Ad
Ad
Ad
Pages: [1] 2 3 »   Bottom of Page
Print
Author Topic: i1 publish problems  (Read 9739 times)
Paul2660
Sr. Member
****
Online Online

Posts: 1673


WWW
« on: April 23, 2011, 09:00:19 AM »
ReplyReply

In late Feb. 2011 I purchased an i1 extreme with the UV cut ( I was told by my dealer I needed the UV cut to help on papers/canvas with optical brighteners)
The software that came with the extreme installed clean but I was using a friends system who has Profile maker 5.0 installed to generate profiles until I received
the upgrade to publish.  I am familiar with the process of creating a profile in 5.0.   The target I used was 4P_TC1728-RGB.  I was able to get excellent scans on any
paper and generate excellent profiles.   My environment is PC win7 and so is the friends machine that has Profile maker 5.0

I have since received publish.  The upgrade to the i1 measuring device went fine.  But the results I am getting with the software are anything but stellar.

On any target scan, (so far I have generated 768, 1585, and 1728 in the new software), by the 16th line I am getting an error and the device will not swipe a color line.
I can still get the data in via spot scans, but this takes way too much time.  On a 4 page target of 1728 I printed on Lexjet esatin from a 9880 I ended up having to do
spot scans 45% of the time. 

When I finally get done with a scan, and save the measurements, then attempt to generate a profile, all of them are clearly bad.  They have have gaping holes in the
3D view and when I proof with them, the colors look shifted  or worse.

I was also under the impression I could import older targets like the 4P_TC1728 (which came from the profile 5.0 software) and use these.    I was able to import the txt file
into publish and generate a target sheet, but when I attempt to create the profile from this scan, the publish software errors out and won't create anything.  You can however
load a previous measurement file done on profile maker 5.0 of 4P_TC1728 and it will generate a clean profile.  So far the only clean one I have seen.

Net, I can't get a complete page of colors swiped without having to go back and manually enter 40 to 45% of the lines spot by spot.  When I finally get done and attempt to
generate a profile, the results are obviously terrible.  i1 was closed Friday as was my dealer, so I have not been able to reach them.

Any ideas would be greatly appreciated.  I am wondering if the i1 device I have with the UV cut is causing the problems and that I should not have been sold this. 

My printing is on RC, Fiber, Canvas, some of which will have OBA's.  Printers are 9880 and 7800.  PC, win7 64 bit is where the publish software is installed.  I also installed it on my older
XP machine but have gotten the same problems.

Thanks
Paul Caldwell
www.photosofarkansas.com


Based on how clean profile maker 5.0 works vs the trash I keep getting from publish, right now I am tempted to back down to profile maker 5.0.   
Logged

Paul Caldwell
Little Rock, Arkansas U.S.
Photography > http://photosofarkansas.com
Blog> http://paulcaldwellphotography.com
Doyle Yoder
Sr. Member
****
Offline Offline

Posts: 500


« Reply #1 on: April 23, 2011, 06:42:14 PM »
ReplyReply

I am not totally impressed with the i1 Pro UV-cut and i1Profiler. Seems to have too many read errors. More than I have ever gotten with Measure Tool or ColorPort. Not sure why but I find I need to read targets much slower and let it sit much longer at the end of a row before releasing the button. I have not had a lot of time to test this but I seem to have more issues with CMYK profiles. It is strange but I am finding like you if I read a target in Measure Tool and bring that into i1Profiler I get great profiles.

I wonder if it is somehow related to this. http://rmimaging.com/information/Chromaxion_Issue_1.pdf

« Last Edit: April 23, 2011, 06:44:19 PM by Doyle Yoder » Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 8763



WWW
« Reply #2 on: April 23, 2011, 08:29:46 PM »
ReplyReply

It is strange but I am finding like you if I read a target in Measure Tool and bring that into i1Profiler I get great profiles.

