Luminous Landscape Forum
Raw & Post Processing, Printing => Colour Management => Topic started by: TeamG on May 15, 2019, 10:28:21 pm
-
Hello,
Much apologies for my first time post (long time lurker) but I can't seem to find the answer I'm looking for
Epson SC P800, Hahnemule Photo Rag 308, Benq SW2700,
When I profile my own paper using the i1Pro 2 and i1Profiler and 2531 targets (sorry if the number is wrong, going on memory) I don't seem to get any better result than the canned profile
I'm determining this by when the profile is built in i1profiler there is the option to compare it to another icc profile, and the graphic that appears shows my profile to be almost identical to the canned one. This compared to a profile that was done by Kayel Australia on some Canson paper, and when the profile is compared the graphic is noticeably larger than the canned one
I've followed all the instructions I can on the workflow through i1profiler, I leave the prints to dry overnight I think I am doing the right things but don't seem to see an improvement
Is there something obvious that I'm missing?
Thanks in advance and apologies if this has been answered before, my search fu may be too weak
Glenn
-
Why not check the copyright tag of the canned profile; could be made from the same software.
-
Hello,
Much apologies for my first time post (long time lurker) but I can't seem to find the answer I'm looking for
Epson SC P800, Hahnemule Photo Rag 308, Benq SW2700,
When I profile my own paper using the i1Pro 2 and i1Profiler and 2531 targets (sorry if the number is wrong, going on memory) I don't seem to get any better result than the canned profile
I'm determining this by when the profile is built in i1profiler there is the option to compare it to another icc profile, and the graphic that appears shows my profile to be almost identical to the canned one. This compared to a profile that was done by Kayel Australia on some Canson paper, and when the profile is compared the graphic is noticeably larger than the canned one
The gamut size differences tell little about the quality of the profiles. Bigger may, or may not, be better.
I've followed all the instructions I can on the workflow through i1profiler, I leave the prints to dry overnight I think I am doing the right things but don't seem to see an improvement
What are you looking for when you say you don't seem to see an improvement? What is defective or what are you looking for that you feel would be an improvement?
When you make a profile with I1Profiler there is an option to print an image of a colorchecker that you can look at with one in the I1Profiler package that has holes it in so you can position one over the other to see how well the colors match. Have you done that and/or do they match?
Is there something obvious that I'm missing?
Thanks in advance and apologies if this has been answered before, my search fu may be too weak
Glenn
-
The gamut size differences tell little about the quality of the profiles. Bigger may, or may not, be better.What are you looking for when you say you don't seem to see an improvement? What is defective or what are you looking for that you feel would be an improvement?
To be honest I have been exclusively looking at the comparison graphic and getting frustrated there is no apparent difference, feeling like I've done something wrong or missed a crucial step. Looks like I might have to burn a few pages in comparison prints instead.
When you make a profile with I1Profiler there is an option to print an image of a colorchecker that you can look at with one in the I1Profiler package that has holes it in so you can position one over the other to see how well the colors match. Have you done that and/or do they match?
No I haven't done that as yet due to not being able to get mentally past the above - but now I will
Thanks for taking the time with your reply, I appreciate it
-
Why not check the copyright tag of the canned profile; could be made from the same software.
Thanks, Andrew, I'll check that out
-
Check out Andrew Rodney's site and videos. His stuff is the best on the web about color management.
http://www.digitaldog.net/
-
The graph of the profile's gamut volume shows you three things: (i) the size of the gamut volume, (ii) whether there are kinks that could be indicators of discontinuous performance of colour rendition, and (iii) the shape of the gamut (the areas of the colour spectrum capable of more or less saturated colour rendition).
The four most important quality characteristics that you don't get from looking at that graph, or in fact from an unaided visual comparison of your printed colours relative to X-Rite's colour checker chart are: (i) whether the profile is allowing for accurate rendition of the file values on paper, (ii) the smoothness of the tonal gradations on paper where the file numbers indicate they should be smooth (skies, skin tones and the like), (iii) whether the shape of the gamut is well-suited to the colour characteristics of your photographs and (iv) the quality of shadow and highlight detail achievable with your profile, especially for Black and White prints where this characteristic would be most noticeable. While these characteristics depend heavily on the kind of paper and ink you are using, the profile is a key participant in the whole set of conditions that determine print quality. To assess these other factors you need other methods of analysis than those provided in X-Rite software. You can check several of my more recent printer/paper reviews on this website to get some ideas. But cutting to the bottom line of your observations to date, let me confirm what Doug Gray said above: the size of the gamut volume doesn't tell you much about the overall quality of a profile. I have produced custom profiles whose gamut volumes are a bit smaller than those of the canned profiles, but whose colour rendition is more accurate in my printing conditions. I'll take the accuracy over the variances of gamut volume any day.
(Typo corrected)
-
It's possible that you've just come across a good canned profile that as Andrew said could be made with the same profiling package. And this would indicate that you've done everything correctly.
I think you'll find that not all canned profiles are good so with a different paper you may see more differences and improvement.
-
It's possible that you've just come across a good canned profile that as Andrew said could be made with the same profiling package. And this would indicate that you've done everything correctly.
I think you'll find that not all canned profiles are good so with a different paper you may see more differences and improvement.
I think it's fair to suggest that today's canned profiles from the more prominent manufacturers and suppliers of printers and papers are generally better made (often much better made) than they were in the earlier days of inkjet printing, as a result of which the evidence of print quality improvement from custom profiling can be less obvious in many cases than it was in the more distant past. I still prefer custom profiles for my printing (with an Epson SC-P5000), because they do provide that last ounce of reproductive fidelity I seek to achieve, but I acknowledge that for numerous papers I could "make do" with supplied profiles.
-
Not all canned ICC profiles are equal in quality even when supplied from the same vendor as shown here:
Not all ICC profiles are created equally
In this 23 minute video, I'll cover:
The basic anatomy of ICC Profiles
Why there are differences in profile quality and color rendering
How to evaluate an ICC output profile
Examples of good and not so good canned profiles and custom profiles on actual printed output.
High resolution: http://digitaldog.net/files/Not_All_Profiles_are_created_equally.mp4
Low resolution (YouTube): https://www.youtube.com/watch?v=TNdR_tIFMME&feature=youtu.be
-
But cutting to the bottom line of your observations to date, let me confirm what Doug Gray said above: the size of the gamut volume doesn't tell you much about the overall quality of a profile. I have produced custom profiles whose gamut volumes are a bit smaller than those of the canned profiles, but whose colour rendition is more accurate in my printing conditions. I'll take the accuracy over the variances of gamut volume any day.
Ok, that bit about the gamut even being a bit smaller also resonates, thank you. I'll suck it up and burn a bit of paper for a more informed opinion of how my profiles are going and report back over the weekend
-
Not all canned ICC profiles are equal in quality even when supplied from the same vendor as shown here:
Thanks, Andrew. I'll make space to watch that oer the weekend
-
Thanks, Andrew. I'll make space to watch that oer the weekend
Game of Thrones has competition :D
-
Lol, I actually have some print competitions coming up I need to prep, and a quarterly market at the end of June where I sell my prints that I need to restock for - so GoT will have to wait :)
-
Apologies all I forgot to close the loop on this
In short, focusing on the gamut grapphic post calibration was not the way to go
I've done a number of prints now and can clearly see my profiles are better than the canned profile. In particular shadow seperation and colour vibrancy are better
Thanks for taking the time to reply
-
Why not check the copyright tag of the canned profile; could be made from the same software.
How does one do that?
-
How does one do that?
I use the Colorsync utility.
-
I use the Colorsync utility.
Is there an equivalent for the Windows OS?
-
Is there an equivalent for the Windows OS?
No idea.
-
ICC Profile Inspector, a free utility downloadable from color.org.
Richard Southworth
-
ICC Profile Inspector, a free utility downloadable from color.org.
Richard Southworth
Very good tool. One can examine the structure and components of a profile's details. LUT tables, pre/post conditioning curves, PCS (Lab or XYZ), RGB profile details. Pretty much anything in the profile.
-
Not all canned ICC profiles are equal...
I just took the time to see some of your videos; very helpful !
also the page on 'Black point compensation'.
so thanks!
I also downloaded the calibration print.
What i still do not understand is that i get so many different values of this colorchecker classic card? (24 colors)
The values from different sources as the one on your target and my own are always slightly different. ( and i use the same colorspace)
-
What i still do not understand is that i get so many different values of this colorchecker classic card? (24 colors)
The values from different sources as the one on your target and my own are always slightly different. ( and i use the same colorspace)
Lab values?
-
Lab values?
I will look into that; so you say it might be the color engine that plays up...?
-
I will look into that; so you say it might be the color engine that plays up...?
The RGB values depend on the RGB color space. Not the case with Lab values (they are device independent). Not that you should expect a prefect match to a reference anyway:
http://www.babelcolor.com/index_htm_files/RGB%20Coordinates%20of%20the%20Macbeth%20ColorChecker.pdf
-
Hello,
Much apologies for my first time post (long time lurker) but I can't seem to find the answer I'm looking for
Epson SC P800, Hahnemule Photo Rag 308, Benq SW2700,
When I profile my own paper using the i1Pro 2 and i1Profiler and 2531 targets (sorry if the number is wrong, going on memory) I don't seem to get any better result than the canned profile
And why should you? The paper manufacturers have highly experienced technicians, top-end equipment, and a lot of motivation to provide highest quality profiles to help boost the reputation of their paper. Of course not every canned profile is ideal, but I have used nothing but for years with great results. But of course, if someone puts a lot of effort and expense into making a custom profile, they will almost always tend to see it as better (psych 101).
-
And why should you? The paper manufacturers have highly experienced technicians, top-end equipment, and a lot of motivation to provide highest quality profiles to help boost the reputation of their paper.
Again no: "All generalizations are false, including this one".-Mark Twain
Further, my video shows (Reply #10) that this isn't the case with one manufacturer (of papers and printers) in some cases, better in others (because they are using a very good off the shelf product to do so).
Again, not all ICC profiles are created equally. Even from the same manufacturer.