« Previous topic | Next topic »  
Reply to topic  
 First unread post  | 20 posts | 
by DOglesby on Mon Sep 23, 2013 10:15 am
User avatar
DOglesby
Lifetime Member
Posts: 979
Joined: 19 May 2008
Location: North Carolina
Member #:01155
I've read that your printer manufacturer's paper with their icc profiles is going to produce a superior print than a third party's paper using the third party's icc profile.  I've witnessed this a couple of times, most notably with Hahnemuhle paper on my Epson 4900.  The print, as compared to the Epson paper, was sub par.  The color was clearly inferior.  I'm wondering why this is. 

I'm assuming the larger paper companies (like Hahnemuhle) are using high quality profile equipment with a significant number of color patches.  So, why aren't these good enough to get perfect color?  Afterall, when I profile a test chart and create my own ICC profile the color is perfect.  At first I was wondering if it was because there could be variances from printer to printer so my personal profile would account for this.  But, the printer manufacturers (like Epson) would have the same problem and their profiles are fabulous.  I understand the printer manufacturer's have the benefit of a better understanding of their printers but is it really as simple as that?
Cheers,
Doug
 

by Royce Howland on Mon Sep 23, 2013 10:34 am
User avatar
Royce Howland
Forum Contributor
Posts: 11719
Joined: 12 Jan 2005
Location: Calgary, Alberta
Member #:00460
I've often wondered myself why paper profiles are hit-and-miss. Some vendors are better than others, on average; like you Doug, I've found Hahnemuhle profiles to be often really sub-par. But even with a given vendor sometimes the results are pretty variable. The only thing I can come up with at the end of the day is that the people producing inferior profiles (whether internal staff or external resources) simply don't know or don't care how good a job they do.

It's pretty straight forward to produce a profile that's at least "good" given halfway decent equipment, a person with some basic knowledge, paying even minimal attention to the process, and doing some simple testing of the resulting profile. The fact that poor profiles continue to be made available indicates that at least one of those 4 things (if not all of them) failed to happen. Understanding the printer itself has little to do with it since the profiling process really doesn't rely on any knowledge of the printer. So the printer vendors have no inherent advantage, other than perhaps caring more that their own papers produce really good looking results on their own printers to justify people sticking with the brand.

Producing exceptional profiles (i.e. better than "good") requires more skill, care, and better tools. So I don't really expect exceptional results from generic downloaded profiles. But I'd like them to be at least as good as I could produce myself in less than an hour with a $500 ColorMunki Photo, and frequently they're not...
Royce Howland
 

by Royce Howland on Mon Sep 23, 2013 10:40 am
User avatar
Royce Howland
Forum Contributor
Posts: 11719
Joined: 12 Jan 2005
Location: Calgary, Alberta
Member #:00460
Oh, printer-to-printer variation could explain some of it; this is why we recommend getting custom profiles for your own printer. But this doesn't explain why some paper vendor's generic profiles for a given printer model are consistently better than some other paper vendor's profiles for that same model. Epson printers in particular are fairly well linearized at the factory, with respect to each other, which means generic profiles should demonstrate less printer-to-printer variation. There will still be a certain amount of it, but I don't think it's enough to explain why some profiles are so bad.

For example, I recently tested out the new Ilford Gold Mono Silk paper which I had high hopes for as a high gloss, bright white companion media for Ilford Gold Fibre Silk which I love. I use the generic Ilford GFS profile on my Epson 4880 because it has always produced very good results; and when I've examined the profile in my diagnostic utilities I've seen nothing to make me want to make a custom version, though I could perhaps do a few % of quality better that way.

But when I tried the new GMS paper on my Epson 4880 with the generic profile, the results were horrid. To the point I thought I must be doing something wrong in my workflow. Nope, it was the profile. I took an hour to build my own GMS profile with the ColorMunki Photo and it nailed the resulting test prints very close to what I'd been seeing along from GFS. So Ilford produced one hit and one miss on the same printer model...
Royce Howland
 

by DOglesby on Mon Sep 23, 2013 11:15 am
User avatar
DOglesby
Lifetime Member
Posts: 979
Joined: 19 May 2008
Location: North Carolina
Member #:01155
Interesting. Thanks for you perspective, Royce.
Cheers,
Doug
 

by signgrap on Mon Sep 23, 2013 1:55 pm
User avatar
signgrap
Lifetime Member
Posts: 1776
Joined: 1 Sep 2004
Location: Delaware Water Gap, PA
Member #:00424
OK Royce, What's the solution?
Dick Ludwig
 

