i1 Profiler worse targets for i1iO ?

[smilem]

HI, I noticed that i1Profiler makes worse targets for i1iO than CP or PM5. The problem is that i1Profiler makes white gap lines more thin than other packages, this results in harder to read targets. Now how does making thinner lines improve patch readability Considering that CP (colorport) is an X-rite software and makes normal PM5 like targets.

See attached image

[VitOne]

I am not an i1iO user but I have no problem in using ColorPort with i1Profiler. I am using the Spectrolino + SpectroScan combo with i1Profiler and because the hardware is not directly supported in i1Profiler I use ColorPort to import patches, print them  and read them with my hardware. Then I go back and load the readings in i1Profiler. So what I can suggest you to do if you are more happy with ColorPort is just to continue to use it, it requires just 2 “drag and drop” operation with the mouse to interact with i1Profiler. I am much more happy with the Spectrolino reading capabilities than I am with the i1Pro REV D (that I am using only as a dongle to unlock the software).

Anyway if you have problems with the i1iO maybe you could email X-Rite, your hardware should work without problems with i1Profiler, maybe they can help you to make a better target.

[smilem]

What is your Spectrolino + SpectroScan combo

Could you share your target comparison, 1st scan vs. 2nd scan of the same target?

Measuretool 5 target comparing:

De2000

Worse 10% - ?
maximum Total - ?

Thanks.

[VitOne]

Spectrolino + SpectroScan is an hardware bundle that can be used to read color patches; a Google search can easily tell you more.

I don’t have the data that you request now. Maybe I can do some tests in the next days. I am sure that you can this information searching also in previous post in this forum. If you are interested in the Spectrolino you could maybe find some info here:

http://www.luminous-landscape.com/forum/index.php?topic=43199.0

Anyway I am not really sure that the DeltaE 2000 between different reading is a complete information. It would be more interesting for me to compare results with a pro-grade spectrophotometer, for example.

[smilem]

I know what Spectrolino + SpectroScan is, so you say it's very accurate, then please provide info I asked for. Don't tell me you know it's accurate and haven't made a measurement data comparisons in profilemaker?

[VitOne]

I used ColorLab 2.8.13 and 2 measurements I made for a profile I was building today from 2132 patches (2 A4 sheets). Comparing them this is what I get (all values in DeltaE 2000):


I don’t think that I have ever said that the Spectrolino is very accurate. Anyway if I said so I am sorry, I didn’t mean that. As I told you I am not interested in DeltaE 2000 average between different readings. I don’t like the Spectrolino for his repeatability, I care more about his accuracy in readings, including the “dark” patches, where I had some poor results with other instruments. I know from previous experience that a color patch printed in different area of the sheet can give me a different reading. And I also know that a Spectrophotometer that is not re-certified can give very bad results. This is for example the report of a re-certification I have recently done for the Spectrolino:



A video for color management “idiots” from a color management “idiot” (Spectrolino in action): http://www.youtube.com/watch?v=ghFviIQbRQE

I have been told that i1Pros can have a max DeltaE of more than 2.5. I have no proof of this, but users that sent their instrument recently maybe can give you some more detailed information.

Do you really think that a comparison between 2 measurements will tell you something about the accuracy of an instrument? I learned that repeatability and accuracy are two very different things. Both important, but repeatability alone doesn’t show much.
I am personally more happy with the results that I get from this instrument combo than I was with other hardware I own and tested. I don’t really need to show or prove anything, this are just my 2 cents, I don’t want to convince anybody. I spoke with some “color gurus” and color management company, including X-Rite email support, and they all confirmed to me that the Spectrolino has a better “reading head” (I don’t know if this is correct, I am sorry but my English is very weak) than the i1Pro and somebody also told me that the Spectrolino is better than the i1iSis…

I am not a “number maniac” and I mainly built profile for myself and test them with prints, not with numbers. I like to have an instrument with a good repeatability, but I learned that this is something that cares if you are sure that your instruments is accurate. Have you ever tried to print more than once a target and read it? The printers I used didn’t behave in the same exact way each time I printed (and I have a “controlled” working space) so it is not very uncommon to get some discrepancies in targets that you print in different moments, with a different roll of paper, with a different set of inks.

I am not saying that repeatability is not an important thing to care about, I am saying that there are many others thing that I worry about when I make profiles, and that the best profile doesn’t necessarily come from different measurements with low DeltaE between them. There are cases where you want to have some DeltaE between the reading, maybe because you are using a “non-perfectly consistent” paper and you want to average its behavior.

[smilem]

Do you really think that a comparison between 2 measurements will tell you something about the accuracy of an instrument? I learned that repeatability and accuracy are two very different things. Both important, but repeatability alone doesn’t show much.

Well I have separate BRCA tile and I checked what reading I got when the i1Pro was new, and I check it periodically, the reading doesn't change - means it is still accurate.

Have you ever tried to print more than once a target and read it?

Yes, intact I now print same target on 2 sheets of paper and average them, this gives better results then averaging measurements from same target measured 2 times.

Where did you re-calibrate , how much did it cost?
I'm afraid not to pay the calibration cost but the fact that shipping is very hard on the instrument no matter how you pack it, and shipping to Switzerland is not a lightly trip.

[smilem]

Well here is my target comparison

[VitOne]

Spectrolino recertification was not cheap, I had to pay about 350 euro including shipping and taxes (to X-Rite).

[digitaldog]

Could you share your target comparison, 1st scan vs. 2nd scan of the same target?

Measuretool 5 target comparing:

De2000

Worse 10% - ?
maximum Total - ?

Sorry if this is obvious but not yet stated. The reports of course are all based on the number of patches you compare. The average, best, worst 10% etc are all going to be different if you feed the software 500 or 5000 patches. For example here’s a report from ColorThink on two measurements off an iSis:


--------------------------------------------------

dE Report

Number of Samples: 567

Delta-E Formula dE2000

Overall - (567 colors)
--------------------------------------------------
Average dE:   0.22
    Max dE:   0.52
    Min dE:   0.02
 StdDev dE:   0.09

Best 90% - (509 colors)
--------------------------------------------------
Average dE:   0.20
    Max dE:   0.33
    Min dE:   0.02
 StdDev dE:   0.07

Worst 10% - (58 colors)
--------------------------------------------------
Average dE:   0.38
    Max dE:   0.52
    Min dE:   0.33
 StdDev dE:   0.04

--------------------------------------------------
--------------------------------------------------

[smilem]

If accuaracy is the upmost goal, then with i1iO I scan in patch mode, that provides superb accuracy. I also should say that right target is very important.

Why X-rite makes non ISO 5-4:2009 compliant targets is beyond me, and then you want to read accurately forget it. ISO 5-4:2009 target needs to me made depending what your instrument aperture is, the smaller the smaller patches you need.

The default X-rite patches for i1io of 6.4x6.4mm is a joke ! They have been pretty good ignoring my support requests and questions about this issue and similar issues where it is clear they should stand behind their products.

Why TC9.18 RGB industry standard target made for i1iO is with 6.4x6.4 patch size Who said it can be read fine with such patch size? X-rite? After merging with Gretagmacbeth they haven't made singe software or hardwire solution worthy of the fact that two greatest (or should I say biggest) companies name.

[VitOne]

Looking in the i1Pro data I found this (values in DeltaE 94):

Concordanza interstrumentale: 0.4 DE*94 media, 1.0 DE*94 massimo (deviazione dal standard di fabbricazione X-Rite a 23 °C per la modalità misurazione singola su 12 ceramiche BCRA (D50, 2°)
I try to translate in English: Concordance between instruments: medium 0.4 DeltaE, max 1 DeltaE.

The instrument that we are discussing about has a very good repeatability (Ripetibilità a breve termine: <= 0.1 DE*94 (D50, 2°), rispetto al valore medio CIELab di 10 misurazioni ogni 3 secondi su bianco). This is good, but I would not say that it is very accurate. The average DeltaE 94 error of 0.4 DeltaE is something to think about.

I would also think about what paper and printer you are using: as I told you, if you try to print the same target in different days, on a different sheet of paper, with a new set of inks (and this are not all the variables that can make a difference) you will easily find that for almost all the paper/inks/settings the DeltaE between the measurements will be not so small.
Again: a repeatable instrument is a good start, but I would care about many other variables when making printer profiles. As I told you I am not trying to convince anybody with my choice of the Spectrolino, but you will never convince me that the i1Pro (also if used with the i1iO table) is a better instrument. A few reasons: with the Spectrolino I get profiles that are more accurate in the shadow details and I can use the polarization filter and the UV filter.

Also making the hypothesis that your instrument is perfect and has no deviation (and this is not true, because from the data I have the i1Pro REV D has a medium error of 0.4DE 94 as I reported) the paper and the ink that you are using will never be perfect. Anyway I really don’t find a real world application where an accuracy of less than 0.9 DeltaE average is needed. If you need to reproduce a particular color or you want a profile to be optimized for a particular patch there are several ways do this. But we should know what is the goal that you want to reach. Otherwise, for general purposes profile, there are, as I said, many thing that I would care about, accuracy and repeatability of my instrument are on the list, but they are not alone.

Before getting “mad” with “perfect measurements” I would care more (just to make some examples) about: lighting conditions of the area where the print will be exposed, storage “box” used to store the prints, type of glass and frame, “right” management of out of gamut color, management of gloss differential and bronzing (if needed), “pleasantness” of the profile in a real-world print (accuracy of the gradients, shadow details), etc...

[VitOne]

Here you can find a link with specifications in English: http://www.northlight-images.co.uk/reviews/profiling/i1-lt.html

[smilem]

As I told you I am not trying to convince anybody with my choice of the Spectrolino, but you will never convince me that the i1Pro (also if used with the i1iO table) is a better instrument. A few reasons: with the Spectrolino I get profiles that are more accurate in the shadow details and I can use the polarization filter and the UV filter.

Yes the Spectrolino is better and can scan transmission with (T table), but it is not longer manufactured and as I read no longer can be calibrated after 2012. The removable filters are nice, I always wanted to have those on my i1iO

The i1Isis is no better than i1Pro : Average 0.4 DE*94 (deviation from X-Rite manufacturing standard at a temperature of 23°C on 12 BCRA tiles (D50, 2°))
They did not wire max value so I assume it's the same as i1Pro, after all they say it is the same technology.

You can't scan super thin film like, thick printable DVD's anyone with i1isis, so why get i1isis when you can get i1iO with i1Pro and profile your monitor too.

[smilem]

Looking in the i1Pro data I found this (values in DeltaE 94):

Concordanza interstrumentale: 0.4 DE*94 media, 1.0 DE*94 massimo (deviazione dal standard di fabbricazione X-Rite a 23 °C per la modalità misurazione singola su 12 ceramiche BCRA (D50, 2°)
I try to translate in English: Concordance between instruments: medium 0.4 DeltaE, max 1 DeltaE.

The instrument that we are discussing about has a very good repeatability (Ripetibilità a breve termine: <= 0.1 DE*94 (D50, 2°), rispetto al valore medio CIELab di 10 misurazioni ogni 3 secondi su bianco). This is good, but I would not say that it is very accurate. The average DeltaE 94 error of 0.4 DeltaE is something to think about.

I would also think about what paper and printer you are using: as I told you, if you try to print the same target in different days, on a different sheet of paper, with a new set of inks (and this are not all the variables that can make a difference) you will easily find that for almost all the paper/inks/settings the DeltaE between the measurements will be not so small.
Again: a repeatable instrument is a good start, but I would care about many other variables when making printer profiles. As I told you I am not trying to convince anybody with my choice of the Spectrolino, but you will never convince me that the i1Pro (also if used with the i1iO table) is a better instrument. A few reasons: with the Spectrolino I get profiles that are more accurate in the shadow details and I can use the polarization filter and the UV filter.

Also making the hypothesis that your instrument is perfect and has no deviation (and this is not true, because from the data I have the i1Pro REV D has a medium error of 0.4DE 94 as I reported) the paper and the ink that you are using will never be perfect. Anyway I really don’t find a real world application where an accuracy of less than 0.9 DeltaE average is needed. If you need to reproduce a particular color or you want a profile to be optimized for a particular patch there are several ways do this. But we should know what is the goal that you want to reach. Otherwise, for general purposes profile, there are, as I said, many thing that I would care about, accuracy and repeatability of my instrument are on the list, but they are not alone.

Before getting “mad” with “perfect measurements” I would care more (just to make some examples) about: lighting conditions of the area where the print will be exposed, storage “box” used to store the prints, type of glass and frame, “right” management of out of gamut color, management of gloss differential and bronzing (if needed), “pleasantness” of the profile in a real-world print (accuracy of the gradients, shadow details), etc...

Hi, could you send me first page of the report you posted?
I have received the re-certified device and I wonder if it has been certified by the same laboratory.

Thank you.