Ad
Ad
Ad
Pages: « 1 [2] 3 »   Bottom of Page
Print
Author Topic: z3100 Colour Issue  (Read 13314 times)
Roscolo
Sr. Member
****
Offline Offline

Posts: 621


« Reply #20 on: June 15, 2007, 04:52:48 PM »
ReplyReply

Quote
I've created a TIFF file in ProPhoto space with the above images and four other images I use to gauge soft proofs and colour accuracy.  It's sized to fit on a 24" wide roll taking up 5 inches of space.  3.7MB Test File

Curious to see the results others get, especially in the skin tones.

Now, if someone would send me a RIP to test....
[a href=\"index.php?act=findpost&pid=123033\"][{POST_SNAPBACK}][/a]

I am about to leave today, but I will try to print some of these tonight or tomorrow (Sat.). I just received my free paper with the Rebate, so I have HP Matte Litho Realistic and HP Super Heavyweight Matte Plus, as well as some Kodak Rapid Dry Lustre.
Logged
neil snape
Sr. Member
****
Offline Offline

Posts: 1432


WWW
« Reply #21 on: June 16, 2007, 01:02:59 AM »
ReplyReply

I would have preferred a JPG as this tiff is pretty low res (100ppi).
There are a lot of out of gamut colours in the synthetic blues, and even the richness of the flowers isn't going to be had on a lot of printers.
Here is the soft preview of relative on from l 2 right, Durst Lambda, Canon iPf5000 on Glossy, Epson 9800 Premium Glossy Bill Atkinson profile, HP Z on ID Glossy.
comparison Lambda,Canon,Epson,HPZ

And yes I print a screen to print match within the gamut boundaries as expected.
Logged
neil snape
Sr. Member
****
Offline Offline

Posts: 1432


WWW
« Reply #22 on: June 16, 2007, 01:27:01 AM »
ReplyReply

Also for the question of a rip, here is a preview of ImagePrint V7 vs HP driver.
Please consider the IP color maps not to be fixed yet. The driver is shipping so it should be considered as such.
Oh I forgot this is Glossy on the driver and Satin in IP as I only have a profile for Satin.

HP driver vs ImagePrint v7 beta
« Last Edit: June 16, 2007, 01:28:14 AM by neil snape » Logged
neil snape
Sr. Member
****
Offline Offline

Posts: 1432


WWW
« Reply #23 on: June 16, 2007, 01:51:15 AM »
ReplyReply

Yet another view on this. Now I didn't have time to make a new cal or profile for Hahnemuhle PhotoRag 308 since this weeks arrival of the new Z, here are two views , soft proofing of relative with simulate BP and Paper White. It is a profile from Print Center, on Collector Canvas settings which no longer will be as good as >250 g/m2 fine art. This clearly shows where HP and Canon cannot and will not perform in the reds as well as Epson.
It also clearly demonstrates the superior  Dmax  on the HP. What  you prefer is up to you. As I have said many times the HP profiles have a problem in the return portion of the rgb from Lab so the prints always look better in print than on screen. There are good things in both previews. In the flowers the Epson handles the mapping much better, but don't forget this is relative mapping from Kodak ProPhoto of a magenta with near clipping in there (some g numbers around 4) so a super saturated magenta is normal for all spaces with a smaller than the Prophoto gamut.

HPR 308 Epson vs HPZ

I also wanted to say this is the weaker side of the Z as I have said enough times already. On rag and relatively uncoated papers the HP pigments are not strong. It requires a more substantial coating to produce best results. It is not such a problem on HP Smooth or similar. The reds are acceptable, and the overall print quality on HP Smooth very much in line with any other printer.
« Last Edit: June 16, 2007, 02:01:12 AM by neil snape » Logged
SeanPuckett
Full Member
***
Offline Offline

Posts: 245


WWW
« Reply #24 on: June 16, 2007, 09:55:33 AM »
ReplyReply

Neil, your efforts are appreciated but do nothing to provide clarification of the problem.

Soft proofs do match not the prints.  I can look at the portrait all day on my monitor in soft proof and see that it is neither too red nor too yellow nor too dark nor too blotchy nor too saturated, and yet when I go to print, it is all of these things when sent through application managed colour via an ICC profile made either by the printer, by Argyll or by ProfileMaker.

My monitor is calibrated.
I have checked for double profiling.
This worked just fine with the previous firmware.
Logged

neil snape
Sr. Member
****
Offline Offline

Posts: 1432


WWW
« Reply #25 on: June 16, 2007, 10:16:53 AM »
ReplyReply

Quote
Neil, your efforts are appreciated but do nothing to provide clarification of the problem.

