Pages: [1]   Go Down

Author Topic: Does i1Profiler print targets with all CM turned off? Like ACPU?  (Read 5384 times)

l_d_allan

  • Full Member
  • ***
  • Offline Offline
  • Posts: 208
    • Berean photos
Does i1Profiler print targets with all CM turned off? Like ACPU?
« on: February 25, 2013, 09:15:05 pm »

I've got the i1Profiler 1.4.2 demo, and trying it out to:
  • create and print targets for iSis
  • to be measured by i1Profiler or PMP5
  • and printer profiles generated by PMP5
  • goofy, non-standard work-flow due to MeasureTool's Test-Chart-Generator not licensed to Save  ... if intererested, see:
  • http://www.luminous-landscape.com/forum/index.php?topic=75630.0
  • also attempting to use ColorPort + PMP5 to make profiles, which may be a better approach

My speculation is that when i1Profiler is used to make a target, it is very much equivalent to using ACPU on the same target "Saved" by i1Profiler. In both  cases, there is "No Color Management" done (as long as the printer-driver settings are appropriate ... see attachment).

Is that correct?

Using an iSis that has very good repeatability, should the measured Lab values of the printed patches  from ACPU and i1Profiler be very, very close to each other (within measurement variability)? If MeasureTool's Compare capability is used, should the same mid-gray patch in the two printed targets be within De2k 0.3 or better?



Logged
retired in Colorado Springs, CO, USA ... hobby'ist with mostly Canon gear

bill t.

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3011
    • http://www.unit16.net
Re: Does i1Profiler print targets with all CM turned off? Like ACPU?
« Reply #1 on: February 26, 2013, 12:20:25 am »

If I understand you correctly, then yes, i1Profiler prints non-color managed targets without help from any other software such as ACPU.

But I believe it will only offer to print targets for whatever supported calibration device is actually plugged and powered up, and for which there is a driver installed on the system.  So I doubt if you can print targets for arbitrary measurement devices to be used on other systems.

Edit...i1Profiler will save targets as tif files, again for whatever device is plugged in.  So if you want to print targets on your system but measure them on a remote system, perhaps somebody could send you a target file in tif format.
« Last Edit: February 26, 2013, 12:25:30 am by bill t. »
Logged

l_d_allan

  • Full Member
  • ***
  • Offline Offline
  • Posts: 208
    • Berean photos
1Profiler demo can print targets for unattached devices
« Reply #2 on: February 26, 2013, 04:32:26 am »

I've had no problems getting the i1Profiler 1.4.2 demo to print test-charts for any device that shows up in the combo-box dropdown, even if they weren't attached.

Also, I've got an iSis,  and iProfiler demo can print test-charts even when it isn't turned on. I don't have a Pro, Pro-2, i1iO, or iSis-XL, but can easily make prints for them.

i1Profiler is inclined to "flip back" to the iSis after defining the number of patches in the Patch-Set, but it is easy to re-select another device, and then make a printed test-chart for the selected device.

A person in a photography Meetup group that I belong to has a Pro-2, and was willing to let me watch him make a profile from a 702 patch test-chart. I printed the 702 test-chart out, let it dry, and was hoping to do that, but we got snowed out.

I was able to "spoof" having an iSis-XL, define a larger paper size, and print a 1505 patch test-chart. As described in another post:
http://www.luminous-landscape.com/forum/index.php?topic=72184.msg605020#msg605020
I was able to use CS4 + Edit + ImageSize to have ACPU print it on letter-size 8.5x11" paper. However, my iSis standard couldn't reliably read all rows. That was possibly because the patches ended up about 5.5 x 5.5mm in size on the ACPU printed test-chart.

See attached

« Last Edit: February 26, 2013, 04:55:18 am by l_d_allan »
Logged
retired in Colorado Springs, CO, USA ... hobby'ist with mostly Canon gear

Rhossydd

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3369
    • http://www.paulholman.com
Re: 1Profiler demo can print targets for unattached devices
« Reply #3 on: March 09, 2013, 05:12:13 am »

I was able to use CS4 + Edit + ImageSize to have ACPU print it on letter-size 8.5x11" paper. However, my iSis standard couldn't reliably read all rows. That was possibly because the patches ended up about 5.5 x 5.5mm in size on the ACPU printed test-chart.
Not sure if you've resolved your issues with this but....