No reason why that wouldn’t be the case and I too prefer to measure in MT and build the profiles in i1P. I can multi-task too. That was possible with ProfileMaker Pro modular. Not so with i1P. With ProfileMaker Pro, I’d measure in MT, build profiles in the PMP at the same time. I’d build targets in i1P! Its got a better target generator. But then you can use MP or ColorPort and then load the measured data in i1P for building the profile.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
terrywyse
Jr. Member
**
Offline Offline

Posts: 89



« Reply #3 on: April 23, 2011, 09:22:50 PM »
ReplyReply


On any target scan, (so far I have generated 768, 1585, and 1728 in the new software), by the 16th line I am getting an error and the device will not swipe a color line.
I can still get the data in via spot scans, but this takes way too much time.  On a 4 page target of 1728 I printed on Lexjet esatin from a 9880 I ended up having to do
spot scans 45% of the time. 

First thing I thought of was a possible scaling issue....maybe your charts are not being printed at exactly 100%. You didn't mention how you're printing the charts but I would probably recommend saving/exporting a TIFF from i1Profiler rather than printing from within the application.

I'd concur with what others are saying....I do most of my measuring in ColorPort or Measure Tool and then bring the measurement data into i1Profiler....I find measuring in i1Profiler with my iSisXL to still be a little buggy.

Terry
Logged

Terry Wyse, WyseConsul
day job...Color Management Consulting
on the side....photoWyse, photography and fine prints
G7 Certified Expert (but that depends on the day)
Paul2660
Sr. Member
****
Online Online

Posts: 1673


WWW
« Reply #4 on: April 23, 2011, 10:30:51 PM »
ReplyReply

I am printing the targets as tifs, resolution  is 101.6 as I recall.  That is the default resolution that publish uses.  I let publish create the tif. 

I don't believe with i1 publish you get the measure tool anymore, like the older profile maker 5.0 has, the modular interface.  If I understand
the interface, I have to use everything withing the tool. 

I don't have profile maker to fall back on, as I went from the i1 extreme to publish.  In i1 extreme, it seems you only get a set
choice of 3 targets to pick from and I can't find anyway to load any others.

From what I have read from the responses so far, it seems I may have to purchase profile maker 5.x. 

Does anyone have a feeling that it may be that my device has the uv cut filter?  I was told by my dealer I had to have the uv cut
if I wanted read papers with OBA's correctly.  From other reading it seems this may not be the case.

I also noticed on the xrite website, that they offer i1publishpro with a uv cut and non uv cut version.  But not i1 publish.  Does anyone
know if there is a difference in the software to allow you to use the uv cut filter? 

I am currently using i1 publish that doesn't seem to have any mention of the uv cut, only i1publishpro.  But again I am wondering if
the difference is only that xrite ships a measureing device with the uv cut filter or not.

Also does anyone know if you can download profile maker 5.x as a trial?  I know that it requires a dongle to get a licencse.  I just would like
to see if my i1 device works with profile maker 5 and gets better scans.  I briefly tried to use on a friends machine with Profile maker 5 and it
locked up the entire PC twice.

Paul Caldwell

Logged

Paul Caldwell
Little Rock, Arkansas U.S.
Photography > http://photosofarkansas.com
Blog> http://paulcaldwellphotography.com
Rhossydd
Sr. Member
****
Offline Offline

Posts: 1891


WWW
« Reply #5 on: April 24, 2011, 01:37:04 AM »
ReplyReply

Not sure why but I find I need to read targets much slower and let it sit much longer at the end of a row before releasing the button.
Yes, seeing that here with a standard i1 too.
Logged
Ernst Dinkla
Sr. Member
****
Offline Offline

Posts: 2824


« Reply #6 on: April 24, 2011, 05:41:45 AM »
ReplyReply

Yes, seeing that here with a standard i1 too.

Referring to the thread on 2000+ patch readings: a more critical algorithm to sieve usable patch data from boundary data may infuence faster measurements, deliver more  scanning errors when too little samples can be made. Something could have been changed there compared to the other tools. If that is the case then the slower reading and improved data contributes to better/more specific profiles. Which in a way has always improved profiles.

Mr. Myers' detective work is delivering an interesting fact but that issue should affect all kinds of tools used for Eye 1 measuring at any speed. Software that does not create or accept the arbitrary FBA+ compensation on a spectrometer that can not measure FBA effects will not suffer. ColorSage used with the Z3200 spectrometer will receive spectral data at 20Nm intervals in a range of 400 to 700 Nm. The spectral data beyond that range is kept at zero. That is what the Z3200 APS exports in my experience. The problem then does not occur, Mr. Myers solution is the same. It all comes back to the statement that I share with ArgyllCMS' approach; better subtract what you measured on FBA effect than add an artificial FBA effect where nothing is measured.

met vriendelijke groeten, Ernst


New: Spectral plots of +250 inkjet papers:

http://www.pigment-print.com/spectralplots/spectrumviz_1.htm
Logged
Rhossydd
Sr. Member
****
Offline Offline

Posts: 1891


WWW
« Reply #7 on: April 24, 2011, 05:56:05 AM »
ReplyReply

a more critical algorithm to sieve usable patch data from boundary data may influence faster measurements, deliver more  scanning errors when too little samples can be made.
It's possible, or it might just be more poorly written software too.
The inability to read charts from certain printer/ink/paper combinations I'm seeing may well be due to an over zealous algorithm looking for errors.
Still waiting to hear back from X-Rite on that one.

Paul
Logged
Doyle Yoder
Sr. Member
****
Offline Offline

Posts: 500


« Reply #8 on: April 24, 2011, 06:28:41 AM »
ReplyReply

No reason why that wouldn’t be the case and I too prefer to measure in MT and build the profiles in i1P. I can multi-task too. That was possible with ProfileMaker Pro modular. Not so with i1P. With ProfileMaker Pro, I’d measure in MT, build profiles in the PMP at the same time. I’d build targets in i1P! Its got a better target generator. But then you can use MP or ColorPort and then load the measured data in i1P for building the profile.

How do you build targets in i1P and then get them to match the printed target in Measure Tool or even import them into ColorPort? A target in i1P that is 20 columns wide in comes into MT at 41 columns wide.
« Last Edit: April 24, 2011, 06:38:28 AM by Doyle Yoder » Logged
Ernst Dinkla
Sr. Member
****
Offline Offline

Posts: 2824


« Reply #9 on: April 24, 2011, 06:42:17 AM »
ReplyReply

It's possible, or it might just be more poorly written software too.
The inability to read charts from certain printer/ink/paper combinations I'm seeing may well be due to an over zealous algorithm looking for errors.
Still waiting to hear back from X-Rite on that one.

Paul

It would be interesting to see whether i1Profiler profiles are that better than for example Profiler profiles if Measure Tool or Colorport is used to collect the samples. Patch reading is of course the standard to compare to. Similar reading errors there should indicate that the flaw is not in the scan reading algorithm.

met vriendelijke groeten, Ernst

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/
Logged
PhilipCummins
Full Member
***
Offline Offline

Posts: 102


« Reply #10 on: April 24, 2011, 10:06:20 AM »
ReplyReply

I am currently using i1 publish that doesn't seem to have any mention of the uv cut, only i1publishpro.  But again I am wondering if the difference is only that xrite ships a measureing device with the uv cut filter or not.

Pretty sure it's just that, the i1 Pro has firmware that says whether it's UV-cut or No Filter so the software can figure it out on its own.

Also does anyone know if you can download profile maker 5.x as a trial?

ProfileMaker Pro is available from X-Rite's site as a trial here.
Logged
Paul2660
Sr. Member
****
Online Online

Posts: 1673


WWW
« Reply #11 on: April 24, 2011, 01:22:22 PM »
ReplyReply

Phillip, thanks for the link.

A few updates since my last post.

My next thought pattern was to drop back to the i1match software just to see if I could create anything with it.
You only get 3 targets to pick from I chose the large one.  1st attempt to scan, the about 2/3's of the way through,
the entire software just locked up.  I had seen this same thing happen on my friends machine when using my device and USB
cable.  Net, I replaced the cable and got a clean scan and match generated a profile that I could look at.

I then went back to publish and using the new USB cable, I attempt to rescan my 1585 patch target I had generated.  This time
it worked near perfect, I had to run slower on a few of the lines, but got all in via normal scanning.
I felt great Smiley   But it was not to last Cry  I then tried to take the new measurements and create a profile and I got the same
error, "can't generate a profile".  Moved the measurement data over to my XP machine and loaded it and got the same error.
So now I hoping I can test the measurement data on profile maker and see if it works. 