Soft proofs do match not the prints.  I can look at the portrait all day on my monitor in soft proof and see that it is neither too red nor too yellow nor too dark nor too blotchy nor too saturated, and yet when I go to print, it is all of these things when sent through application managed colour via an ICC profile made either by the printer, by Argyll or by ProfileMaker.

My monitor is calibrated.
I have checked for double profiling.
This worked just fine with the previous firmware.
[a href=\"index.php?act=findpost&pid=123124\"][{POST_SNAPBACK}][/a]

I did say that the print on ID Gloss looks exactly like the soft proof which in turn looks like the original. I think I explained the conditions too, but I haven't tried on matte paper, which I also said.
Logged
SeanPuckett
Full Member
***
Offline Offline

Posts: 245


WWW
« Reply #26 on: June 16, 2007, 10:29:52 AM »
ReplyReply

Quote
I did say that the print on ID Gloss looks exactly like the soft proof which in turn looks like the original. I think I explained the conditions too, but I haven't tried on matte paper, which I also said.
[a href=\"index.php?act=findpost&pid=123129\"][{POST_SNAPBACK}][/a]
Neil,

I've reread your posts above and don't see anything about actual prints -- I'm sorry if I'm just not understanding you.  I believe what you say about the prints matching, but this is not what I see with any media I try.  I would really like this problem to just go away with a "oh, look, I'm doing it wrong" but I can find nothing, anywhere, in my configuration that indicates that I am doing so.  Remember that I have tried on the Mac and on XP (two different computers) and the failure mode is identical each time.  I have pounds of printout that all look wrong in exactly the same way, except  when they are even more wrong because I have intentionally double-profiled.

From my analysis of the problem, I am wondering if the driver or the printer is ignoring the "application managed" option, when it is selected, and performing some sort of colourspace conversion or automatic print adjustment, so that when the raw RGB comes from the application it is translated even though it is told not to.

I wonder further if my printer, having had multiple firmware upgrades applied, somehow works differently than printers that have more recently shipped with a virgin firmware installation.

I wonder if it is possible to install a "null" profile into the printer for a paper type, such that even if it was told to manage colour itself, it would actually not perform any lookup.  And, thus, it would be impossible for the printer or driver to screw up application managed colour at all.

Does anyone know of a pre-built "null transform" ICM or ICC profile I could try this theory with?  I could probably convince argyll to create one, but if one already exists I would  be more sure of doing it right and not spoiling the test.

Really, all I want is my printer back.  I'm so weary and fed up.
Logged

marclw
Newbie
*
Offline Offline

Posts: 27


« Reply #27 on: June 16, 2007, 12:55:05 PM »
ReplyReply

I think that I'm experiencing the same kind of problem. I just got my 24" Z3100 two weeks ago, and, as long as I print with the printer handling color, then everything is fine. If I let PSCS3 do the color and printing through the HP profile, I get unacceptable results. Here's an example of my problem.

The shot at the left is a preview pdf of the photo being handled by the printer, and the one at the rifht is being processed via PSCS3. Look at the differences! The one on the left matches my calibrated apple 23" monitor.

What gives? BTW, I am using the latest firmware. Should I open up a case with HP too??

Marc Weinberg
Logged
neil snape
Sr. Member
****
Offline Offline

Posts: 1432


WWW
« Reply #28 on: June 16, 2007, 01:01:02 PM »
ReplyReply

Two questions Marc,
is it printing the same from any previous Adobe apps , like CS2 Photoshop?
and is the capture on the right lighter etc, than if you print with PCM?

Normally the PCM should be the lighter of the two.

I'm still trying to get a handle on the problem, trying to see what direction the PCM vs Photoshop managed colors is going. I don't have CS3 yet , it has not been delivered thus I'm still waiting.
And is this PC or Mac?
Logged
Scott Martin
Sr. Member
****
Offline Offline

Posts: 1311


WWW
« Reply #29 on: June 16, 2007, 02:12:59 PM »
ReplyReply

Quote
Here is the soft preview of relative on from l 2 right, Durst Lambda, Canon iPf5000 on Glossy, Epson 9800 Premium Glossy Bill Atkinson profile, HP Z on ID Glossy.
comparison Lambda,Canon,Epson,HPZ
[a href=\"index.php?act=findpost&pid=123077\"][{POST_SNAPBACK}][/a]

I've been doing a number of similar comparisons with prints from these printers and I'm seeing different results from what this comparison would suggest. Are all of these devices profiled with exactly the same profiling software and method? On the Canon IPF printer was the driver or plug-in used? If the driver, then I'd suggest trying it again with the plug-in as the gamut is dramatically improved. And I assume on the Z and K3 the driver was used?
« Last Edit: June 16, 2007, 02:17:37 PM by Onsight » Logged

