Ad
Ad
Ad
Pages: [1] 2 »   Bottom of Page
Print
Author Topic: About shadow details in ICC printer profiles  (Read 6606 times)
MonsterBaby
Full Member
***
Offline Offline

Posts: 133


« on: June 22, 2011, 02:22:35 AM »
ReplyReply

good morning forum,

im generating ICC profiles now for years for my artwork.. and just wanted to open a discussion here about shadow details and how to achieve the best.

im working with a Z3100 and Z3200.

using either the built in spec or a DTP 20.

no metter which i use or which program i use (profile maker, APS) it lacks details in the shadows of a print. just now for the Z32 i did a new profile for HM photo rag baryta. i decided to use more ink and less GE as this turned out to be possible and getting the largest gamut. BUT.. printing an image with a very dark lower half of a new york sidewalk turned out: black!

i dont want to use less ink.. overall.. as the paper can handle it. of course now i edited the ICC with monaco (gave me best results) and opend the shadows quite well..
but is this really necessary ?

or is this really the domain of a RIP where u can use more ink in the channels but then reduce the total ink limit over the printer driver?

thanx for some ideas
Logged
Czornyj
Sr. Member
****
Online Online

Posts: 1419



WWW
« Reply #1 on: June 22, 2011, 03:05:13 AM »
ReplyReply

The perceived reproduction of shadow detail vastly depends on viewing condition. Remember that you're editing images on a bright monitor, in many cases in a dark surround which additionally opens the perceived shadows.

Try to evaluate the print in a strong daylight - it may occur, that the shadow detail is already there, but you can't see it due to dim light in your environment.

The solution to this issue is to calibrate the display to lower luminance and put it on a brighter background, or increase the brightness of the print lighting. You can also calculate the profile using ArgyllCMS with viewing condition compensation, see flag -c and -d description:
http://argyllcms.com/doc/colprof.html
Logged

Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6931


WWW
« Reply #2 on: June 22, 2011, 07:06:49 AM »
ReplyReply

I would explore everything you can do in Photoshop to improve the rendition of shadow detail before resorting to playing with the print pipeline. There are all kinds of techniques, do a search on this website.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
gromit
Full Member
***
Offline Offline

Posts: 133


« Reply #3 on: June 22, 2011, 08:11:23 AM »
ReplyReply

no metter which i use or which program i use (profile maker, APS) it lacks details in the shadows of a print.

Before you bother looking at profiling issues, use Photoshop’s Info panel and set the secondary read-out to Lab Color. Move your mouse over shadow areas in your image and observe the L value. Anything less than an L of 2 will print as black or near black, irrespective of how the shadows appear on your monitor. Only if the values are higher would you then have a profiling and/or ink loading/linearization issue.
« Last Edit: June 22, 2011, 08:16:41 AM by gromit » Logged
MonsterBaby
Full Member
***
Offline Offline

Posts: 133


« Reply #4 on: June 24, 2011, 08:59:40 AM »
ReplyReply

thnx everyone.

i was not lookin for help actually on how to view a print and to set up monitor lighting.

i was hoping for a top of the line discussion about shadow details - printer driver vs. rip f.e.

and i am just surprised that even with a large atkinson's chart i loose shadow details in the print through f.e. qimage..

Logged
TylerB
Sr. Member
****
Offline Offline

Posts: 358


WWW
« Reply #5 on: June 24, 2011, 12:41:26 PM »
ReplyReply

there are a great many contributing factors to this issue. Rendering intent is a big one to start with, there is no standard with regard to perceptual rendering and it's behavior with regard to shadow detail, amongst many other things. I use two different profiling softwares(sp?), on matte (I'll get back to that) one is wide open and the other more as expected. Additionally both are more open than their corresponding relative colormetric conversions even ( and most relevant) with black point compensation. We do have a standard for colormetric conversions and they should be very similar from one profiler to another, and the big BUT is... there is no standard for black point matching, and 99.99% of the time, black point matching, or compensation, is the only viable way of using relative colormetric for most printing. So the low value contrast may change from one cmm/RIP or driver/application/profiler to another even in colormetric conversions.
Another issue rarely discussed but common to all good printmakers I know- our spectros tend to see into gloss and matte surfaces quite differently, and result in much more dumping down of shadow contrast on photo surfaces than matte surfaces. Because of the way they illuminate and detect during measurement, resulting profiles even from the same software, with as much equal as possible in the build, still result in darker shadows down into black with gloss than matte. You can see it right on the monitor with softproof and ink black turned off to avoid that obvious difference in display. With prints it's obvious. My more "open" profiler works better for me on photo surfaces then...
Editing profiles is generally not recommended as other problems tend to crop up. Your mention of inking controls, either those supplied in your driver, or selecting different media settings, or RIP setup possibilities, are not terribly relevant to internal shadow contrast, but possibly are relevant to the dumping of values down to black. No profiler can know how to map lower values that all read black in the first place.
So first of all, whatever ink controls you have must be set for no "clipping", or dumping of values, even at the expense of gamut. A sweet settings spot must be found with no profile, then profile over that.
I hope I'm not spewing stuff you already know and wasting your time...
So once all that is in place and you have a good profile with good continuous tone down to black, and given all the variables above, IMHO the only option to keep shadows as open as you want are a good color managed workflow, calibrated monitor, and use soft proofing with the profile and rendering intent, then control shadows in the file itself with editing.
Hope some of this helps.
Tyler
Logged
Iliah
Sr. Member
****
Offline Offline

Posts: 410


« Reply #6 on: June 25, 2011, 12:32:46 PM »
ReplyReply

L* is a relative measure, it is a scale where 100 is the whitest white the media can render.
Logged
GamutGirl
Newbie
*
Offline Offline

Posts: 37


« Reply #7 on: July 21, 2011, 04:17:01 PM »
ReplyReply

We've been using X-Rite's i1 Match software/hardware to make printer profiles for years with our large format Epson printers. Recently, we compared stock Epson profiles vs custom ICC profiles using our i1 Match software. Hands down the detail in the Epson stock profiles is apparent. The X-Rite custom ICC profiles, while color accurate, seemed flat and lacked shadow detail. What can we do to get the same detail as the Epson stock profiles provide? Will upgrading to i1 Profiler work? Should we invest in ProfileMaker software?
Logged
stretchdcanvas
Guest
« Reply #8 on: July 21, 2011, 05:30:48 PM »
ReplyReply

+1 for Editing.

"Just do it"

Logged
tony22
Full Member
***
Offline Offline

Posts: 105


« Reply #9 on: July 21, 2011, 06:07:05 PM »
ReplyReply

The X-Rite custom ICC profiles, while color accurate, seemed flat and lacked shadow detail. What can we do to get the same detail as the Epson stock profiles provide? Will upgrading to i1 Profiler work? Should we invest in ProfileMaker software?

I'm also very interested in a response to this observation.
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9097



WWW
« Reply #10 on: July 21, 2011, 07:05:47 PM »
ReplyReply

Recently, we compared stock Epson profiles vs custom ICC profiles using our i1 Match software. Hands down the detail in the Epson stock profiles is apparent. The X-Rite custom ICC profiles, while color accurate, seemed flat and lacked shadow detail.

Assuming you are using Epson profiles created in the US, those were built using X-Rite PROFILER.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
JeffKohn
Sr. Member
****
Offline Offline

Posts: 1671



WWW
« Reply #11 on: July 21, 2011, 10:08:22 PM »
ReplyReply

I'm surprised to hear the problem is with photo-black media, in my experience muddy shadows is far more likely to be a problem with matte media because of the much lower DMax, which means the tonal range has to be compressed more heavily (particularly in the shadows). The Hahnemuhle Baryta papers are particularly good, probably some of the easiest papers to profile due to the excellent gamut and dmax.

You might try the Logo Chroma Plus perceptual intent option, I've found that tends to open shadows up a bit more than the Logo Colorful option.

You mentioned you raised the ink limits to get more gamut volume, but gamut volume is not the only indicator of profile quality, or even necessarily the most important one. It may be that while the higher ink limit gives you richer midtones and larger overall gamut, the ink density curve is resulting in too much ink in the shadows - and PMP does not handle that well at all. The ink density curves are not something you can change when using the standard drivers (not for Canon/Epson, anyway; I assume the same is true for HP). So if you really want to go with the higher ink limits you may need to try some different media types to see if one of them has a more suitable curve. Only other option would be if a RIP allows this sort of control.
Logged

shewhorn
Sr. Member
****
Offline Offline

Posts: 540


WWW
« Reply #12 on: July 21, 2011, 10:26:18 PM »
ReplyReply

Assuming you are using Epson profiles created in the US, those were built using X-Rite PROFILER.

I'll venture a guess that they didn't scan the targets with an Eye One Pro either which doesn't do the best job with regards to shadow detail.

Cheers, Joe
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9097



WWW
« Reply #13 on: July 22, 2011, 07:51:24 AM »
ReplyReply

I'll venture a guess that they didn't scan the targets with an Eye One Pro either which doesn't do the best job with regards to shadow detail.

I know they didn’t but I’d like to know details about this issue with shadows and an EyeOne Pro. I’ve never seen such an issue comparing it with say an iSis, using of course the same target.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
shewhorn
Sr. Member
****
Offline Offline

Posts: 540


WWW
« Reply #14 on: July 22, 2011, 09:41:30 AM »
ReplyReply

I've noticed that compared to the Spectroscan/lino, the Eye One Pro is a bit lacking when it comes to shadow detail.

Cheers, Joe
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9097



WWW
« Reply #15 on: July 22, 2011, 09:42:33 AM »
ReplyReply

I've noticed that compared to the Spectroscan/lino, the Eye One Pro is a bit lacking when it comes to shadow detail.

I guess I’ll have to dust off mine and see.
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Scott Martin
Sr. Member
****
Offline Offline

Posts: 1311


WWW
« Reply #16 on: July 22, 2011, 10:00:04 AM »
ReplyReply

I've noticed that compared to the Spectroscan/lino, the Eye One Pro is a bit lacking when it comes to shadow detail.

I haven't seen that, unless I used the polarizer filter (which is a wildcard anyway). I would expect the EyeOnePro data to be a bit "dirty" in comparison the Lino or DTP70 but I haven't seen that translate to the real world visual results you're describing. Interesting. Do you have more than one EyeOnePro to compare with? Does it pass Diagnostics?
Logged

digitaldog
Sr. Member
****
Offline Offline

Posts: 9097



WWW
« Reply #17 on: July 22, 2011, 10:34:23 AM »
ReplyReply

I haven't seen that, unless I used the polarizer filter (which is a wildcard anyway).

Good to know because after 45 minutes, I’m giving up on even trying to get the Spectroscan to show up via the old Keyspan USB adaptor. Tried on an old G5 running 10.5 and a newer 10.6 machine, both with updated drivers. Keyspan is seen, Spectroscan isn’t. That shows how long its been since I had to hook one up!

Anyone want a screaming deal on a green Spectroscan? I’m not futzing around with this old stuff anymore <g>
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
Scott Martin
Sr. Member
****
Offline Offline

Posts: 1311


WWW
« Reply #18 on: July 22, 2011, 10:39:11 AM »
ReplyReply

Anyone want a screaming deal on a green Spectroscan? I’m not futzing around with this old stuff anymore <g>

Oh, but it looks so good next to your letter press in the lobby - LOL! Yeah, the Purple Lino's are a little more handy to have around but, like you, I'm looking more to the future.
Logged

Czornyj
Sr. Member
****
Online Online

Posts: 1419



WWW
« Reply #19 on: July 22, 2011, 10:51:37 AM »
ReplyReply

Anyone want a screaming deal on a green Spectroscan? I’m not futzing around with this old stuff anymore <g>

If you wouldn't mind shipping it to EU, I'd take it - my wife was treating me bad lately, this should give her a lesson Cheesy
Logged

Pages: [1] 2 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad