rel. Col. color shift from non-linear effects of inking on uV OBA interactions, this is really the determining factor on whether prints with OBAs will be impacted by using M0/1 or M2 profiles. It likely varies between ink and/or paper types. This can be easily quantified and I have done some ad hoc work on this with my printers and the results show little impact. Especially compared to the large impact from spectrally spikey illuminants. However, they are both pigment printers and a small sample at that.

I have searched for work testing this issue of the importance of the recent trend of supplying M2 only profiling equipment compared to the past practice of typically providing mostly M0 profiles but have found nothing. Seems to me this would be rather significant for critical printing. Has anyone seen anyone testing this issue?

I'm considering putting a measurement protocol together to determine if and when a printer/paper M2 profile can be used with minimal effect in an environment with significant (D50) levels of uV. It's a pretty simple process with the right tools but does require profiles made from the same patch set with uV and no uV readings.