by Trev on Mon Sep 23, 2013 2:50 pm
User avatar
Trev
Forum Contributor
Posts: 626
Joined: 20 Oct 2008
Location: New Zealand
Royce mentioned that, do your own or have custom profiles made.
Trevor Penfold
Website http://www.trevorpenfold.com
Facebook https://www.facebook.com/trevorpenfoldphoto
 

by Royce Howland on Mon Sep 23, 2013 4:41 pm
User avatar
Royce Howland
Forum Contributor
Posts: 11719
Joined: 12 Jan 2005
Location: Calgary, Alberta
Member #:00460
Yes, exactly. If you do a lot of paper experimentation, a ColorMunki Photo might be a good investment. It makes good printer profiles relatively easily; I'm pleasantly surprised how good its results are given the simplicity of using it. If you do fewer profiles, a custom profiling service usually costs $30 - $50 per paper type. But you do have to wait for some period of days or weeks while things are being emailed and snail-mailed around. Finding a good service may be challenging, though; search past discussion threads here for a mention of a few custom profiling services...
Royce Howland
 

by Bill Chambers on Tue Sep 24, 2013 11:52 am
User avatar
Bill Chambers
Forum Contributor
Posts: 4015
Joined: 8 Feb 2006
Location: Milton, Florida
As for companies doing profiles, I have used Chromix in the past and was very pleased with the results, although Chromix is more expensive ($99)than most. I don't have personal experience with Michael Gordon's profiles but I've heard from several (4 or 5) friends that his profiles are excellent and very modestly priced. Michael's website is here - http://www.michael-gordon.com/page/customRGBprofile/.
Please visit my web site, simply nature - Photographic Art by Bill Chambers
Bill Chambers
Milton, Florida
 

by ronzie on Sat Sep 28, 2013 2:24 am
User avatar
ronzie
Forum Contributor
Posts: 459
Joined: 26 May 2011
Location: 40 miles North of Minneapolis, MN, US
Here's a tip on using the ColorMunki Photo:

Be sure the latest drivers are used. For best accuracy let the first target dry in a light protected area several hours before measuring. The latest driver allows you to close the program right after the reference target is printed and resume it at the measuring process in the next session.

For whatever reason they do not allow this for drying time for the correction patch set. A work around is to start at the same resume point where it will print the correction patch set but when you get to the print dialog just cancel it and hit next and it will take you to the measuring point of the correction patch set.

I use dye inks and it takes a long time for them to stabilize and not react to light, much more than the ten minute clock in the CMP countdown. I don't think you want the CMP software running on your workstation in the background for several hours while the ink stabilizes.

Another tip on a Color Munki Photo is in the intial sensor calibration process, let the process idle a few minutes to let the LED stabilize (it will illuminate in cycles to heat up) before you run the calibration.

A final hint on the CMP and other X-Rite products:

If you are a Mac user be aware that the CMP software will not run on the latest Mac O/S nor will it be upgraded per X-Rite.

https://xritephoto.com/ph_product_overview.aspx?ID=788&Action=Support&SupportID=5084 for Win users and

https://xritephoto.com/ph_product_overview.aspx?ID=788&Action=support&SupportID=5041 for Mac users.
 

by Royce Howland on Sat Sep 28, 2013 10:51 am
User avatar
Royce Howland
Forum Contributor
Posts: 11719
Joined: 12 Jan 2005
Location: Calgary, Alberta
Member #:00460
The info I have indicates ColorMunki Photo software runs on Mac OSX at least as far as 10.8.2, although there are some installation problems to work through. Are you saying it doesn't run at all on 10.8.5? That would be unfortunate. X-Rite has not upgraded the Munki software since 2009. That's a long time, and does make one wonder if the product is on the way out. Hmm... I don't really see an equivalent product in their current lineup. Too bad if it goes away and isn't replaced by something equivalent that's as easy to use with good quality results.
Royce Howland
 

by ronzie on Sat Sep 28, 2013 4:56 pm
User avatar
ronzie
Forum Contributor
Posts: 459
Joined: 26 May 2011
Location: 40 miles North of Minneapolis, MN, US
Royce Howland wrote:The info I have indicates ColorMunki Photo software runs on Mac OSX at least as far as 10.8.2, although there are some installation problems to work through. Are you saying it doesn't run at all on 10.8.5? That would be unfortunate. X-Rite has not upgraded the Munki software since 2009. That's a long time, and does make one wonder if the product is on the way out. Hmm... I don't really see an equivalent product in their current lineup. Too bad if it goes away and isn't replaced by something equivalent that's as easy to use with good quality results.
I'm not saying, X-Rite is. I'm a Win user but I still receive compatibility issue notes from them for all platforms. I just pointed to the compatibility matrices on the X-Rite site. However it appears that some parts of the matrices are not complete. For example, under the current CM Photo specs it states:

"
Macintosh
  • Intel processor
  • Mac® OSX 10.6, 10.7, or 10.8 (with the latest updates installed)
  • 1GB RAM
  • 500 MB of free disk space
  • Powered USB Port
  • 1024 x 768 or higher display resolution with 16-bit video card
  • User must have Administrator rights to install and uninstall the application
  • Internet connection for software updates
  • Broadband Internet connection required for online services*
  • Network card required
Windows
  • Microsoft® Windows® XP 32 bit, Microsoft® Windows Vista® 32 or 64 bit, Microsoft Windows 7® 32 or 64 bit or Microsoft Windows 8® 32 or 64 bit. All operating systems should have latest Service Pack installed
  • Intel[sup]®[/sup] Core 2 or AMD Athlon™ 64 X2 or better CPU
  • 1GB RAM
  • 500 MB of free disk space
  • Powered USB Port
  • 1024 x 768 or higher display resolution with 16-bit video card
  • User must have Administrator rights to install and uninstall the application
  • Internet connection for software updates
  • Broadband Internet connection required for online services*
  • Network card required
So if the only issue was Rosetta being dumped from Mac 10.7 onward, then it appears it should also work on 10.8. They should update their Matrices.

For Win users, Win 8 is stated to work.

It also looks like I1 Publish and I1 Photo Pro 2 products are current as well. (I only quickly looked for both monitor and printer calibration products. In the Publish products some have CMYK profiling as well.)

This Mac product matrix is only good up to 10.7 but apparently applies to 10.8 and I quote:

"Apple has announced that the new OS 10.7 Lion will no longer support software that relies on Rosetta. This directly affects X-Rite's retired i1Match, ProfileMaker and MonacoPROFILER applications. However, the current i1Profiler and ColorMunki software does not rely on Rosetta and is fully compatible with Apple OS systems from 10.5.8, 10.6 and the new 10.7 Lion."

BTW: As far as updating the CM Photo app, a long time ago I requested the resume feature that is available after printing the standard target ( in the software 1.11 version) be also available after printing the correction color patch target explaining about the long drying time needed for dye ink stability and they replied it would be considered. However as noted there have been no new releases since 1.11.

They have released new products though such as the I1 Publish, etc., and that is where they are concentrating their efforts aiming toward new hardware and apps for them.

Sorry for misleading the thread about this but they should have updated the matrices.
 

by rnclark on Mon Oct 14, 2013 10:58 am
rnclark
Lifetime Member
Posts: 864
Joined: 7 Dec 2010
Member #:01978
Note that color is dependent on the light illuminating the print. One reason for "hit or miss" print results can often be due to the light on the print. For example, if the ICC profile for a paper is made using a certain color temperature tungsten lamp, then the print viewed under fluorescent lighting, the colors will appear off. Color perception also varies with the intensity of the light. So with varying light level, spectral response of the light and spectral response of the inks in the paper as well as the underlying paper spectral response, there are a lot of variables for how the colors appear, even with a perfect icc profile. Check with the manufacturer of the icc profile to see if they specified a light source. If you intend to view a print under a different light source, you need an icc profile for the print under that light source.

Roger
 

by bjs on Mon Oct 14, 2013 12:22 pm
bjs
Forum Contributor
Posts: 362
Joined: 18 Jan 2004
Location: Vancouver, BC
rnclark wrote:Check with the manufacturer of the icc profile to see if they specified a light source.
AFAIK, every printer profile lists it's assumed illumination in one of its header fields (use the ICC's free tool "ICC Profile Inspector").

All the common paper profiles I've seen assume X=0.9642, Y=1.0, Z=0.8249 in CIE XYZ space.   The correlated color temperature equivalent is 5000K.
 

by bjs on Sun Nov 10, 2013 12:38 pm
bjs
Forum Contributor
Posts: 362
Joined: 18 Jan 2004
Location: Vancouver, BC
ronzie wrote:For whatever reason they do not allow this for drying time for the correction patch set. A work around is to start at the same resume point where it will print the correction patch set but when you get to the print dialog just cancel it and hit next and it will take you to the measuring point of the correction patch set.
It may not be that simple.  The Munki software forces you to rescan the first patch and then it generates a new second patch set.  If this new patch set is at all different (e.g. because you scanned the now day old first patch set and got slightly different results) then cancelling out and using your original "second" patch set will generate a worse profile.

Have you asked Xrite if this works?   Or measured the two "second" patch sets under a variety of conditions to verify they are always the same?  Otherwise it sounds a bit risky to me.

The Munki's handling of printing and rescanning patches is one of the most boneheaded, unfriendly things I've seen in any software for a long time.   Probably the single biggest negative I have with the package.   It's enough to make one learn Argyll.
 

by ronzie on Sun Nov 10, 2013 10:51 pm
User avatar
ronzie
Forum Contributor
Posts: 459
Joined: 26 May 2011
Location: 40 miles North of Minneapolis, MN, US
I think you are misunderstanding me. I do not generate two first or second patch sets. Be sure you are using version 1.11

I print the first patch set, then exit as allowed by option, and resume at a later time using the aged first patch set.

After the first patch set is measured after resuming I let it print the second patch set and force the application to exit.

After the second patch set has aged I again enter at the same resume point but when the print dialog comes up to generate the second patch set (again) I cancel out of the print dialog and hit next without printing another second patch set. This takes me to measure the second (aged) patch set.

I'm using Win XP to do this and have not had any problems.

I contacted X-Rite about two years ago about adding an optional exit and resume for the second patch set explaining about the lengthy aging time for dye ink stability. I was told it might be considered for the next version but there was no version beyond 1.11 for Win or Mac.

I do the same when optimizing.
 

by bjs on Mon Nov 11, 2013 1:33 am
bjs
Forum Contributor
Posts: 362
Joined: 18 Jan 2004
Location: Vancouver, BC
Yes, I misunderstood you. I thought you were initially generating non-aged versions and then switching. Since you are only generating one set and both are off aged versions it should be fine. A good idea actually given the current limitations.
 

by Royce Howland on Mon Nov 11, 2013 5:09 pm
User avatar
Royce Howland
Forum Contributor
Posts: 11719
Joined: 12 Jan 2005
Location: Calgary, Alberta
Member #:00460
That's pretty much how I do it with the ColorMunki Photo as well. It's a pity they never updated the Munki software after 2009... it's a neat product that produces really good results from a deceptively simple process. With a few simple updates, it could be more usable. And of course they could extend the functionality quite a bit if they so chose, to make it more of an upper mid-tier product for those balking at jumping all the way up to an i1 Pro 2 kit.
Royce Howland
 

by ronzie on Tue Nov 12, 2013 7:29 pm
User avatar
ronzie
Forum Contributor
Posts: 459
Joined: 26 May 2011
Location: 40 miles North of Minneapolis, MN, US
Royce Howland wrote: ..snip...  With a few simple updates, it could be more usable. And of course they could extend the functionality quite a bit if they so chose, to make it more of an upper mid-tier product for those balking at jumping all the way up to an i1 Pro 2 kit.
I think the marketing folks want a big separation from the upper tiers.

If you add features then they want you to purchase a different product at a much higher price.
 

by Royce Howland on Tue Nov 12, 2013 8:35 pm
User avatar
Royce Howland
Forum Contributor
Posts: 11719
Joined: 12 Jan 2005
Location: Calgary, Alberta
Member #:00460
Yeah, bad thinking of marketing geniuses...
Royce Howland
 

by DOglesby on Sun May 17, 2015 9:12 am
User avatar
DOglesby
Lifetime Member
Posts: 979
Joined: 19 May 2008
Location: North Carolina
Member #:01155
As an update to this year-and-a-half-old post, I've been using Hahnemuhle, Moab and Ilford profiles with their paper in my Epson 9900 and the prints are turning out amazing.  I don't know if this is explainable by the fact that the Epson 9900 naturally results in greater care in the making of the profiles, or that the profiles just naturally improved over time.  (My 4900 has died so I obviously cannot test out the latest 4900 profiles).  Regardless, I am extremely happy with these third party manufacturers' profiles.  So much so that I don't see a need to make custom profiles myself.  Has anyone else noticed an improvement?

By the way, Royce - I see why you like Ilford Galerie Gold Fibre Silk.  Gorgeous paper.  Not as much texture as the Hahnemuhle FA Baryta but I think I may actually prefer the GFS.  
Cheers,
Doug
 

Display posts from previous:  Sort by:  
20 posts | 
  

Powered by phpBB® Forum Software © phpBB Group