SeanPuckett
Full Member
***
Offline Offline

Posts: 245


WWW
« Reply #30 on: June 16, 2007, 02:22:15 PM »
ReplyReply

Quote
.... as long as I print with the printer handling color, then everything is fine. If I let PSCS3 do the color and printing through the HP profile, I get unacceptable results.
[a href=\"index.php?act=findpost&pid=123144\"][{POST_SNAPBACK}][/a]

Marc,

Yes, exactly, this is the problem.  However, and further, profiles created by third party tools (ColorElite, ProfileMaker, Argyll) also do not preform properly -- even when they work well when stuffed into the printer via ICC profile installation.

The problem with installing them into the printer, of course, is the limited gamut available, and the poor mapping of out-of-gamut colours.

Glad to know I'm not insane.

-s
Logged

dkeyes
Sr. Member
****
Offline Offline

Posts: 259


WWW
« Reply #31 on: June 16, 2007, 03:06:54 PM »
ReplyReply

Sean,
Printed your files with Mac OSX, Latest driver and firmware (this was my 1st driver and 3rd firmware upgrade). I am using APS 1.21 (just released a few days ago). Paper is HP ID gloss.
Printed both percetual +BPC and relative + BPC. Both match respective soft proofs on screen (never been an issue for me). Relative seems closest to the native Prophoto space with obvious clipping in the saturated blues and magentas when viewed against soft proof.
The prints look decent (for 100dpi). The skin tones are slightly red on shadow side of face with relative, even more with perceptual.

From what I understand, your issue is prints not matching soft proof since last firmware/driver upgrade? The three things that changed for me were:
1. Serious network problems. Finally figured it out after a week of HP support and my own noodling.
2. Colors are a bit more saturated with perceptual than they were before. I always print perceptual since it seems to match my existing prints best (from Epson 9800) and matched Prophoto scans.
Now all the work seems to match best with relative color intent.
3. I noticed my profiles weren't being updated automatically in my paper list. Now I go into printer utility and manually update/add profiles I've made.

- Doug
ps. If you want the prints I can mail them to you.
Logged

rdonson
Sr. Member
****
Offline Offline

Posts: 1420


WWW
« Reply #32 on: June 16, 2007, 03:43:04 PM »
ReplyReply

Quote
I've created a TIFF file in ProPhoto space with the above images and four other images I use to gauge soft proofs and colour accuracy.  It's sized to fit on a 24" wide roll taking up 5 inches of space.  3.7MB Test File

Curious to see the results others get, especially in the skin tones.

Now, if someone would send me a RIP to test....
[a href=\"index.php?act=findpost&pid=123033\"][{POST_SNAPBACK}][/a]

A very interesting test image, Sean.  An 8 bit TIFF in ProProto space.  Anyway, thanks for making it 5"x24".  I create a custom size of 6"x24" and printed out some app managed color tests.  Next are the printer managed color tests.

My tests were run on HP Premium Instant-Dry Satin with GE On using Perceptual Rendering.  Perceptual looked better in soft-proof than relative colorimetric.  These were also done on a PC (XP Pro) with the latest firmware and drivers.

The first round yielded an interesting finding.  Based on Neil's recommendation I used both CS3 and CS2 and threw in a print from QStudio (Qimage) as well.  I screwed up the QStudio print though by leaving final sharpening on.

The prints I got were pretty darn close to the soft-proofing in Photoshop.  There were some interesting differences between CS2 and CS3 though.

Original Image


Soft-proofed


Out of Gamut in CS3


Out of Gamut in CS2


I'm loathe to report much until the prints have more time to dry but I think CS2 yields the better print.  Closer to what you're looking for.  The difference is subtle and really only noticeable on the b&w portrait and the color portrait.  The color portrait is slightly less "ruddy" in the skin tone.

More later....
« Last Edit: June 16, 2007, 03:59:51 PM by rdonson » Logged

[span style='font-size:14pt;line-height:100%'][span style='font-family:Arial'][span style='font-family:Geneva'][span style='font-size:8pt;line-height:100%']Regards,
Ron[/span][/span][/span][/span]
Ernst Dinkla
Sr. Member
****
Offline Offline

Posts: 2805


« Reply #33 on: June 16, 2007, 03:50:56 PM »
ReplyReply

Quote
Sean

Could you check the settings by accessing the driver properties from QImage?

Make sure that the Advanced>Graphic>Image Color management>ICM Method is set to ICM disabled.

This should allow QImage to handle color management properly, as long as the proper source and output profiles are set up in QImage, and the color tab setting for the printer is set for app managed color.


Regards,

Marc Sitkin
[{POST_SNAPBACK}][/a]

There are some odd choices and obscuring texts in the Z3100 driver and Qimage's choices on CM. This is the Windows combination of course.

For instance the full Z3100 driver choices gives two spots where you can select CM on, off, etc. First in the start>printer+faxes>Z3100>preference settings>advanced>graphic>ICM there are 3 choices: ICM off, ICM by host (I guess that application is meant but it could be network related), ICM by printer. I have the Dutch driver and translate it here so it can be phrased different on your systems. It is set on off in my settings. Then there is the choice between printer controlled ICM and application controlled ICM under the color tab in the driver menu. The one you also can change at the last stage before printing in the driver menu that pops up in the application. I have that one set on application controlled ICM in the preferred settings too so the whole setup should default to application controlled ICM normally.

In Qimage I get the same choices offered as the one first described for the Z3100 driver: ICM enabled (with the green text: Qimage will manage color via the printer profile below), ICM disabled (off) (with the same green text though so not correct, it should simply disappear or read: no ICM), it is only when you click the printer icon in that menu that the green text says: Printer/driver will manage color: use the ICM option in driver. The profile choice window in Qimage shows the sRGB profile that is actually the default "space" profile for printer/driver color management in my case. But 1st the printer/driver may not have a real ICM option, 2d you can still change the printer profiles in that Qimage profile choice window and after OK that profile pops up in the general profile window of Qimage and Qimage will control ICM again (which is understandable but opens up confusing choices for newbies, the profile choice window should simply grey out and not be accessible after you let the printer/driver do CM). Also the driver menu still is showing that it uses application controlled ICM when you open it in Qimage while you have set printer/driver ICM control on in Qimage itself. So while it is possible to have two profiles on top of one another in many workflows in this case it is also possible to fall between two stools with this application, or at least not get the right information back which program is doing what. Custom profiling per image in the Qimage print page is still open for selections but defaults to printer/driver ICM control after the OK so should not have been available after the general choice for printer/driver CM control.

There are more problems of cooperation between the two programs and I can not write who is to blame as I am not capable to find what causes it. But at least be aware that you need a consistent method to avoid contradictions in the  menus and hopefully you will get consistent working of ICM as a result. I'm keeping the printer/driver away from any CM control but the media settings.

Ernst Dinkla

try: [a href=\"http://groups.yahoo.com/group/Wide_Inkjet_Printers/]http://groups.yahoo.com/group/Wide_Inkjet_Printers/[/url]
Logged
neil snape
Sr. Member
****
Offline Offline

Posts: 1432


WWW
« Reply #34 on: June 16, 2007, 03:59:43 PM »
ReplyReply

For the question of the poster on what was used for the 4 up soft proofs, yes all GMB PM except the Lambda which is Binuscan. The Canon profile was from an early version of the driver I believe. If you have a new one from the plug-in by all means send it my way.

It may well be that CS3 is not working as expected rather than the HP driver and firmware.

And yes the skin tones in the test image are quite red. As much as 11 % more magenta than cyan on the shadow cheek. Yet that shows very well on screen, so I'll assume it is intended.
Logged
SeanPuckett
Full Member
***
Offline Offline

Posts: 245


WWW
« Reply #35 on: June 16, 2007, 06:32:48 PM »
ReplyReply

My problems began after loading the latest firmware / driverset. Before that, I was quite content with soft proof consistency whether or not I used AMC or PMC.

The AMC to softproof mismatch is much more obvious on low gamut (e.g. art or watercolour) papers.  See sample images in first post.  As my other prints show, it still reveals itself on high gamut media.

I don't have CS3, but do my non-qimage testing with CS2.  The output is essentially the same, which suggests to me that it's not QImage related.
Logged

Roscolo
Sr. Member
****
Offline Offline

Posts: 621


« Reply #36 on: June 17, 2007, 02:46:12 PM »
ReplyReply

Quote
My problems began after loading the latest firmware / driverset. Before that, I was quite content with soft proof consistency whether or not I used AMC or PMC.

The AMC to softproof mismatch is much more obvious on low gamut (e.g. art or watercolour) papers.  See sample images in first post.  As my other prints show, it still reveals itself on high gamut media.

I don't have CS3, but do my non-qimage testing with CS2.  The output is essentially the same, which suggests to me that it's not QImage related.
[a href=\"index.php?act=findpost&pid=123196\"][{POST_SNAPBACK}][/a]

Sean, I just printed your sample image on Kodak Rapid Dry Lustre (it's what I had loaded in the printer).