You could either print via CS4 using the 'no colour management' option to avoid ACPU resizing the chart.

Or you probably need to edit the reference data in ColorPort so that the iSis expects the resized chart correctly. (you might need to also change the name and use that to measure the chart, leaving teh original data for creating the image data).
Logged

l_d_allan

  • Full Member
  • ***
  • Offline Offline
  • Posts: 208
    • Berean photos
Re: 1Profiler demo can print targets for unattached devices
« Reply #4 on: March 12, 2013, 04:47:53 pm »

Not sure if you've resolved your issues with this but....

I'm really struggling with the iSis, and set it aside for the last week or two ... starting to read Digital-Dog's "Color Management for Photographers". Lots of Bruce Fraser's "Real World Color Management" was over-my-head.

So far, I can get the iSis to reliably read targets that include BarCodes ... sometimes ... but the profiles tend to be randomly flawed. Without the BarCode, the patch alignment seems to get off as the scan-head moves to the right.  The fields in iSis_INFO seem ill-behaved as far as defining widths and heights. grrr

The applicable fields in the ColorPort .xml files (thickness and spacing) seem to be ignored.

I have gotten decent at using CS4/5 Image + ImageSize to get MyTarget.tif to fill the page symetrically for ACPU  From another post, I believe you are wrestling with ACPU also.
Logged
retired in Colorado Springs, CO, USA ... hobby'ist with mostly Canon gear

Rhossydd

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3369
    • http://www.paulholman.com
Re: 1Profiler demo can print targets for unattached devices
« Reply #5 on: March 12, 2013, 06:49:35 pm »

but the profiles tend to be randomly flawed.
It might help other watchers of this thread help you if you can specify what actual problems you're seeing with the profiles generated, rather than just saying 'flawed'.
Logged

l_d_allan

  • Full Member
  • ***
  • Offline Offline
  • Posts: 208
    • Berean photos
Re: Does i1Profiler print targets with all CM turned off? Like ACPU?
« Reply #6 on: March 13, 2013, 02:13:10 am »

Good advice.

I'm getting ready to put some screen snap-shots together to illustrate. Briefly, some printer profiles seem to come out fine for Perceptual intent, but White areas are light to medium gray with the same profile when Relative Colorimetric is selected. Pretty much all prints made from profiles that use Relative have white being quite gray. This shows up in Print-Preview from CS4 and CS5 (and LR4) when I click on "Match Print Colors".

Those are the usable profiles, at least for Perceptual Intent. Another profile made with very similar "work flow" will have both Perceptual and Relative with what should be pure white being quite gray. But other colors aren't nearly as much gray ... ie. darker, so it isn't that the print-preview in CS5 is simply darker. I'm very puzzled. I haven't figured out how to distill my observations to coherent questions yet.

I'm clearly doing something wrong, and trying to get some consistency and understanding on what causes what.
Logged
retired in Colorado Springs, CO, USA ... hobby'ist with mostly Canon gear

Wayne Fox

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4237
    • waynefox.com
Re: Does i1Profiler print targets with all CM turned off? Like ACPU?
« Reply #7 on: March 13, 2013, 08:02:37 am »

those sound like some odd issues. I have no problem with iSis, getting great results.  I have a perfect  match between a 4900 and a 9900, and my Epson luster is now so close to my exhibition fiber prints I can use the cheaper luster for testing on the 4900, then print the larger prints on EEF on the 9900.  I  did 4000 patch targets, easy to print and read.

No bar codes, but I did save the settings when I made the target so I could be sure to get the identical settings the next day when I read them.  (I always wait at least 24 hours before reading a target from an inkjet printer).

I've also built profiles for a ZBE chromira and Immetto silver halide/chemical printers that have worked great.

Perhaps the iSis has a problem?
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: Does i1Profiler print targets with all CM turned off? Like ACPU?
« Reply #8 on: March 13, 2013, 10:52:21 am »

Briefly, some printer profiles seem to come out fine for Perceptual intent, but White areas are light to medium gray with the same profile when Relative Colorimetric is selected. Pretty much all prints made from profiles that use Relative have white being quite gray. This shows up in Print-Preview from CS4 and CS5 (and LR4) when I click on "Match Print Colors"

Either a Version 4 profile and a bug in the software or a bad series of readings.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

l_d_allan

  • Full Member
  • ***
  • Offline Offline
  • Posts: 208
    • Berean photos
Re: Does i1Profiler print targets with all CM turned off? Like ACPU?
« Reply #9 on: March 14, 2013, 10:49:35 am »

Those sound like some odd issues. Perhaps the iSis has a problem?

I sure hope not ... but I'm much more inclined to suspect the CM newbie in the mirror is doing something wrong.

The n.i.b. iSis passes the i1Diagnostics fine, and seems to get reasonable values on pure white patches and pure black patches. I looked at the iSis' bright White calibration tile, and it was fine, then carefully cleaned it. Same with the white strip under where the scan-head glides that provides the equivalent of several sheets of white paper.

I'm wondering how a person could do an inexpensive at-home calibration check of an iSis.

So as to not hijack my own thread that has veered OT, I'll start a new thread with the subject of
"HowTo? inexpensive at-home calibration check of iSis?"
http://www.luminous-landscape.com/forum/index.php?topic=76304.0
« Last Edit: March 14, 2013, 11:35:18 am by l_d_allan »
Logged
retired in Colorado Springs, CO, USA ... hobby'ist with mostly Canon gear

l_d_allan

  • Full Member
  • ***
  • Offline Offline
  • Posts: 208
    • Berean photos
Re: Does i1Profiler print targets with all CM turned off? Like ACPU?
« Reply #10 on: March 14, 2013, 11:12:37 am »

Either a Version 4 profile and a bug in the software or a bad series of readings.

Probably a bad series of readings by the CM newbie in the mirror. The iSis should be pretty idiot-proof. But is some village missing its idiot?

Earilier I had problems with the iSis getting out of alignment ... off-by-one ... as it moved left to right. That reflected my ignorance of the iSis_Info parameter line.  Wow, were those profiles bad. I think I've mostly resolved that problem (using BarCodes now). Also, the supplied TC918 with BarCode was less-than-usable because of file-name mismatch ... that may have been PEBKAC.

When I make a printer profile with PMP5, there are "Preferences" for ver.2 and ver.4 ICC spec profiles. There is also an option for "Optimize image preview in PhotoShop".
[Edit]  I looked at this closer, and it appears to apply to profiles for scanners and cameras:
"Rendering intents in digital camera and scanner profiles[/Edit]

Does CS4/5+ expect ver.2 or ver.4 profiles? How about LR 4.3? What would be the correct PMP5 "Preferences" for LR4.x and CS4/5+? Or does it not really matter?

I've tried all four combinations of the PMP5 Preferences above,
  • but that doesn't seem to make any difference
  • as far as the White background of test-prints turning gray with all CS5 Print-Previews and prints using Relative-Colormetric intent
  • and most Print-Previews and prints using Perceptual intent.
  • I've gotten maybe 10% of the profiles I've made to work well with Perceptual Intent
  • white background stays white and colors of the prints actually look good.
  • I've tried to carefully keep track of just how I've made test-chart prints and done the measurement scans, but getting semi-random profiles.

I'm part way through the "Color Management for Photographers" book. It has details for CS and CS2, and I'm trying to translate to LR4/CS4/CS5.

Also, I'm shifting over to a printer-profile workflow using ColorPort and then maybe ArgyllCms to see if the profiles come out better.
« Last Edit: March 14, 2013, 01:05:26 pm by l_d_allan »
Logged
retired in Colorado Springs, CO, USA ... hobby'ist with mostly Canon gear

l_d_allan

  • Full Member
  • ***
  • Offline Offline
  • Posts: 208
    • Berean photos
Re: Does i1Profiler print targets with all CM turned off? Like ACPU?
« Reply #11 on: March 14, 2013, 01:22:01 pm »

Earilier I had problems with the iSis getting out of alignment ... off-by-one ... as it moved left to right. That reflected my ignorance of the iSis_Info parameter line.  Wow, were those profiles bad.

I was trying to recall just how I got "so far out in the weeds".  With the problems with the supplied TC918 test-chart (pebkac?), I then beat my head against the wall trying to get the MeasureTool to generate Test-Charts. Eventually learned my license/dongle didn't enable that functionality.

Then learned the hard way that test-charts from ColorPort aren't all that compatible with PMP5. And that test-charts from i1Profiler without a BarCode aren't all that compatible with PMP5. Lots of trial-and-error on how the iSis_INFO parameter line does and doesn't work (finally gave up and relying on barCodes with no iSis_INFO parameter line in the MyPmpTestChart.txt file).

Sorry for the whining.
Logged
retired in Colorado Springs, CO, USA ... hobby'ist with mostly Canon gear

l_d_allan

  • Full Member
  • ***
  • Offline Offline
  • Posts: 208
    • Berean photos
Re: Printer=profiles bugs in PMP.5.0.10?
« Reply #12 on: March 14, 2013, 01:58:20 pm »

a bug in the software

I've been proceeding with the assumption that my problems are self-inflicted. However, are their known bugs in PMP5.0.10 that can cause flawed printer profiles? My speculation has that PMP5 has the status of very mature software with few if any "show stopper" bugs. Or not?
Logged
retired in Colorado Springs, CO, USA ... hobby'ist with mostly Canon gear

Schewe

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6229
    • http:www.schewephoto.com
Re: Printer=profiles bugs in PMP.5.0.10?
« Reply #13 on: March 14, 2013, 04:07:31 pm »

I've been proceeding with the assumption that my problems are self-inflicted. However, are their known bugs in PMP5.0.10 that can cause flawed printer profiles? My speculation has that PMP5 has the status of very mature software with few if any "show stopper" bugs. Or not?

Pretty sure Andrew was talking about bugs in apps or print pipeline causing a tint to be printed on a white background...

Personally, I think you are biting off more than you can chew with all the gyrations you are doing trying to generate test charts. I think you would be better off just printing the RGB 9.18 iSis chart with Photoshop CS3 or CS4 with No Color Management enabled and see if you can get a good measurement reading and start making some good profiles that work decent before going down the rabbit hole of custom charts and different methods of porting or moving readings and stuff back and forth. It's still not clear whether or not your used iSis is actually working correctly–which if it isn't, is gonna screw up anything you are trying to do.
Logged

l_d_allan

  • Full Member
  • ***
  • Offline Offline
  • Posts: 208
    • Berean photos
Re: Printer=profiles bugs in PMP.5.0.10?
« Reply #14 on: March 14, 2013, 05:14:58 pm »

I think you would be better off just printing the RGB 9.18 iSis chart with Photoshop CS3 or CS4 with No Color Management enabled and see if you can get a good measurement reading and start making some good profiles that work decent before going down the rabbit hole of custom charts and different methods of porting or moving readings and stuff back and forth. It's still not clear whether or not your used iSis is actually working correctly–which if it isn't, is gonna screw up anything you are trying to do.

Agree.

As I write, I'm rescanning some previous TC918-iSis test-charts I hadn't pitched. The Reference Data is supplied by "Chart with Barcode", rather than the iSis_INFO line within TC918.txt that I still don't grok.

They were printed with ACPU, and now don't have the "off-by-one" drifting mis-alignment that I didn't notice when I first started using PMP5. So I perhaps gave up too soon on TC918, and flailed away at MeasureTool Test-Chart-Generator, then flailed away at ColorPort, then flailed away at i1Profiler 1.4.2 demo.

Hopefully, they will result in decent profiles. Or not?

I've got another post on trying to figure out if the n.i.b. iSis is measuring ok, without sending it to X-Rite for re-certification:
http://www.luminous-landscape.com/forum/index.php?topic=76304.0
Logged
retired in Colorado Springs, CO, USA ... hobby'ist with mostly Canon gear

Rhossydd

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3369
    • http://www.paulholman.com
Re: Does i1Profiler print targets with all CM turned off? Like ACPU?
« Reply #15 on: March 15, 2013, 03:16:01 am »

White areas are light to medium gray
I'm pretty certain this is down to a particular combination of settings in PMP. I seem to recall having a similar issue some years back, but can't find the detailed explanation.

For now;
Use version 2 profiles
Large profile size
paper grey perceptual rendering intent
Logo colorful Gamut mapping.
D50

Don't mess with custom targets until you can reliably get good profiles from the default charts. This should work fine.
« Last Edit: April 02, 2013, 05:46:04 pm by Rhossydd »
Logged
Pages: [1]   Go Up