Ad
Ad
Ad
Pages: « 1 [2] 3 4 ... 8 »   Bottom of Page
Print
Author Topic: Adobe Color Printer Utility  (Read 35396 times)
digitaldog
Sr. Member
****
Offline Offline

Posts: 9191



WWW
« Reply #20 on: December 01, 2010, 02:42:45 PM »
ReplyReply

right, it needs the null transform technique to do this in InDD.

Thanks. Looks like the target from ID using sRGB>sRGB match the null technique out of Photoshop CS5. That’s visually, I have to let it dry down and measure.

Man, ID’s color management setup and print setup sure is crude and in need of the Photoshop or LR team to help them clean it up UI wise.
Logged

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

Posts: 623


« Reply #21 on: December 01, 2010, 02:52:37 PM »
ReplyReply

Thanks. Looks like the target from ID using sRGB>sRGB match the null technique out of Photoshop CS5. That’s visually, I have to let it dry down and measure.

Would be great if it does, but I didn't get that result, and hence the thread hijack to indd versus PSCS5 output. Anyway, I will go back and try it again. As I had noted earlier, the target output looks normal enough that it doesn't raise a red flag, but upon measuring I found the errors were pretty significant.

Man, ID’s color management setup and print setup sure is crude and in need of the Photoshop or LR team to help them clean it up UI wise.

Agreed, but it get's the job done. Would definitely be nicer if the whole CS5 suite had a little more UI conformity. It is after all CS5 not CS1 Undecided
Logged
Doyle Yoder
Sr. Member
****
Offline Offline

Posts: 500


« Reply #22 on: December 01, 2010, 03:20:44 PM »
ReplyReply

Indd/CS5's "print path" different from PSCS5? So, the measured values of targets printed from these two apps won't match even when applying the same null transform technique for target printing? Yikes, now that is a problem because it means a profile built from PSCS5 using the null transform method or this new utility (if the new utility in fact matches PSCS5 output) won't work well for Indd and vice versa, nor for older versions of Mac OS and PS, nor for other non Adobe color managed apps. And, likewise, all older profiles built prior to PSCS5 won't work well in PSCS5.  So, I'm with Doyle on this matter.  Seems like it's a really important issue that needs to be resolved, indeed more important than whether the new utility matches PSCS5.

FWIW, after upgrading to OS10.6 and PSCS5, printing from PSCS5 became very suspect to me. It almost works right but not really. Like Doyle, my only answer so far has been to quit printing from PSCS5 and both profile and print using InDD.

First off I was not saying that there is a difference, just that given the problems that we have seen with Apple's new printing path this needs to be considered. In my experience profiles created from targets printed with ID work perfectly in CS4, CS5 and LR. That is if the printer driver works correctly with Apple's new path in 10.6.5 and for that matter with 10.5.8. We have seen many times the moving target from Apple throw a kink in printing from apps that use the new path. Note here my fix for Canon iPF drivers for ACPU. http://www.luminous-landscape.com/forum/index.php?topic=49207.0  This Canon fix is not really an Apple issue/bug, but Canon's way of dealing with this. I can see the pro and cons of this approach. At least you know that the driver is defaulting to No Color Correction seemingly regardless of what changes Apple makes.

The only (old) profiles I can see that might have issues are from targets that were printer through drivers (not correctly written for Apple's new path) that required a workaround. This may be what you have been dealing with printing from CS5. I have seen no color variations printing from CS5 or ID as long as CS5 is listed in Canon's special casing file.

Doyle


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

Posts: 500


« Reply #23 on: December 01, 2010, 03:25:15 PM »
ReplyReply


I concluded at that time the targets printed from InDDCS5 with a null transform technique did match targets I had routinely printed from PSCS3/10.5 on an older PowerPC Mac using PSCS3's no color management mode, but targets printed from PSCS5 under Snow Leopard differed by an average of several delta E.  Both sets of targets looked like they were plausibly printed, nothing way out of whack to indicate a target printing problem, but they just didn't match with reasonable accuracy. That means printing targets using PSCS5 appears to make app specific profile builds, perhaps suitable for use only with PSCS5 and Lightroom, but nothing else including other Adobe apps like InDD. So, if all you want to do is print from PS or perhaps lightroom, then maybe that's fine. But for every other app on the Snow Leopard platform, well it seems like we need to use Indd to make decent profiles. Oh, boy. Two steps forward, two steps back.



You should not be using a null transform technique to print targets from ID. ID has the capability to turn off CM. Use "Emulate Adobe InDesign 2.0 CMS Off" in the Color Settings menu.

Of course with ID before 10.6.2 be sure "Print as Bitmap" is checked.

Doyle
« Last Edit: December 01, 2010, 03:45:15 PM by Doyle Yoder » Logged
MHMG
Sr. Member
****
Offline Offline

Posts: 623


« Reply #24 on: December 01, 2010, 03:50:54 PM »
ReplyReply

You should not be using a null transform technique to print targets from ID. ID has the capability to turn off CM. Use "Emulate Adobe InDesign 2.0 CMS Off" in the Color Settings menu.

Doyle


Thanks for the tip Doyle. Even if I'd noticed that menu setting I would not have trusted it without someone like yourself having OK'd that it works. Good to know it equals "no color management" like in the classic PS.  The null transform method also works fine in ID, but your method is indeed easier to set up and recommend to others.
Logged
MHMG
Sr. Member
****
Offline Offline

Posts: 623


« Reply #25 on: December 01, 2010, 04:03:00 PM »
ReplyReply


The only (old) profiles I can see that might have issues are from targets that were printer through drivers (not correctly written for Apple's new path) that required a workaround. This may be what you have been dealing with printing from CS5. I have seen no color variations printing from CS5 or ID as long as CS5 is listed in Canon's special casing file.

Doyle


Can you elaborate on what you mean by Canon's special casing file? I have an x100 series ipF printer, and I can't get around the scum dot problem in the margins when printing to this printer from PSCS5. It occurs from PSCS5 and not IDCS5 when printing to my canon ipf8100.  I was on the phone with Canon for about an hour when I first upgraded to CS5 and Snow leopard.  The canon techs tossed it around and couldn't recommend a cure. So, I have no choice but to print from ID at the current time or use the 16 bit plug in in 32 bit PS mode (bummer), but I have grown to like the Canon driver's free layout utility better and that rules out PSCS5 until Apple, Adobe, and/or Canon update something.
Logged
teddillard
Sr. Member
****
Offline Offline

Posts: 664


WWW
« Reply #26 on: December 01, 2010, 04:11:12 PM »
ReplyReply

This is beyond lame.  What the heck is going on between Adobe and Apple that they can't fix something like this?
Logged

Ted Dillard
Doyle Yoder
Sr. Member
****
Offline Offline

Posts: 500


« Reply #27 on: December 01, 2010, 04:17:07 PM »
ReplyReply

Can you elaborate on what you mean by Canon's special casing file? I have an x100 series ipF printer, and I can't get around the scum dot problem in the margins when printing to this printer from PSCS5. It occurs from PSCS5 and not IDCS5 when printing to my canon ipf8100.  I was on the phone with Canon for about an hour when I first upgraded to CS5 and Snow leopard.  The canon techs tossed it around and couldn't recommend a cure. So, I have no choice but to print from ID at the current time or use the 16 bit plug in in 32 bit PS mode (bummer), but I have grown to like the Canon driver's free layout utility better and that rules out PSCS5 until Apple, Adobe, and/or Canon update something.

The special casing file (AppColorMatchingInfo.xml) is a file listing the applications that use Apple's new printing path. If the application (internal executable) is not listed in this driver file when you choose to have the app manage color the driver under Main/Color Mode will default to "Color" which results in double profiling. If the application is listed, Main/Color Mode with default to "No Color Correction" which prints correctly without the driver managing color.

This has nothing to do with the printing of the scrum dot in the none image area of the page size. For a better understanding (but potentially more confusing) of the scrum dot issues check out this thread. http://www.luminous-landscape.com/forum/index.php?topic=48860.0

Doyle
« Last Edit: December 01, 2010, 04:19:02 PM by Doyle Yoder » Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9191



WWW
« Reply #28 on: December 01, 2010, 05:50:17 PM »
ReplyReply

You should not be using a null transform technique to print targets from ID. ID has the capability to turn off CM. Use "Emulate Adobe InDesign 2.0 CMS Off" in the Color Settings menu.

Actually, based on a test just done, they produce the same results:

dE Report

Number of Samples: 930

Delta-E Formula dE2000

Overall - (930 colors)
--------------------------------------------------
Average dE:   0.19
    Max dE:   0.72

    Min dE:   0.01
 StdDev dE:   0.10

Best 90% - (836 colors)
--------------------------------------------------
Average dE:   0.16
    Max dE:   0.31
    Min dE:   0.01
 StdDev dE:   0.07

Worst 10% - (94 colors)
--------------------------------------------------
Average dE:   0.39
    Max dE:   0.72
    Min dE:   0.32
 StdDev dE:   0.07

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

Note that the Emulate 2.0 target has not had time to dry down as much as I’d like (I think I’ll measure everything tomorrow). That one max dE of 0.72 is RGB 64/0/32, a dark tone which I’ve seen in the past and would attribute to dry down issues. Yes its close to 1 but I suspect by late tonight or tomorrow, the delta will be far lower.

But here’s the curious part. Lets assume for the time being ID using either method above is fine, how does this compare to using CS4 (No Color Management):

CS4 vs. ID5:
--------------------------------------------------
Average dE:   0.28
    Max dE:   1.06

    Min dE:   0.02
 StdDev dE:   0.16

Best 90% - (836 colors)
--------------------------------------------------
Average dE:   0.24
    Max dE:   0.49
    Min dE:   0.02
 StdDev dE:   0.11

Worst 10% - (94 colors)
--------------------------------------------------
Average dE:   0.64
    Max dE:   1.06
    Min dE:   0.49
 StdDev dE:   0.13

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

These targets have dried a few hours, I don’t like seeing a max dE of 1.06! Interestingly its RGB values are 160/255/96, kind of teal green. Next worst offender at a dE of 1.03 is 255/128/0, an orange. I see the next 5 or so colors listed alternating between a green and orange with high deltaE’s over 1.

OK, how about CS5 vs. CS4 (using CS5 null profile hack):
--------------------------------------------------
Average dE:   0.33
    Max dE:   1.18

    Min dE:   0.02
 StdDev dE:   0.19

Best 90% - (836 colors)
--------------------------------------------------
Average dE:   0.29
    Max dE:   0.59
    Min dE:   0.02
 StdDev dE:   0.14

Worst 10% - (94 colors)
--------------------------------------------------
Average dE:   0.73
    Max dE:   1.18
    Min dE:   0.59
 StdDev dE:   0.14

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

Again, not happy about the high max dE patches. The top 5 are all a teal green (192/255/96 or there about). These are all over a deltaE of 1.


OK, lets examine CS5 with the hack vs. the new APCU product:
--------------------------------------------------
Average dE:   0.24
    Max dE:   0.75

    Min dE:   0.02
 StdDev dE:   0.14

Best 90% - (836 colors)
--------------------------------------------------
Average dE:   0.21
    Max dE:   0.44
    Min dE:   0.02
 StdDev dE:   0.11

Worst 10% - (94 colors)
--------------------------------------------------
Average dE:   0.53
    Max dE:   0.75
    Min dE:   0.44
 StdDev dE:   0.07

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

OK, this is what I expect to see among ALL the various products. Max dE of 0.75 (probably will be less tomorrow). But sort by max dE and again, a green (64/255/64). Next in line is very dark (0/22/0) which again I would attribute to dry down. Keep in mind that even if I printed the same target using the same print process, there will still be differences just from the iSis noise itself (I’d expect something in the range of 0.2 or less deltaE).

So we have CS5 and APCU producing what I’d call today, the same results. We have CS5 and CS4 not producing the same results but close. And ID5 is different still.

To round out the matrix, here is ID5 vs. CS5:
--------------------------------------------------
Average dE:   0.33
    Max dE:   1.08

    Min dE:   0.02
 StdDev dE:   0.16

Best 90% - (836 colors)
--------------------------------------------------
Average dE:   0.29
    Max dE:   0.55
    Min dE:   0.02
 StdDev dE:   0.12

Worst 10% - (94 colors)
--------------------------------------------------
Average dE:   0.66
    Max dE:   1.08
    Min dE:   0.55
 StdDev dE:   0.10

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

Worst offenders are green again (96/224/0) then an orange (255/128/0). Back and forth, green then orange, all at 1.0 or slightly below.
And lastly, to give you an idea of really low detla’s in the example just above, RGB 48/0/255 have a deltaE of 0.02! So yes, we can get really low values as the average dE figure provides. Its the max dE colors that I can’t figure out. Perhaps its printer, don’t know. Everything was output on a 3880 on Luster.

What would be really, really useful for many parties is if those of use with Spectrophotometers could test this matrix. If you don’t have ColorThink Pro, I can build the deltaE reports from saved CGATS files. Sorting by deltaE helps greatly as we can see, based on this one “standard” target (TC918) where in color space the errors are showing.
Logged

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

Posts: 500


« Reply #29 on: December 01, 2010, 05:56:23 PM »
ReplyReply

What printer did you use for this?
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9191



WWW
« Reply #30 on: December 01, 2010, 05:58:39 PM »
ReplyReply

Epson 3880.
Logged

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

Posts: 500


« Reply #31 on: December 01, 2010, 06:27:11 PM »
ReplyReply

Epson 3880.

It would be interesting to see if the Canon iPF6300 showed similar differences.
Logged
MHMG
Sr. Member
****
Offline Offline

Posts: 623


« Reply #32 on: December 01, 2010, 07:25:20 PM »
ReplyReply


...These targets have dried a few hours, I don’t like seeing a max dE of 1.06! Interestingly its RGB values are 160/255/96, kind of teal green. Next worst offender at a dE of 1.03 is 255/128/0, an orange. I see the next 5 or so colors listed alternating between a green and orange with high deltaE’s over 1.

... Keep in mind that even if I printed the same target using the same print process, there will still be differences just from the iSis noise itself (I’d expect something in the range of 0.2 or less deltaE).

IMHO, given that the max deltaE2000 for this many patch values printed on four different prints was less than 1.5, my interpretation is that all target printing techniques produced essentially identical results, all within the experimental error attributable to instrument reading combined with print-to-print inking/substrate variability. I couldn't achieve any where near that result when I tried several months ago, but there has been at least one OS10.6 update and perhaps other system tweaks since then, and alternatively, I could have just plain screwed up in my tests. Operator fatigue can get to be very high in these mind numbing target making exercises, especially for those of us who don't have an Isis Cheesy (Or, as the saying goes "I thought I made a mistake once, but I was wrong!).

Notwithstanding, I've still got the additional but probably unrelated scum dot problem printing from PSCS5 to my Canon printer, so printing from ID is still my only decent work around at the moment.

I will give it another go, and see how I do this time around in matching ID target output to PSCS5 output.  Andrew, thanks for sharing these results.
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9191



WWW
« Reply #33 on: December 01, 2010, 07:31:21 PM »
ReplyReply

IMHO, given that the max deltaE2000 for this many patch values printed on four different prints was less than 1.5, my interpretation is that all target printing techniques produced essentially identical results, all within the experimental error attributable to instrument reading combined with print-to-print inking/substrate variability.

I don’t think so! Those delta’s are way too high. The printer and the Spectrophotometer are going to produce far, far lower max delta’s if printed from the same app using the same driver.

Quote
Notwithstanding, I've still got the additional but probably unrelated scum dot problem printing from PSCS5 to my Canon printer, so printing from ID is still my only decent work around at the moment.

That’s a totally different printing issue (at least for many of us, using V4 profiles from some packages with some print drivers and apps).
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6976


WWW
« Reply #34 on: December 01, 2010, 07:46:54 PM »
ReplyReply

My experience with this utility is "so far, so good". My printer profiling package is the now discontinued XRite Pulse Elite software and DTP-20 Spectrophotometer. The software provides the option to download and save the 729 patch target on two letter-sized pages in TIFF format. I did so, opened them in the new Adobe application and printed them. The process is pretty seamless and simple, although the instructions provided with the app do not exactly match how the application presents itself on my system (Mac version on OSX 10.6.4). But one can figure it out quite easily because there isn't that much that needs to be verified or done. I compared the appearance of the printed sheets with previous successful profiling exercises using the same printer and paper combination (Epson 3800 with Ilford Gold Fibre Silk) and they look pretty much the same. I did not make comparative measurements, because I expect there to be minor differences, given that two different versions of the Epson driver (Windows vs Mac) are involved between the previous (comparator) exercise and the one I am attempting now. I then noted the dimensions of the printed material in Photoshop, and measured the same surface areas with a very accurate steel ruler on the prints. If there is any difference at all, it is very, very small. Hence I expect that the Pulse Elite should be able to read these patches successfully. I shall let the prints dry over-night before attempting that, and I'll report back on both the profile creation experience and my impression of the profile performance with real-world images.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
digitaldog
Sr. Member
****
Offline Offline

Posts: 9191



WWW
« Reply #35 on: December 01, 2010, 08:04:36 PM »
ReplyReply

It would be interesting to see if the Canon iPF6300 showed similar differences.

Doing that now so targets dry. But here’s an odd one. I can’t print the untagged target through CS4 to the iPF6300! Everything acts normal, printer acts like its actually printing, paper comes out without a drop of ink!
Using the same app, a tagged image and (in this one test) Printer Manages Color, I get a print. I can also print the untagged target in CS5, the new Adobe app for this task and InDesign CS5. But CS4, nada. The Export module works as well (but I don’t want to go that route). Anyone else with this printer and CS4 able to print a target using version 2.2.1?
Logged

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

Posts: 500


« Reply #36 on: December 01, 2010, 08:14:51 PM »
ReplyReply

Doing that now so targets dry. But here’s an odd one. I can’t print the untagged target through CS4 to the iPF6300! Everything acts normal, printer acts like its actually printing, paper comes out without a drop of ink!
Using the same app, a tagged image and (in this one test) Printer Manages Color, I get a print. I can also print the untagged target in CS5, the new Adobe app for this task and InDesign CS5. But CS4, nada. The Export module works as well (but I don’t want to go that route). Anyone else with this printer and CS4 able to print a target using version 2.2.1?

No problem here with the 2.17 driver for the iPF9000 for CS4. I printed a TC9.18 target with IDCS5, PSCS4, and the ACPU. Visually they look identical. I will let them dry and read them tomorrow.

I am assuming you updated the AppColorMatchingInfo.xml file to add the ACPU so that it defaults to No Color Correction?
Logged
shewhorn
Sr. Member
****
Offline Offline

Posts: 540


WWW
« Reply #37 on: December 01, 2010, 11:19:19 PM »
ReplyReply

Doing that now so targets dry. But here’s an odd one. I can’t print the untagged target through CS4 to the iPF6300! Everything acts normal, printer acts like its actually printing, paper comes out without a drop of ink!

Holy crap... I had the EXACT same thing happen to me today on a 6100. Try this...

Using the regular driver set the depth to 16 bits, then try again but setting it to 8 bits.

I saw the following behavior... at 16 bits it just fed the paper through the printer. Didn't even more the head. At 8 bits it moved the printer through the paper AND moved the heads this time... it just neglected to lay down any ink. If you did a print to preview it would just show a blank page.

I called Canon support. They were truly puzzled.

This was working fine a few days ago. Only change... I ran that Adobe utility.

A reinstall of the driver brought it back to life but now I'm kind of wondering if anything else was changed. I have a backup from a few days ago that I can run off of to see if the charts it spits out are different.

Cheers, Joe
Logged
Ernst Dinkla
Sr. Member
****
Offline Offline

Posts: 2867


« Reply #38 on: December 02, 2010, 02:25:01 AM »
ReplyReply

This becomes a typical Adobe CM thread again.

I'm paying 6x the price of a Qimage license to get an update on a CS suite. Adobe can not create a CM ghostbusters team to bring one UI + the same CM effects on all its applications including this "cure" tool? Of course the last only needs a CM-Off button on Windows and a CM-Off button on OS-x with a color space loop hidden so no Mac user has to use his brain or might get the impression Apple isn't devine.


met vriendelijke groeten, Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6976


WWW
« Reply #39 on: December 02, 2010, 05:54:56 AM »
ReplyReply

This becomes a typical Adobe CM thread again.

I'm paying 6x the price of a Qimage license to get an update on a CS suite. Adobe can not create a CM ghostbusters team to bring one UI + the same CM effects on all its applications including this "cure" tool? Of course the last only needs a CM-Off button on Windows and a CM-Off button on OS-x with a color space loop hidden so no Mac user has to use his brain or might get the impression Apple isn't devine.


met vriendelijke groeten, Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/

Ernst, sorry, but this kind of smearing might improve the taste of your bread and butter, but it does ZERO to advance the discussion in any kind of constructive manner. I'm probably more sharply critical of several major providers in this industry because of their sheer failure to get their acts together and collaborate properly at a technical level on some pretty fundamental issues which affect a great many people who spend large sums of money on their products. That is the larger scale issue which needs to be dealt with at a level and in a manner where perhaps more effective solutions could be forthcoming in the future, in order to mitigate the risk of these costly mishaps. Meanwhile, what we are trying to do in this thread is share experience to help each other and in the process perhaps help Adobe "get it right" wherever that may be necessary.

As for your smear on Apple computer users, I recently switched from Windows to Mac; on the whole I'm satisfied I did the right thing, and not because I am afraid of either technology after being an intensive user of Windows systems since the 1980s; at the same time, I have several major issues with Apple Computer on account of things which don't work properly; and not being a fan-boy of any corporation or any technical approach, they are hearing from me loud and clear to clean it up. All of them and all of it is only as good as it delivers the goods. You seem to imply that you know the answer to this problem and you seem to imply the whole thing is really very simple, so why don't you contact Adobe and offer them your services to fix it.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Pages: « 1 [2] 3 4 ... 8 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad