Pages: [1]   Go Down

Author Topic: HP Z3100 CMY-based RIP profiling  (Read 6838 times)

adiallo

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 87
HP Z3100 CMY-based RIP profiling
« on: February 26, 2007, 03:19:31 pm »

With the isues posted here about color gamut in mind I have compared prints here on the Z3100 using
1) the OEM driver with the built-in (non-APS) profiler
2)  the OEM driver with PrintFix Pro profiler
3) StudioPrint RIP (pre-release HP driver) with their GPS profiler

A low-res of one of the test images used in each scenario is attached just to give you an idea of the out-of-gamut colors I've been testing.

Early conclusions here show that there is a strong quality benefit to driving the Z3100 as a CMYK device. Of course this requires a RIP and CMYK linearization and profiling software, but this approach produces accurate and pleasing results on images that are less than satisfactory when printed through an RGB driver pipeline.

If I had to rank things I would put PrintFix Pro in second place and the on-board setup a distant third. Now I will qualify this by saying I have printed numnerous color images that are just fine with the built-in profiler and where the differences with StudioPrint are relatively minor. My biggest issue is in compressed shadow detail with the built-in solution. As you can image some images show this more than others.

Where the CMYK RIP profiling really shines is in images with out of gamut and low luminance reds and to a lesser extent low density yellows and greens. CMYKRGB printers have a lot of inks to be controlled and I think the OEM drivers will improve over time.

And Black and White?
If black and white is your thing you should be very pleased with the built-in profiler. The Vivera ink set in quad K mode gives a great DMax on matte and the tone of the ink is quite pleasing, less warm than the UCK3s. There is some stepping on the shadows, but a gamma  adjustment to the image's working space is a quick fix. Compared to third party quad black output on an Epson, for example, the differences in the Z3100 with on-board calibration/profiling and a gamma adjustment are very slim. (Comparisons with six and seven ink dilutions like those from Cone are another story, which I am still investigating).

So the good news is that it seems likely that issues users are having here with gamut are not endemic to the hardware and ink, and it seems there is certainly room for improvements in drivers and perhaps firmware. When you consider this is essentially HP's v1.0 for large format fine art I think all in all they have done a remarkable job. Usability, judicious use of paper and ink, etc are all positives. I certainly don't expect an RGB profiling solution to beat properly linearized and profiled CMYK options. But I do think that with some adjustments to drivers and firmware a very large number of users can be quite happy with the Z3100 without shelling out for a RIP/CMYK profiler combo.

As an aside, I've made the same test prints on a Canon IPF5000, so far only through the RGB driver, and the Canon does indeed handle out of gamut reds in a much more pleasing manner than the HP built-in profiler. Again the RIP-driven HP is markedly superior. One of my curiosities here was to see how relevant RIPs remain with n-color printers and recent driver improvements. For me the answer is a RIP that allows linearization and CMYK profiling offers demonstrable advantages. Do they cost too much? Yes, but the best of them do deliver the goods.

If anyone on the list would like to compare prints either of their test files or a high-res version of the one here, I'd be happy to send my test file or print yours for comparison. Email me off list.
« Last Edit: February 26, 2007, 04:14:37 pm by adiallo »
Logged
amadou diallo
[url=http://blogfiftygreat

Panascape

  • Full Member
  • ***
  • Offline Offline
  • Posts: 214
HP Z3100 CMY-based RIP profiling
« Reply #1 on: February 26, 2007, 03:56:25 pm »

Unfortunately the target you are printing is not really going to show the worst of the problem. A test chart with saturated reds like the one floating around with the strawberries shows the problem immediately. Looking at the saturation of the red in the colour ramps the profile you are using will have the same problem we are seeing with the strawberry test chart.
Logged

adiallo

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 87
HP Z3100 CMY-based RIP profiling
« Reply #2 on: February 26, 2007, 04:13:18 pm »

The attachment in my post is not a scan from a print. It is a digital file in ProPhoto RGB. If whoever posted the strawberry file you are referring to wants to email it to me off list I'd be happy to print it out through the RIP and compare results.
Logged
amadou diallo
[url=http://blogfiftygreat

Panascape

  • Full Member
  • ***
  • Offline Offline
  • Posts: 214
HP Z3100 CMY-based RIP profiling
« Reply #3 on: February 26, 2007, 04:22:16 pm »

5mb file ok?
« Last Edit: February 26, 2007, 04:24:16 pm by Panascape »
Logged

adiallo

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 87
HP Z3100 CMY-based RIP profiling
« Reply #4 on: February 26, 2007, 04:36:21 pm »

Sounds good. Thanks.
Logged
amadou diallo
[url=http://blogfiftygreat

Christopher

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1499
    • http://www.hauser-photoart.com
HP Z3100 CMY-based RIP profiling
« Reply #5 on: February 26, 2007, 07:05:43 pm »

Quote
Sounds good. Thanks.
[a href=\"index.php?act=findpost&pid=103339\"][{POST_SNAPBACK}][/a]

Yes I'm really intrested in the results you are getting for saturated reds.
Logged
Christopher Hauser
[email=chris@hauser-p

marty m

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 335
    • http://
HP Z3100 CMY-based RIP profiling
« Reply #6 on: February 26, 2007, 09:46:08 pm »

Quote
A low-res of one of the test images used in each scenario is attached just to give you an idea of the out-of-gamut colors I've been testing.

If I had to rank things I would put PrintFix Pro in second place and the on-board setup a distant third. Now I will qualify this by saying I have printed numnerous color images that are just fine with the built-in profiler and where the differences with StudioPrint are relatively minor.

But I do think that with some adjustments to drivers and firmware a very large number of users can be quite happy with the Z3100 without shelling out for a RIP/CMYK profiler combo.

[a href=\"index.php?act=findpost&pid=103324\"][{POST_SNAPBACK}][/a]

Thanks very much for posting your results and conclusions.  It is greatly appreciated.

(1)  You posted only one test image, not three, so there is no basis for comparing.

(2)  I'm still interested in any reports that compare the APS to the on-board setup. Such a direct comparison has not been made.

The other thread contained discussion about the contents to the box, but no one offered a direct comparison.  Does the $1000 for APS make any difference?  

Setting aside the issue of the number of patches, is the APS Gretag software more sophisticated and more capable than the on-board Gretag setup?  

Does doubling the number of patches matter?  (The overall conclusion appears to be no.)

I'm not eager to drop $1000 on APS, but the StudioPrint RIP appears to cost $1300 without the GPS option.  And Image Print for a 24" printer is $1500.  I just don't need the features of a RIP.  

If APS is worth it, either because of an improved engine, or the improved software and more patches -- if those who bought it report that it is superior to the on-board setup -- I'd consider it.  

But I just haven't seen any concrete reports.

(3)  As an advanced amateur, my conclusion (maybe this is wishful thinking) is that the Z3100 will be satisfactory for the vast majority of landscapes that I print.  

I'm also hoping that HP will issue new firmware and software to address these issues in the near future.

There are others in the forum, and on the web, who appear to be perfectly happy with the Z3100.

So I have kept my order to replace an Epson 4000 with the Z3100.  Once I get it, I'll give you all a report.  Including prints made with very saturated fall colors.
Logged

BlasR

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 760
    • http://BMRWorldPhotos.com
HP Z3100 CMY-based RIP profiling
« Reply #7 on: February 28, 2007, 07:30:52 am »

Adiallo, do you got a chance to look the file the Robert send you?

Thanks




 Michael, are you going to take a second  look to the hp z3100 printer in see if working ok for you?  If you do may ask when that going happen.

Thank you

BlasR



Quote
The attachment in my post is not a scan from a print. It is a digital file in ProPhoto RGB. If whoever posted the strawberry file you are referring to wants to email it to me off list I'd be happy to print it out through the RIP and compare results.
[a href=\"index.php?act=findpost&pid=103335\"][{POST_SNAPBACK}][/a]
Logged
BlasR
  [url=http://www.BMRWORLDPHOTOS.CO

adiallo

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 87
HP Z3100 CMY-based RIP profiling
« Reply #8 on: February 28, 2007, 01:01:55 pm »

Haven't gotten a file yet.
Quote
Adiallo, do you got a chance to look the file the Robert send you?

Thanks

BlasR
[a href=\"index.php?act=findpost&pid=103723\"][{POST_SNAPBACK}][/a]
Logged
amadou diallo
[url=http://blogfiftygreat

Christopher

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1499
    • http://www.hauser-photoart.com
HP Z3100 CMY-based RIP profiling
« Reply #9 on: February 28, 2007, 01:47:12 pm »

Quote
Haven't gotten a file yet.
[{POST_SNAPBACK}][/a]


[a href=\"http://www.pixentral.com/show.php?picture=1pSIRzke8wxxM6StJeTs7NOVV13lK0]Here[/url] it is.
Logged
Christopher Hauser
[email=chris@hauser-p

Panascape

  • Full Member
  • ***
  • Offline Offline
  • Posts: 214
HP Z3100 CMY-based RIP profiling
« Reply #10 on: February 28, 2007, 02:21:59 pm »

Hi Adiallo, I mailed the files to the address on your website on Monday.
« Last Edit: February 28, 2007, 02:22:15 pm by Panascape »
Logged

adiallo

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 87
HP Z3100 CMY-based RIP profiling
« Reply #11 on: March 01, 2007, 05:16:13 pm »

Robert, et al,
I just printed out your strawberry image, among others with the 4.1 firmware (recalibrated and reprofiled the paper). Reds I had an issue with in my own test files are greatly improved. They went from unusable to pleasing. Still not so accurate, compared to RIP output but a big improvement. On the strawberry image, I have to say the print looks pleasing to me. Nothing that would throw up a red flag. Softproofing with the new profile I do get less saturation than your supplied Epson comparisons but on paper I have no real objection, likely because I don't have your Epson prints here. Over the weekend I'll relinearize with the RIP and print your test file. Also curious to see if the firmware makes any difference outside the OEM driver and profiler.
Logged
amadou diallo
[url=http://blogfiftygreat

Greg_E

  • Full Member
  • ***
  • Offline Offline
  • Posts: 113
HP Z3100 CMY-based RIP profiling
« Reply #12 on: March 03, 2007, 07:03:41 pm »

It comes as no real surprise to me that you can get better results out of the RIP. It may not be all about the extra ink control, some of it may be the color engine used in the RIP, or the quality of the profile. Amadou, did you buy the RIP or is it the demo? Also what spectro are you using? If the spectro is supported, it should be easy enough for me to gererate a CMYK profile from Xrite/Monaco Profiler 4.8 for you. All you would need in the free utility Colorport from Xrite. All the Profiler targets are in the application, so you would just need to print the target, measure, and send the measurement file to me. I would advise using a spectro with a UV filter as Profiler does not compensate for any OBA in the paper or inks. It's probably easiest to grab me at my Yahoo email right now, you've used it to ask me about the Epson v750 back in October-November. If you don't have a spectro with UV filter, I can send you a link to profiler CMYK targets and measure the prints for you. Since I know all this takes time, that's why I asked if you own the RIP as the demo would run out by the time everything is done.

I'll send an email to your yahoo address just so you have my address.



Also in the link to that image above, what is the color space?
« Last Edit: March 03, 2007, 07:07:22 pm by Greg_E »
Logged

adiallo

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 87
HP Z3100 CMY-based RIP profiling
« Reply #13 on: March 03, 2007, 08:01:15 pm »

Quote
Also in the link to that image above, what is the color space?
[a href=\"index.php?act=findpost&pid=104465\"][{POST_SNAPBACK}][/a]
ProPhotoRGB.
Will follow up with you about the rest offlist.
« Last Edit: March 03, 2007, 08:01:26 pm by adiallo »
Logged
amadou diallo
[url=http://blogfiftygreat

Jim Cole

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 377
    • Jim Cole Photography
HP Z3100 CMY-based RIP profiling
« Reply #14 on: March 18, 2007, 03:01:59 pm »

Amadou,

Have you made any further discoveries here? I am seriously looking at StudioPrint to drive my Z3100. I just posted some questions in this forum about this software, it's use of the onboard spectro and the number of patches the GPS profiler uses to build targets.

I would appreciate any further feedback you might have on this RIP.

Thanks,
Jim
Logged
Jim Cole
Flagstaff, AZ www.jimcolephoto.

adiallo

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 87
HP Z3100 CMY-based RIP profiling
« Reply #15 on: March 19, 2007, 05:04:09 pm »

Jim,
Having run a number of prints and comparing StudioPrint output to the HP driver output (without APS, which I don't own) I am seeing:
1. On b/w prints, more accurate shadow detail when using a quad K print envronment in StudioPrint
2. On general color images, like portraits, PDI test images, slightly cleaner and more accurate colors with StudioPrint
3. On difficult color gradations, ie saturated colors transitioning to black, Granger rainbows, etc the nod goes to the HP driver. It is not more accurate, but more pleasing, as StudioPrint shows some posterizing in the transitions.

Currently, StudioPrint does not run through the built-in spectro. My understanding is that it is in the plans, but can't say how far out that is. If I build a quad K environment in StudioPrint I can hit very subtle shadow separation that get blocked by the HP driver. We're talking very subtle shadow detail, and to be honest, it took a number of images before I hit this limitation with the HP driver running the 4.1.02 firmware. Overall, with the papers optimzed in the newest firmware, I think the HP driver and profiler does a very good job for an RGB profiling solution. At least now it compares much better against PrintFix Pro, which is surprisingly good, considering the price point.
There are benefits to driving a CMYK printer as a CMYK device, but there is a lot of work involved too. You've got to know how to linearize, set ink limits, and then have the equipment to measure fairly large patch counts for color profiles. Is it worth it? On some images, you say yes, definitely. On others, the differences are so close, you say why bother? And in a few instances I am getting more pleasing results from the HP driver. My sense is that all of the RIP companies have their work cut out for them with the CMYKRGB printers. A much different beast that the standard 4c inksets. I'm still working through some issues with StudioPrint profiling behavior. Anyone who's complaining about feeling like a guinea pig with the HP driver/profiling would not want to all of a sudden jump into a RIP right now. RIP support is still limited to very few vendors. A few weeks or a couple of months from now, it will likely be a different story.
I think that for a large majority of images the HP driver/profiler does a commendable job. StudioPrint offers some advantages today, but is not the magic bullet for all types of images yet. I'm still reserving judgement as to the route I will ultimately take with driving the Z. Unless you are completely unhappy with the HP OEM solution I would give it a bit of time to settle out.
Logged
amadou diallo
[url=http://blogfiftygreat

Jim Cole

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 377
    • Jim Cole Photography
HP Z3100 CMY-based RIP profiling
« Reply #16 on: March 19, 2007, 05:38:19 pm »

Amadou,

Thanks for your detailed explanation of the comparative benefits of the HP and StudioPrint profiling solutions. The only trouble I'm experiencing now is with the black transitions to deep colors in shadow areas on HM PhotoRag. The transitions are abrupt and nothing close to smooth. I have calibrated and profiled the paper as both Thick Fine Art and as HP Smooth Fine Art which was the obvious better of the two. I am finally supposed to receive my HP Smooth Fine Art today so I can find out if it transitions better than the HM PhotoRag. If this paper solves the shadow transition issue, then I should be able to do quite well with the Smooth Fine Art and Pro Satin papers for all of my work...at least for now. We'll see what happens with the evolution of the HP APS solution. Right now it is of limited use.

However, the other reason I was interested in StudioPrint for was the multiple image layout options to allow me to utilize 24" and 44" roll papers better. I wouldn't even think of buying StudioPrint until it made use of the built in spectro on the 3100. I really do not want to invest in another one. If and when they do, the ink mixing may be even better and would probably make it worth the purchase price for me. I wish QImage allowed you to define roll paper sizes (width, but no length), but they do not. That would be a good interim fix for me as far as paper utilization goes.

Again, thanks for your time and knowledge.

Jim
Logged
Jim Cole
Flagstaff, AZ www.jimcolephoto.
Pages: [1]   Go Up