Moderator: E.J. Peiker

All times are UTC-05:00

  
« Previous topic | Next topic »  
Reply to topic  
 First unread post  | 19 posts | 
by Rocky Sharwell on Sun Dec 08, 2013 6:57 pm
Rocky Sharwell
Lifetime Member
Posts: 2994
Joined: 21 Aug 2003
Member #:00054
Having read this thread dealing with less than ideal manufacturer paper profiles:
http://www.naturescapes.net/forums/view ... 4&t=237126

I was wondering which papers users have liked and were satisfied with stock profiles.  Thanks for any thoughts and advice...
Rocky Sharwell
 

by Royce Howland on Mon Dec 09, 2013 4:34 pm
User avatar
Royce Howland
Forum Contributor
Posts: 11719
Joined: 12 Jan 2005
Location: Calgary, Alberta
Member #:00460
It depends on your standards. :) I suspect some people are out there who aren't even printing with profiles at all, and yet are satisfied. Whereas I've looked at many stock paper profiles and honestly have only ever thought a few were satisfactory, with respect to my Epson 4880. Several of the Epson paper profiles are good, but I rarely print on Epson papers. Harman Gloss FB Al had a good profile, but the paper is no longer available. The replacement for it is called Harman by Hahnemuhle Gloss Art Fibre, and the generic profile there is poor, as seems to be the case for many Hahnemuhle profiles I've looked at.

Papers that I like is a different question. I pick papers I want to use first & foremost, then look at the generic profiles. If I'm not satisfied (which is almost always) I either get custom profiles created or make them myself with my ColorMunki. The ColorMunki is the only entry-level printer profiling tool I ever used that works, and it works surprisingly well considering how easy it is to use. With it, I can make profiles that are better than most generic ones. The Munki is not perfect, though, and sometimes there's a case for getting a custom profile made by somebody that really knows their stuff and uses high-end gear...
Royce Howland
 

by signgrap on Mon Dec 09, 2013 5:03 pm
User avatar
signgrap
Lifetime Member
Posts: 1776
Joined: 1 Sep 2004
Location: Delaware Water Gap, PA
Member #:00424
Royce Howland wrote:. . . 
Papers that I like is a different question. I pick papers I want to use first & foremost, then look at the generic profiles. If I'm not satisfied (which is almost always) I either get custom profiles created or make them myself with my ColorMunki. The ColorMunki is the only entry-level printer profiling tool I ever used that works, and it works surprisingly well considering how easy it is to use. With it, I can make profiles that are better than most generic ones. The Munki is not perfect, though, and sometimes there's a case for getting a custom profile made by somebody that really knows their stuff and uses high-end gear...
Royce, I have used mostly Epson Luster and have found that I can match my NEC SpectraView very closely with the Epson stock Luster profile. Recently I have been trying some Breathing Color paper and have found the profile a bit lacking i.e. a bit too saturated and too warm. Would a ColorMunki be a tool to consider to correct these problems? Or would a custom profile be the way to go? I tend to be fussy about color.
Dick Ludwig
 

by Royce Howland on Mon Dec 09, 2013 6:15 pm
User avatar
Royce Howland
Forum Contributor
Posts: 11719
Joined: 12 Jan 2005
Location: Calgary, Alberta
Member #:00460
If you only want to get a few custom profiles made, it's probably more economical to have somebody else (who is good) do them for you. I make profiles all the time, so at ~$450 the cost of the Munki was easy to justify. If you only need 2 or 3 it may be tougher to justify the expense.

A ColorMunki profile could produce a better result with a particular Breathing Color paper... assuming it is the BC generic profile that's the issue and not something else. :)

The profiling procedure with the Munki produces decent results but where the profiles are lacking is in the deepest shadow tones. There's no way to directly edit profiles with the Munki software; you can iteratively refine existing profiles with a clever process but if the shadows aren't working the greatest they'll probably stay that way through the iterative profile refinement. A profile made with high end tools by somebody who's good at it should produce better shadow tone results than the Munki can do.

That's one example of why I said there's still a case for getting really good 3rd party custom profiles made, even for somebody with a ColorMunki or something else like it. It's a question of good, better, best. Sometimes you may need or want "best"...
Royce Howland
 

by RoyH on Wed Dec 11, 2013 11:48 am
RoyH
Forum Contributor
Posts: 31
Joined: 13 Mar 2012
Location: Denver, CO
To the OP,

My printer is the Epson 3880 and my paper of choice at this time are the papers from Moab and I am very happy with the results from the stock profiles. My everyday paper is the Moab Lasal Exhibition Luster, but I have also use the Entrada Rag bright and natural with the stock profiles also. It addition to the Moab papers I do use papers from Epson and Red River on occassion and mostly happy with the Epson profiles but I have had some issues with the Red River profiles.

To Royce,

I wonder if you would give some explanation of exactly what you look for when you first get a new paper and test it with the stock profile and what areas do you find the most issues with?

Some background, before switching to a career as a software engineer about 15 years ago I spent the previous 17 years in professional photo-finishing. I worked for a few years as a custom printer making custom prints from 5x7 inches to 8x12 feet, then moved into the professional portrait/wedding lab area, working both in labs and for equipment manufacturers. During this time I worked with engineers developing lab calbration procedures and teaching these procedures to labs and worked for several major pro labs overseeing all areas of the production including final color-correction and quality control. Over those years I developed a pretty good eye for evalulating prints visually. I understand, at least in the old days of chemical photographic processes, using reflective and transmission color meters to calibrate for film and print processing. So I am curious in what do you look for visually in your print tested with new papers and just how much of a change are you able to visually achieve with your custom profiles.

Thanks in advance.

Roy
Roy H
[url]http://www.reflectivephoto.net[/url]
 

by Rocky Sharwell on Thu Dec 19, 2013 10:09 pm
Rocky Sharwell
Lifetime Member
Posts: 2994
Joined: 21 Aug 2003
Member #:00054
As luck would have it I forgot about several sampler boxes of paper purchased from Atlex. I will experiment after the holiday.
Rocky Sharwell
 

by Royce Howland on Thu Dec 19, 2013 10:27 pm
User avatar
Royce Howland
Forum Contributor
Posts: 11719
Joined: 12 Jan 2005
Location: Calgary, Alberta
Member #:00460
I haven't looked at many Moab papers for quite some time. I believe I know who creates the profiles for them, or at least who used to do it anyway, and if it's still the same person they should be fine. :)

Roy, essentially the profile determines (or at least heavily influences) qualities such as Dmax (maximum ink density), contrast, greyscale neutrality / linearity, colour gamut, and colour cast. I look at all of these qualities in the paper's generic profile using some analysis tools before printing anything. Very often I will see some things that could be improved. Surprisingly often I will see things that are poor or outright broken.

For example, there's somebody I'm helping through some printing workflow matters right now. We've been looking at some printer/paper profiles in detail and finding issues with each one so far. The generic profile for Ilford Gold Fibre Silk appears to have results from the Perceptual rendering intent that are identical to Relative Colorimetric. This is unusual, and though not "wrong" as such it's definitely not optimal given what a person might have as an expectation for how Perceptual would do things differently than Relative Colorimetric. This profile has some significant non-uniformity in the green hues as well.

Another paper looked at was Hahnemuhle Photo Rag Pearl. It's got some greyscale neutrality issues. And in the Perceptual rendering intent, it actually potentially produces some hues such as deep yellow and blue that are MORE saturated than the colours coming from the source image colour space (e.g. Adobe RGB). That's just plain broken behavior, IMO.

I recently evaluated the new (ish) Ilford Gold Mono Silk paper on my Epson 4880. I found the profile there looked odd for greyscale neutrality. Sure enough, B&W prints have a terrible magenta cast to them, making the generic profile totally unusable for B&W printing.

I could go on with many more examples. Most of these issues can be corrected even with a basic profile generated by the ColorMunki Photo package. It has some challenges in the deep shadow tones but is otherwise amazingly capable for something so simple to use. More advanced profile generation approaches using high-end kit like the i1 Pro 2 would be able to produce better profiles including in the shadow tones.
Royce Howland
 

by RoyH on Thu Dec 26, 2013 10:30 am
RoyH
Forum Contributor
Posts: 31
Joined: 13 Mar 2012
Location: Denver, CO
Thanks for the information Royce!
Roy H
[url]http://www.reflectivephoto.net[/url]
 

by RoyH on Tue Jan 07, 2014 11:06 am
RoyH
Forum Contributor
Posts: 31
Joined: 13 Mar 2012
Location: Denver, CO
Royce’s comments in this thread intrigued me so I wanted to come back and update this thread with where it has led me.

Being that I bought an Epson 3880 a couple of years ago because I wanted control over my prints, and having been very satisfied with the results using stock profiles, based on Royce’s comments it seems the next logical step of control over one’s prints is the creation of custom profiles. After further research over the last few weeks and finding that devices to create one’s own profiles are now reasonably priced I decided to jump in and begin to create my own profiles.

I elected to go with the SpyderPrint from Datacolor instead of the popular Colormunki. This decision was primarily motivated by the fact that I have been a SpyderPro user for some time and was already getting ready to upgrade to the Spyder4Elite for monitor calibration, so instead I upgraded to the SpyderStudio kit which includes the Sypder4Elite for monitor calibration, SypderCube for help in setting white balance during raw processing, and the SpyderPrint for created printer profiles.

I have only had a few evenings to this point to create some basic profiles, so this information is preliminary at this time until I can do more complete testing and evaluation. But bottom line is that I have easily created 4 custom profiles and for all of them I can visually see improvement in evaluation prints over the stock profiles.

At this point I have created profiles for 2 Epson papers, glossy and ultra-premium luster, along with profiles for 2 Moab papers, Slickrock Metallic Pearl and Lasal Exhibition Luster, all photo black ink papers. For these profiles I used the basic target of 255 patches, although SpyderPrint does provide a target with 729 patches and you can have the inclusion of extended gray patches with either the 255 or 729 color patches for creation of black and white profiles.
Since Ihave only had time for basic evaluation at this point, I kept things simple and used the printer evaluation image created by Uwe Steinmueller and Jack Flesher at Digital Outback Photo. I intend to do further testing with some of my own images in the near future. For all 4 papers I was able to see a more neutral tone overall, the prints from the stock profiles were all slightly cool in tone when compared to the prints from the SpyderPrint profiles. These resulted in more accurate color reproduction in the skin tones of the people image, along with more realistic color of the strawberry image, and better highlight color in the lava rock image. In addition, there was better color saturation overall. It should be noted that the prints were evaluated under a full spectrum light rated at 5900K and 96 C.R.I from Blue Max lighting.

So, what conclusion can I draw at this time? My initial question was that while a custom profile might very well be better than a profile provided by the paper supplier based on scientific measurements, would the difference be visually apparent?  Based on this early testing I believe I can say that yes the difference can be visually noticeable. However, I want to emphasize that while the differences were noticeable to me they were very slight and for many they would not be enough to justify the cost and time of creating your own profiles. Remember, I have been printing with my 3880 and the stock profiles for over 2 years now and have been very satisfied with my prints and they have received excellent comments from others. For me, I find the work involved in creating the print as satisfying as taking the shot and like having the control and ability to get every last bit out of the image. After investing in cameras, lenses, processing software, printer, and papers, I see the additional few hundred dollars necessary for created my own printer profiles as a minimal investment.

At this time I have satisfied myself that yes, creating custom profiles can show an improvement over the stock profiles, and even the sub $500 systems now available to create printer profiles are of value.
Roy H
[url]http://www.reflectivephoto.net[/url]
 

by Royce Howland on Tue Jan 07, 2014 12:15 pm
User avatar
Royce Howland
Forum Contributor
Posts: 11719
Joined: 12 Jan 2005
Location: Calgary, Alberta
Member #:00460
Good followup Roy. I must say I'm not surprised that even your "preliminary" review with small patch counts was able to immediately produce visible improvements. As you say, some people will care more or less about the degree of improvement that's possible, but it is there to be had. It will be more significant with other papers because the unfortunate fact is the quality of the generic profiles is highly variable. In all likelihood many are worse than the ones you selected for your initial review.

Of course it's not necessary that each person buy & use their own custom profile creation gear, since there are services that will do it for a cost that's much lower if only a few papers are involved. But for those who really want to get hands-on control of their printing, including a wide variety of media, I think it's very beneficial to be able to do this work oneself.

More accurate hues, more saturation, deeper blacks, better shadow detail, more neutral grey tones... these are characteristics of a better profile. Even the lower cost entry level printer profiling gear usually can do a better job than most generic profiles, as you found.
Royce Howland
 

by bjs on Tue Jan 07, 2014 3:18 pm
bjs
Forum Contributor
Posts: 362
Joined: 18 Jan 2004
Location: Vancouver, BC
RoyH wrote: the prints from the stock profiles were all slightly cool in tone when compared to the prints from the SpyderPrint profiles.

Seems strange all the stock profiles are cooler (given different manufacturers).   Perhaps the SpyderPrint is being affected by the OBA's?

Datacolor recommends using their Saturation intent which has secret sauce baked in to be "pleasing" so I suppose that could account for some difference too.
 

by enrique patino on Fri Jan 10, 2014 11:31 am
enrique patino
Forum Contributor
Posts: 319
Joined: 6 Apr 2012
Excellent discussion... and thanks for the info!!

Do the ColorMunki or SpyderPrint work well with textured matte papers?

Thanks!
 

by Royce Howland on Fri Jan 10, 2014 11:43 am
User avatar
Royce Howland
Forum Contributor
Posts: 11719
Joined: 12 Jan 2005
Location: Calgary, Alberta
Member #:00460
I can't speak to SpyderPrint, but the ColorMunki Photo should work fine with textured matte. It takes multiple readings from each colour patch as you're scanning the targets, and averages them together. So small irregularities from paper surface should be normalized when the profile is created.

One thing the ColorMunki can't handle is very far-off-centre media such as Moab Slickrock Metallic Silver. Colour reproduction on this paper is far from normal, and the Munki software actually crashes when trying to generate a profile for it.
Royce Howland
 

by enrique patino on Fri Jan 10, 2014 5:41 pm
enrique patino
Forum Contributor
Posts: 319
Joined: 6 Apr 2012
Thanks Royce
 

by bjs on Wed Jan 22, 2014 2:55 pm
bjs
Forum Contributor
Posts: 362
Joined: 18 Jan 2004
Location: Vancouver, BC
Royce Howland wrote:One thing the ColorMunki can't handle is very far-off-centre media such as Moab Slickrock Metallic Silver. Colour reproduction on this paper is far from normal, and the Munki software actually crashes when trying to generate a profile for it.
I recently tried some RR Polar Pearl Metallic (I'm told identical to the Slickrock Metallic Pearl).   Red River's profiles are made by Chromix (who also provide a custom profiling service) and are generally decent.  However this one had enough issues that I pretty quickly made my own custom profile with much better results.

As an example with Gamutvision and custom profiling, I've included some before and after Gamutvision plots.

First a "rainbow" plot for the RR profile:
 
Image
There is a lot of banding evident.  The blue is particularly grim.

A look at the overall gamut plot highlights the blue issues (wire frame is AdobeRGB, solid is RR profile):
Image
The profile also struggles a bit with color neutrality as shown in this plot:

Image
The horizontal dotted lines in the above plot show the neutral color error (shown as 10x the LAB a* b* values).  This result suffers compared to top flight profiles.

So, given all that, I made a custom profile using 150 patches on a small 4x6 inch paper (using the RR profile as a pre-conditioning profile to help increase the patch efficiency).

The Munki "rainbow" plot below looks a lot better, less banding and no blue issues:
Image
The Munki profiles color neutrality is improved also.  In my experience it is extremely hard to see improvement beyond this level.  The deviations in the deepest blacks could be improved with more patches but probably only worth it for the fiendishly fussy printing black and white images (and even then maybe not...I have a very hard time seeing that amount of color shift in such deep blacks).
Image
It is also worth noting the difference in Dmax.   The Red River goes to Dmax of 2.03 while the Munki profile goes to a Dmax of 2.34.   This is a significant difference!   On photo papers I often achieve a deeper Dmax (with more gamut volume in the lower tones) with custom profiles.   I'm not sure the manufacturers put a lot of effort into finding the optimum paper/ink settings.

These differences are easily shown by comparing the two gamut plots:
Image
The wire frame is the Munki profile, the solid is the Red River profile.  The significant increase in Dmax and  gamut volume is clearly evident.  As is the blue hook of the original Red River profile (the Munki profile has none of that).

Not everyone can see these differences, or if they can, not everyone cares.  For example, my first prints with the stock RR profile showed issues.  But my simple 150 patch profile on 4x6 inch paper looks fine to me.  So, until I see a problem, I may never make a profile using a more typical 1500 patch set printed onto a 11x17 size paper.   That, in my opinion, is one of the nice aspects of the Munki package.  One can use the stock Xrite software knowing that if it doesn't do the job they can easily upgrade to the much more powerful Argyll software (the Munki hardware itself is very good).  Personally I use the Argyll exclusively since it is faster and uses a lot less paper given similar quality.

Unfortunately I don't have any of the new Slickrock Metallic Silver Royce mentioned.  It would be interesting to do a custom profile for that!  Especially since the stock Dmax seems pretty grim...Dmax = 1.6.    The color gamut does seem impressive though.
 

by bjs on Wed Jan 22, 2014 3:26 pm
bjs
Forum Contributor
Posts: 362
Joined: 18 Jan 2004
Location: Vancouver, BC
I forgot to mention...

If the Moab Slickrock Metallic Pearl is indeed the same paper as Red River Polar Pearl Metallic (as I'm told...apparently made by Mitsubishi) then using the Moab profile on the RR paper is a good option.  The Moab profile doesn't exhibit the Red River problems and overall is much closer to the Munki profile.
 

by Royce Howland on Wed Jan 22, 2014 5:16 pm
User avatar
Royce Howland
Forum Contributor
Posts: 11719
Joined: 12 Jan 2005
Location: Calgary, Alberta
Member #:00460
That's a good summary of another interesting case. That original RR profile does look poor, while the custom profile of the paper looks much better. More like what I would expect for this media if it is the same as Moab Slickrock. The Argyll software looks to be doing a good job driving the ColorMunki spectro hardware. I'll have to take a serious look at Argyll at some point; up to now I've only casually skimmed the documentation for it but haven't tried installing or using it.
Royce Howland
 

by bjs on Thu Jan 23, 2014 2:16 pm
bjs
Forum Contributor
Posts: 362
Joined: 18 Jan 2004
Location: Vancouver, BC
bjs wrote:If the Moab Slickrock Metallic Pearl is indeed the same paper as Red River Polar Pearl Metallic (as I'm told...apparently made by Mitsubishi)


I decided to quit being lazy and checked if Ernst Dinkla had measured these papers.  Indeed he has.

The SpectrumViz plots are identical...both front and back sides...with white or black backing...clearly the same paper!

As an aside, SpectrumViz it is very useful for these kind of things...identifying papers that may take the same profiles or looking for cheaper sources for a given paper.  And of course checking for the level of OBAs in a paper (something the Munki hardware, being a UV cut instrument, doesn't do well...although a black light works pretty well if one actually has the paper).
 

by bjs on Tue Jan 28, 2014 1:03 pm
bjs
Forum Contributor
Posts: 362
Joined: 18 Jan 2004
Location: Vancouver, BC
Royce Howland wrote:I'll have to take a serious look at Argyll at some point;  
 
 
It's definitely worth a look.  If you do, my main tip would be to start with the default settings (i.e. don't be overwhelmed by the plethora of options). It is hard to improve over the default settings except in specialized cases (and then one has to know what they are doing!).
 
The Argyll toolset also complements Gamutvision.   Gamutvision is good at comparing profiles but ultimately doesn't show which profile is more accurate.  Argyll has the tools to actually measure profile accuracy (i.e. compare profile with what actually gets printed).
 

Display posts from previous:  Sort by:  
19 posts | 
  

Powered by phpBB® Forum Software © phpBB Group