Thanks to all the help so far.
Paul Caldwell
Logged

Paul Caldwell
Little Rock, Arkansas U.S.
Photography > http://photosofarkansas.com
Blog> http://paulcaldwellphotography.com
VitOne
Full Member
***
Offline Offline

Posts: 114


« Reply #12 on: April 24, 2011, 03:13:50 PM »
ReplyReply

I don't understand why an UV-Cut instrument should help with papers that have OBAs.
I suggest to read this link: www.argyllcms.com/doc/FWA.html and ask more information to your vendor. For what I read an insturment without UV reading cababilities will not compesate OBAs. Can you tell me what do you think?
Logged
Shane Webster
Jr. Member
**
Offline Offline

Posts: 73


« Reply #13 on: April 25, 2011, 06:50:04 AM »
ReplyReply

Quote
No reason why that wouldn’t be the case and I too prefer to measure in MT and build the profiles in i1P. I can multi-task too.

Unfortunately, moving from PMP5 to i1P is a tad painful for me.  The issue was repeatable by X-Rite on the tech's XP machine (I'm on a Mac) so I know it's not user error (which it generally is).  I received the free upgrade to i1P from MP.  I also own PMP5.  The i1P dongle and my PMP5 dongle cannot be connected to the computer at the same time.  If I need to get into PMP5, I must remove the i1P dongle and vice versa.  I have no idea how X-Rite handles their licenses within their dongles but find the conflict between the two rather annoying.  The tech was going to put in a request to find out whether this was an expected and/or intended result.  Hopefully, they'll get it resolved.  Luckily, I can still measure in MT without the dongle, if memory serves. . .
Logged
terrywyse
Jr. Member
**
Offline Offline

Posts: 89



« Reply #14 on: April 25, 2011, 09:01:17 AM »
ReplyReply

I am printing the targets as tifs, resolution  is 101.6 as I recall.  That is the default resolution that publish uses.  I let publish create the tif. 

I don't believe with i1 publish you get the measure tool anymore, like the older profile maker 5.0 has, the modular interface.  If I understand
the interface, I have to use everything withing the tool.

Just download ProfileMaker 5.x from X-Rite's website, install it and use the Measure Tool application. A license/dongle isn't required if you just want to measure and save the data
 
Quote
From what I have read from the responses so far, it seems I may have to purchase profile maker 5.x. 

Nope...just download and use Measure Tool in demo mode.

Quote
Does anyone have a feeling that it may be that my device has the uv cut filter?  I was told by my dealer I had to have the uv cut
if I wanted read papers with OBA's correctly.  From other reading it seems this may not be the case.

Your issue has nothing to do with the filtration of your i1Pro.

As far as a UVcut instrument measuring papers with OBAs more "correctly", that's an ongoing debate. I would say that *most* users would be safer with a UVcut instrument but if you're doing press proofing and want to adhere to standards then you should be using a "no filter" instrument.  Personally, I measure using my iSisXL in both UVex (UV exclude) and UVinc (UV include) modes and average the results. This gives me the best visual coorelation.....the effects of OBAs are still in the measurement data but they are "tempered" or tamed a bit by mixing in the UVcut data...bottom line, works for me....but this is not always pratical for folks that can't afford an iSis or purchase two i1Pro's with each filter.


Quote
I also noticed on the xrite website, that they offer i1publishpro with a uv cut and non uv cut version.  But not i1 publish.  Does anyone
know if there is a difference in the software to allow you to use the uv cut filter? 

Nothin' to do with the software at all......i1PublishPro is simply a bundle that includes i1Publish and an i1Pro instrument....the difference is simply the choice of the instrument.

Quote
I am currently using i1 publish that doesn't seem to have any mention of the uv cut, only i1publishpro.  But again I am wondering if
the difference is only that xrite ships a measureing device with the uv cut filter or not.

Yes.

Quote
Also does anyone know if you can download profile maker 5.x as a trial?  I know that it requires a dongle to get a licencse.  I just would like
to see if my i1 device works with profile maker 5 and gets better scans.  I briefly tried to use on a friends machine with Profile maker 5 and it
locked up the entire PC twice.

You'll be downloading the entire PM5 package but only using the Measure Tool application....you could throw away the rest of the software and just keep Measure Tool. Like I said above, no dongle required, just run it in demo mode to do your measurements.