Five of six images in your sample are spot-on perfect match to my monitor. The only image that looks slightly different, is the 3rd image from the left. My monitor looks a bit more blue - the print appears more purple. I haven't calibrated my monitor in 3 weeks and it is aging as well. There is not over-saturation in the red flowers - they are a perfect match. There is no over-saturation or red in the shadow side of the color portrait - it's a match. The other 3 images (b&W's and the 2nd image from the left are a match as well. Before printing, I did have to lighten the image in Photoshop CS2 just using Brightness scale, up to +35, because it just appeared too dark on my monitor.

I am using XP Pro SP2, latest driver and firmware for the z3100. I printed with Application Managed Color using the custom profile I made for Kodak Rapid Dry Lustre using the z3100's profiling system. In Photoshop I printed with Rendering Intent set to Perceptual and Black Point Compensation Checked. I have no RIP or QImage installed on my machine.

Maybe something is up with your individual machine or your set-up.

I hope this helps. If you don't resolve the problem, I will be happy to print this again on some HP Matte or Litho-Matte when I load one of those papers.
Logged
marclw
Newbie
*
Offline Offline

Posts: 27


« Reply #37 on: June 17, 2007, 03:32:44 PM »
ReplyReply

Quote
Two questions Marc,
is it printing the same from any previous Adobe apps , like CS2 Photoshop?
and is the capture on the right lighter etc, than if you print with PCM?

Normally the PCM should be the lighter of the two.

I'm still trying to get a handle on the problem, trying to see what direction the PCM vs Photoshop managed colors is going. I don't have CS3 yet , it has not been delivered thus I'm still waiting.
And is this PC or Mac?
[a href=\"index.php?act=findpost&pid=123145\"][{POST_SNAPBACK}][/a]
It's a Mac, and I'm using CS3. I don't have CS2 installed anymore. The photos would print out exactly as they appear. It seems that only some of the colors are affected . . . .

Marc
Logged
rdonson
Sr. Member
****
Offline Offline

Posts: 1420


WWW
« Reply #38 on: June 17, 2007, 06:14:14 PM »
ReplyReply

Quote
A very interesting test image, Sean.  An 8 bit TIFF in ProProto space.  Anyway, thanks for making it 5"x24".  I create a custom size of 6"x24" and printed out some app managed color tests.  Next are the printer managed color tests.

My tests were run on HP Premium Instant-Dry Satin with GE On using Perceptual Rendering.  Perceptual looked better in soft-proof than relative colorimetric.  These were also done on a PC (XP Pro) with the latest firmware and drivers.

The first round yielded an interesting finding.  Based on Neil's recommendation I used both CS3 and CS2 and threw in a print from QStudio (Qimage) as well.  I screwed up the QStudio print though by leaving final sharpening on.

The prints I got were pretty darn close to the soft-proofing in Photoshop.  There were some interesting differences between CS2 and CS3 though.

Original Image


Soft-proofed


Out of Gamut in CS3


Out of Gamut in CS2


I'm loathe to report much until the prints have more time to dry but I think CS2 yields the better print.  Closer to what you're looking for.  The difference is subtle and really only noticeable on the b&w portrait and the color portrait.  The color portrait is slightly less "ruddy" in the skin tone.

More later....
[a href=\"index.php?act=findpost&pid=123167\"][{POST_SNAPBACK}][/a]

Update:  after drying over night the prints look virtually identical - CS2, CS3 and QStudio.  In all cases they are very close matches to what I see on my LCD (calibrated and profiled with Eye-One Display 2).  As Neil mention the ruddy complexion in the color portrait is in the image.
Logged

[span style='font-size:14pt;line-height:100%'][span style='font-family:Arial'][span style='font-family:Geneva'][span style='font-size:8pt;line-height:100%']Regards,
Ron[/span][/span][/span][/span]
marcsitkin
Full Member
***
Offline Offline

Posts: 125


WWW
« Reply #39 on: June 18, 2007, 03:20:58 PM »
ReplyReply

Sean-

I ran your test file on some Ilford Galerie Gloss, which is a material recently added to my media after the last firmware upgrade, and profiled with the APS.

Results were pretty much as I expected. Color matches between QImage and CS3PS when using application controlled color.

Softproofs were very close to the prints.

CS3 Photoshop printing after conversion from your source to Adobe RGB and printed with printer managed color yielded a print slightly lighter and less ruddy on the portrait.

Please send me shipping info and I'll send these out to you.

Marc Sitkin
Logged

Regards,

Marc Sitkin
www.digitalmomentum.com
Pages: « 1 [2] 3 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad