i1Profiler can't optimize PM5 generated targets???

[smilem]

Why i1Profiler can't optimize PM5 generated targets is beyond me, it gets bad gradients.

See for yourself, notice the artifacts. Use attached image to simulate your ICC profile optimization in photoshop softproof mode.

PM5 target optimized in i1Profiler Relative:


PM5 target optimized in i1Profiler Perceptual:

[smilem]

So nobody tried this?

[digitaldog]

Why i1Profiler can't optimize PM5 generated targets is beyond me, it gets bad gradients.

I don’t understand what you are talking about. Optimize a target built in PM5?

You should be generating new targets with the newer (better) target generator.

[smilem]

I don’t understand what you are talking about. Optimize a target built in PM5?

Yes I have many targets built using PM5 (optimized to fit into certain paper sizes, use max patch count etc.)

You should be generating new targets with the newer (better) target generator.

Like what? i1Profiler and colorport can't seem to generate small targets.
See here: http://www.luminous-landscape.com/forum/index.php?topic=58092.0
X-rite seems like a company wanting to remove current PM5 userbase, by making unavailable certain features, not supporting to save to CGATS file formats etc. That is absurd.
Oh did I mention OBA in i1Profiler not available? I'm lucky to have UV-cut i1Pro

[digitaldog]

Like what? i1Profiler and colorport can't seem to generate small targets.

Build a target in i1P with the patch count you want (don’t worry about the page size). The newer color patches for the number of patches selected is superior to the same number specified in ProfileMaker Pro. Then in PMP’s MeasureTool, in Chart Generator, select Load Reference and load the file saved out of the Chart Workflow area. MT will load those colors. Now create a page size you wish but with the newer patch generation. Measure in MT if you want, save out CGATs and build profile in i1P (it CAN do this).

Oh did I mention OBA in i1Profiler not available?

Can you prove by building two profiles in both packages that PMP did anything useful and that i1P needs this (or doesn’t have it?).

I'm lucky to have UV-cut i1Pro

Only in cases where you want the unit to put its hands over its eyes and ignore (filter) UV which in many cases isn’t desirable. Sometimes you do want that data.

[smilem]

select Load Reference and load the file saved out of the Chart Workflow area

The MT can't open cxf or pxf files saved in i1Profiler "Patch set" what do you mean "Chart Workflow area"?

Can you prove by building two profiles in both packages that PMP did anything useful and that i1P needs this (or doesn’t have it?).

No since I have UV-cut version. I only read about many users "on this forum too" that are frustrated by this lacking feature.

Sometimes you do want that data.

Any examples please?

[digitaldog]

The MT can't open cxf or pxf files saved in i1Profiler "Patch set" what do you mean "Chart Workflow area"?

Go into the Test Chart workflow (2nd button) AFTER creating the patch set. Save that from Page Data button (its a txt file). MT can open that as a reference and load those colors. Then you use MT to configure the test chart as you wish with those new colors. Save out a new reference and TIFF from there. Measure there. Save out CGATs from MT to load into i1P to build a profile.

[digitaldog]

Any examples please?

The best approach is to measure cut and non cut and build a profile each way and test. I’ve done this many times, most recently with a digital press who’s paper had a large amount of OBAs (bStar of -7!). The only way to know what will work best is to have both sets of data and test on a case by case basis from the resulting profiles built from each set of data. But having an iSis, that’s super easy for me to do.

[smilem]

Go into the Test Chart workflow (2nd button) AFTER creating the patch set. Save that from Page Data button (its a txt file). MT can open that as a reference and load those colors. Then you use MT to configure the test chart as you wish with those new colors. Save out a new reference and TIFF from there. Measure there. Save out CGATs from MT to load into i1P to build a profile.

Second button lets me save a txf file, and that is not text file, more like xml file. And I can't load that into MT.

[smilem]

The best approach is to measure cut and non cut and build a profile each way and test. I’ve done this many times, most recently with a digital press who’s paper had a large amount of OBAs (bStar of -7!). The only way to know what will work best is to have both sets of data and test on a case by case basis from the resulting profiles built from each set of data. But having an iSis, that’s super easy for me to do.

It is nice to have both, but living in europe means double the price of i1isis, would you still buy it??? When I consider it can read only thin paper like targets I can't justify the price.

[digitaldog]

Second button lets me save a txf file, and that is not text file, more like xml file. And I can't load that into MT.

Wrong button. The txt file I save loads fine in MT as a reference.

[smilem]

Wrong button. The txt file I save loads fine in MT as a reference.

See attached file, what version are you using in v1.1.1 this is what I see:

[digitaldog]

On 1.1.1 on Mac. Click the button Save under Page Data. There is no extension shown on my end in the dialog. TWO files are saved to disk, a txt file and a txf file. You have to load the txt file in MT.

[smilem]

On 1.1.1 on Mac. Click the button Save under Page Data. There is no extension shown on my end in the dialog. TWO files are saved to disk, a txt file and a txf file. You have to load the txt file in MT.

Seems I found a reproducible bug, when I save to thumb drive there is only txf file saved, when I save on folder located on my HDD two files are saved. The file opens fine in MT, thank you.

[smilem]

Colorport has new algorithm like i1profiler that re-reads hard to read rows slower or by patch mode. But when I import my patches into colorport the target doesn't fit onto single page with margins set to zero. It fits inside 1 sheet in Measuretool.

This is because colorport add additional info on the side of the page, is there way to remove it by editing config files etc.
I would like to make and measure my targets with colorport and then generate profiles with i1profiler, all this trouble because i1profiler makes targets with smaller white gap than black gap.

Perhaps the i1profiler i1iO target gap is adjustable in config files?
Does anyone know if latest PM5.0.10 has this re-reads hard to read rows slower or by patch mode thing?

From Xrite site colorport page "When measuring with the i1iO, measurements will automatically switch to spot mode if the instrument fails to detect the patches correctly while reading in strip mode. This will improve measurement results for low-resolution output and/or highly textured substrates."

[tony22]

The best approach is to measure cut and non cut and build a profile each way and test. I’ve done this many times, most recently with a digital press who’s paper had a large amount of OBAs (bStar of -7!). The only way to know what will work best is to have both sets of data and test on a case by case basis from the resulting profiles built from each set of data. But having an iSis, that’s super easy for me to do.

Andrew, I know you discussed this a number of times, but in this case of your recent comparison with this high OBA paper I'd love to hear what happened. Especially for Perceptual and Relative.

[digitaldog]

Andrew, I know you discussed this a number of times, but in this case of your recent comparison with this high OBA paper I'd love to hear what happened. Especially for Perceptual and Relative.

Simple, we preferred the output from a number of test images with the non cut data (Perceptual).

[smilem]

OK here is my workflow that I'm testing.

I generated target using i1Profiler, export as cgats file.
I loaded the target in Colorport and generated tiff file to my taste.
I measured the target in Colorport two times saving cgats file.
I averaged the target in PM5 Measuretool.

Now the problem:

I tried to build the ICC profile using i1Profiler.
Loaded the averaged cgats file as source "patch set"
Loaded the averaged cgats file as measurement file.

But I noticed the "test chart" button is not correct, the patches seem not the right color, generating icc gives error.
Where can I get the correct "test chart" file for i1Profiler? I could save directly to xrga format but I need averaged data, PM5 can only open cgats.

[digitaldog]

I generated target using i1Profiler, export as cgats file.
I loaded the target in Colorport and generated tiff file to my taste.

You could have saved out the TIFF from i1P...

I measured the target in Colorport two times saving cgats file.

And you could have measured the target in i1P...

I averaged the target in PM5 Measuretool.

Good.

I tried to build the ICC profile using i1Profiler.

Load what profile?

Loaded the averaged cgats file as source "patch set"

I think you need to load the reference (not the measured data).

[smilem]

You could have saved out the TIFF from i1P...

Yes but I need to have even color separators, and I can't seem to hack i1profiler

I think you need to load the reference (not the measured data).

Yes I loaded the reference and then measurement in i1Profiler and it worked. Somehow it doesn't matter for PM5, so I only was loading the measurement data, as it suppose to have both.

It seems reading in colorport then saving to profilemaker5 xrga give larger gamut and lower neutral rendering values compared to saving to non-xrga format using colorport.

Will compare this with PM5 and i1Profiler.