X-Rite i1iO review
X-Rite i1iO profiling review
Automated patch reading for printer ICC profiling
Keith has been looking at the the X-Rite i1iO robot arm for the i1Pro and i1Pro 2.
Although mostly the same as the original one from our i1iO review in 2009, ours has been upgraded to fit both the original i1Pro and current i1Pro2 spectrophotometers.
The X-Rite i1iO automated scanning table
The i1iO first appeared in 2009 as the Gretag Macbeth Eye One iO, using the i1Pro spectrophotometer to measure profiling targets. Our device has been modified by X-Rite so it’s the same as the current i1iO and takes either the original i1Pro spectrophotometer or the newer i1Pro 2.
You’ll notice that they didn’t change the Gretag Macbeth logo – much of the rest of the device was upgraded, especially the clear plastic foot that holds the spectrophotometer.
I’ll give a quick overview of assembling and using the device in a basic mode with i1Profiler and then cover some of the refinement and measurement options available in the advanced mode.
2019: There is a new version of the i1iO to use with the new larger i1Pro 3 plus spectrophotometer. The two work together exactly the same way as shown here, but with the additional features of the i1Pro 3 Plus, such as M3 measurements (polarised + UV Cut). See my 1iPhoto Pro 3 Plus review for more details.
The i1iO hardware
I’m showing the i1iO here, working attached to my MacBook Pro. It’s shown on a table to both give a feel for the space you need and to save me tidying up my office desk enough for the photos…
The baseplate has attachment points for the arm assembly and a white measuring area. This white area has an electrostatic system that will hold sheets of paper steady, although anything with strong curl will benefit from using a bit of masking tape as well. The static holding is controlled by a button in the top RH corner of the base.
Note: The screens look a bit blue due to the tungsten lighting in the room (~2700K) not matching the whitepoint of the screens (~6500K)
There are three lugs that engage with the arm base, along with an electrical connection that will automatically lock when you attach the arm.
The metal base, showing the locking lever.
At the back are USB and power connections.
I’ve shown them connected here, but don’t power up the i1iO until everything is fully assembled.
Take care with the power connector, especially if trying to plug it in from the back without a clear view. You don’t want to bend/break pins or even damage the socket on the i1iO.
The arm has two rotation axes allowing the measurement of any patch in the marked area.
You can see the difference in design to the original (i1Pro only) fitting.
The back end of the i1Pro or i1Pro2 fits on the USB plug at the back.
The spectrophotometer is secured by moving the black tab on the end ring.
The i1iO has its own calibration tile (white circle)
The dial at the top of the arm’s base lets you adjust the flying height of the spectrophotometer.
Here, I’ve raised it to match the thickness of the cork coaster. Note how the tile holder rises too.
Now, I’ve got everything assembled and am ready to scan some profiling targets.
Using i1Profiler – Basic Mode
There are quite a lot of options in the i1Profiler software which is one reason the interface has ‘Basic’ and Advanced’ modes. I’ll start with the basic printer profiling process, to make an ICC profile for a paper. In this case I’m just using a cheap glossy photo paper for an RGB inkjet printer (see also my Epson P5000 review for more)
If you’re unfamiliar with i1Profiler, I’ve quite a lot more detail in my i1Photo Pro 2 review and my review of RGB printer profiling in i1Profiler. These include even more advanced features such as Optical Brightening Agent compensation (OBC). There are many other functions in the software, but for showing the i1iO I’m sticking to basic printer profiling.
For the basic mode, I’ll just pick printer profiling.
I’ll pick a ‘medium’ sized target for profiling.
The software has detected the i1iO 2 and offering me a number of measurement modes. These are whether the device measures individual patches at a time (spot), or scans across the whole chart in one movement, one row at time.
There is also a choice of whether to use the simple M0 measurement (such as I’d get using my old i1Pro) or the multiple datasets that the i1Pro2 gives me (M1,M1, M2, OBC)
The same target prints differently for different measurement modes, on one A4 sheet or two.
There is a help feature in the software, but I’m afraid it’s not comprehensive.
Much as I dislike saying it, X-Rite documentation has always been somewhat sparse and to my mind assumes the user already has a lot of knowledge. As I’ll show in the advanced section, there are multiple adjustments you can make to settings, but you’ll search in vain for details about just what they really mean, and -why- you might change them. X-Rite have got better over the years, but making videos is (IMHO) a poor substitute for proper reference documentation.
2019 V3 of i1Profiler now supports the i1Pro3 Plus, but I’m afraid still no manual…
You’ll need to print the target and let it dry thoroughly before measuring anything.
You can save targets and print them later, or use the print setting in i1Profiler.
Printing from i1Profiler means that colour management is correctly turned off in the printing process.
Make sure you select the right media setting for the paper you are using – profile quality is critically dependent on getting this right.
Once your prints are dry, you can measure them with the i1iO.
The process starts with calibration of the i1Pro using the white tile.
You then need to manually move the arm to tell it where the corners of the target are.
Note the little alignment cross – this needs to fit in the middle of the patch.
In this photo it’s not properly centered.
In dim lighting, this can be unclear, so I use a small LED torch to see it clearly.
Clicking the i1Pro button records the position and moves to the next corner.
Here’s the last corner showing an accurate position.
After the third location, the software may let you continue.
Take care with the positioning…
Once all is well, the measurement process will just happen…
This is the bit that always impresses people.
As the measurement progresses, the display shows the measurement data.
Once done, you can save the data for later use, and proceed to making a profile.
You’ll need to set the type of light source that the prints are to be viewed under.
Normally just select the default D50 setting.
You can measure lighting and use the measurement sets for making custom print profiles, but I’d suggest it will take quite a bit of experimentation to see what makes a real difference, if at all. I’ve used it a few time to counteract some nasty fluorescent lighting, but if anyone ever ‘fixes’ the lighting, my custom profile is useless and may make the print look worse.
Next, give your profile a meaningful name, and let the software create it for you.
Just for good measure, there is a 3D plot of some of the profile data.
Mostly a pretty picture, but spikes or holes in the shape should give a strong hint that something has been amiss in your profiling workflow…
I’ll show a few of the more advanced features available when I’m using the i1iO with i1Profiler – some are also available if using the spectrophotometer manually – or with a reader like the i1iSis [review]
I used the ‘medium’ sized patch set for profiling.
You can make your own custom patch sets for profiling targets, although there are diminishing returns once you get over a thousand, for good quality printers and media. It takes longer to measure and you’ve more sheets to worry about.
With an i1iSis I can read some 2800 patches on a single A3+ (13″ x 19″) sheet which can give some nice profiles, but is more than I’d probably use with the i1iO.
Once I’ve made a profile, I can get the software to analyse it and come up with a second set of patches to refine or optimise a profile I’ve made with i1Profiler.
The source for the new patch set can be automatically generated and can include spot colours or even from an image.
The new patches are assembled into a target. A few hundred is often enough to give an improvement, especially if you’re doing this to cope with awkward colours.
The patch order can be scrambled, so as to make the target easier to read in scanning mode, where there needs to be a significant change between each patch.
There are a number of additional options for the measurement process – some of which just add extra data into the profile whilst others may change measurement/profiling behaviour.
I have to say ‘may’ since there simply isn’t any true reference documentation I’ve ever seen.
Most items are relatively clear though, and one I do appreciate is the ability to average multiple measurements for each patch.
Unfortunately there is no information on what sort of averaging is happening, but it seemed to improve profiling quality for some rough matte art papers I looked at.
Measured data can be saved for analysis, although individual patches can be selected for viewing.
Take this first patch – 255.255.255 or paper white [click to enlarge]
The graph at the side shows the effect of OBAs in the paper, as viewed using different measurement conditions.
The green line shows how measuring without UV doesn’t show the peak from the brightening agent.
Another patch shows a distinct purple shift, which wasn’t noticeable in a test print – remember that any data shown needs context to be meaningful.
There are quite a few options when making your profiles.
For my usual profiles, I select the ‘optimise quality’ setting.
This makes large 16 bit profiles.
But what about the settings for the Perceptual rendering intent? What’s called ‘Custom’ here is actually the default for me.
“Colorful” reflects settings you’d have got from earlier profiling software.
Saturation seems to be for more colourful images where you want to keep detail.
Both feel a bit too ‘strong’ for photo printing IMHO
But hold on a bit – what do those numbers actually mean? After several years of making profiles I have to say that I simply don’t know for sure.
I make profiles at 0/0/0 for photo papers, with the saturation setting going up to 20 or so for papers with smaller gamuts (matt fine art types). It’s worth noting that increasing contrast or saturation up from zero runs the risk of losing detail in darker areas, so you need to experiment.
This gives me a profile with a Relative Colorimetric rendering intent (which is created untouched by these settings) and a Perceptual rendering intent that seems to give a bit more punch to some prints (but too much for others).
You can see a bit of this in an article when I looked at the printing of ‘strong colours’.
I’ve created a specific black and white test image to help with B/W printing.
This includes some grey patches, set up so as to be readable with the i1iO in spot reading mode (allows for averaging)
If this is interest, I’ve an article describing the process (and how to use it) in some detail. Note that with a slightly different target design it’s quite possible to use an i1Pro or i1Pro 2 manually for this.
The device works well, after almost a decade of use. I’d note that quite a bit of it was upgraded to work with the i1Pro 2, but I’ve never had any mechanical or electrical issues.
It’s not a big robust bit of machinery, so needs to used with some care if accuracy and reliability are to be maintained. You won’t have had to make many profiles ‘by hand’ to appreciate the automation.
The i1Profiler software works well, although it is rather lacking in documentation if you are new to profiling or want to know more detail.
The cost makes this a device more for commercial profiling than occasional use by photographers, but watching it work still impresses me ten years after first seeing it…
Never miss a new article or review - Sign up for our Newsletter (2-4 a month max.)
Enjoyed this article?
All the latest articles/reviews and photo news items appear on Keith's Photo blog
We've a whole section of the site devoted to Digital Black and White photography and printing. It covers all of Keith's specialist articles and reviews.
Articles below by Keith (Google's picks for matching this page)