Terry
« Last Edit: April 25, 2011, 09:06:35 AM by terrywyse » Logged

Terry Wyse, WyseConsul
day job...Color Management Consulting
on the side....photoWyse, photography and fine prints
G7 Certified Expert (but that depends on the day)
digitaldog
Sr. Member
****
Offline Offline

Posts: 8763



WWW
« Reply #15 on: April 25, 2011, 09:06:34 AM »
ReplyReply

How do you build targets in i1P and then get them to match the printed target in Measure Tool or even import them into ColorPort? A target in i1P that is 20 columns wide in comes into MT at 41 columns wide.

Its a bit of a kludge but doable. In MeasureTool, in the Chart Generator you can click to load a reference you saved out from i1P. Use the txt file. Then you need to play around a bit in specifying the size of the page, patches etc but you can get them to sync up with the target you generated in i1P. I’ve done this successfully a number of times with iSis targets, I suspect it should work with other instruments. Again, key is making sure that after you import the reference of patches and their values, you setup patch size and page layout until you get that magical visual match. Then the reference you save from MT should allow you to read that i1P target. Yes, it should have been a LOT easier going backwards.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Doyle Yoder
Sr. Member
****
Offline Offline

Posts: 500


« Reply #16 on: April 25, 2011, 10:12:07 AM »
ReplyReply

Its a bit of a kludge but doable. In MeasureTool, in the Chart Generator you can click to load a reference you saved out from i1P. Use the txt file. Then you need to play around a bit in specifying the size of the page, patches etc but you can get them to sync up with the target you generated in i1P. I’ve done this successfully a number of times with iSis targets, I suspect it should work with other instruments. Again, key is making sure that after you import the reference of patches and their values, you setup patch size and page layout until you get that magical visual match. Then the reference you save from MT should allow you to read that i1P target. Yes, it should have been a LOT easier going backwards.

Using an i1 Pro, with Chart Generator there is no way to sync to a i1P target. The best you can do is create a new target from the i1P values, but that will not work (can't save) for those that want to download the demo just for MT. In ColorPort you can import the values and make a new target from them, read that target in CP and drop the measurement file into i1P.

Seems there should be a way to make the i1P generated txt file load in MT just like other txt reference files.

Logged
rasworth
Jr. Member
**
Offline Offline

Posts: 90


« Reply #17 on: April 25, 2011, 02:52:55 PM »
ReplyReply

I added one line to the i1P generated txt file, the "LGOROWLENGTH  54", seems to be enough to allow Measure Tool to function, although more parameters would make it less kludgey (still have to tell it to measure in strip mode, can only scan one direction).

Here's an example of the file with the added parameter:

CGATS.5

ORIGINATOR "X-Rite, Inc."
DESCRIPTOR "RAS 810 85x11"
CREATED "April 22, 2011"
INSTRUMENTATION "Not specified"
MEASUREMENT_SOURCE "Not specified"

LGOROWLENGTH   54
NUMBER_OF_FIELDS 4
BEGIN_DATA_FORMAT
SAMPLE_ID RGB_R RGB_G RGB_B
END_DATA_FORMAT

NUMBER_OF_SETS 810
BEGIN_DATA
1 0.0000 0.0000 0.0000
2 31.875 0.0000 0.0000
3 63.750 0.0000 0.0000
4 95.625 0.0000 0.0000
5 127.50 0.0000 0.0000
6 159.38 0.0000 0.0000
7 191.25 0.0000 0.0000
8 223.13 0.0000 0.0000
9 255.00 0.0000 0.0000
10 0.0000 31.875 0.0000
etc.

Richard Southworth
Logged
Doyle Yoder
Sr. Member
****
Offline Offline

Posts: 500


« Reply #18 on: April 25, 2011, 08:51:11 PM »
ReplyReply

Hey that works. You just have make sure the number for "LGOROWLENGTH" is the same as the rows in your target.
Logged
rasworth
Jr. Member
**
Offline Offline

Posts: 90


« Reply #19 on: April 25, 2011, 09:50:38 PM »
ReplyReply

Yeah, forgot to point that out.

It's still iffy; I generated a measurement file with MT, brought it into i1P as a CGATS file, saved it back out again as an i1P .mxf file, then shut down i1P.  Brought i1P back up again and successfully generated a profile.  If I tried to generate the profile straight from the CGATS file or skipped the shutdown step i1P would abend (Windows shuts it down) at the 96% point.

This program still has a way to go to be out of beta test stage.

Richard Southworth
Logged
Pages: [1] 2 3 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad