Topic: Magenta-ish skies with Ilford Galerie Smooth Pearl (compared to Epson Semigloss) (Read 8202 times)

pompon44 · « **on:** April 16, 2011, 06:42:03 pm »

Hi,

I've been playing a bit with my new Epson 3880 I received a few days ago (so please bear with me, I'm a newbie in the usage of a "real" photo printer, my previous one being a Canon IP4200)...

Before getting in the A3/A2 land, I'd want to settle on a couple of papers. So I started with a few A4 (sample or not) packs from Epson (for reference, as I assumed their profiles would be top quality), Ilford (because I like their Pearl paper) as well as Canson and Hahnemüle (for my first attempts in matte papers, but that will be for later on...)

I started with Epson Semigloss and was very pleased with the results. Next I tried the Ilford Galerie Smooth Pearl, and noticed a moderate (but still really visible) magenta cast in the (blue) skies.

I'd tend to conclude that the Ilford profile provided for the 3880 is not 100% optimized and that would explain the worst rendering of the blues, as compared to the Epson semigloss profile. Would that be a valid conclusion, or could a pilot error explain such an observation ?
Do you guys observed such a thing as well, for those who've used both papers ?

Thanks, and sorry for the long question...

Mark D Segal · « **Reply #1 on:** April 16, 2011, 06:45:38 pm »

It's possible that the Ilford profile isn't great, but before concluding on that, there could be what you call "pilot error". Each paper has its own tint, so my question is whether you readjusted your image under softproof in Photoshop with the Ilford profile loaded as the softproof condition?

pompon44 · « **Reply #2 on:** April 16, 2011, 07:01:54 pm »

Hi Mark,

No, I did nothing to the image prior to print it (from Lightroom 3, btw), nor for the Epson semigloss, nor for the Ilford Smooth Pearl.

What you're saying is that I should try to soft-proof first to see if I can overcome the problem ?

Regards,

Mark D Segal · « **Reply #3 on:** April 16, 2011, 07:28:37 pm »

Quote from: pompon44 on April 16, 2011, 07:01:54 pm

Hi Mark,

No, I did nothing to the image prior to print it (from Lightroom 3, btw), nor for the Epson semigloss, nor for the Ilford Smooth Pearl.

What you're saying is that I should try to soft-proof first to see if I can overcome the problem ?

Regards,

Yes.

But you cannot do that from Lightroom. You will need to export the image to Photoshop, activate the View>Softproof>Custom, load the Ilford profile, make sure that "Simulate Paper Color" is selected in the softproof dialog, accept this softproof condition, then look at the softproof of the photo on your display and see whether that magenta cast in the sky is noticeable. If it is, you would want to remove the cast in any one of a number of ways in Photoshop, using either a Color Balance Adjustment Layer, or a Curves Adjustment Layer (going into the Green channel and shifting the mid-point of the curve a bit more toward Green), or if you think the problem is limited to the sky, you could use a Selective Color Adjustment Layer, select the offending color (Blue) and decrease the Magenta content as shown in that dialog - a number of possible solutions depending on the exact nature of the problem, which obviously I cannot see. This is all just for your guidance in case you haven't done much of this in Photoshop before.

Now, just in case the profile is problematic (I hope it is the correct profile for your printer model!), you may also benefit by having a custom profile made if you intend to do a lot of printing with this paper.

Garnick · « **Reply #4 on:** April 16, 2011, 11:57:31 pm »

Another possible culprit may be the rendering intent. Most, if not all printer profiles that I use require the rendering intent to be set to "perceptual". However, as per Ilfords specs, the Galerie Smooth Pearl requires that the intent be set to "relative colorimetric". That's for the Epson 9900, so I don't know if it relates to your printer. However, it's certainly worth a try. Simply change the rendering intent and do another test print, it might work. Also, check the info that was sent with the profile and see what the rendering intent should be for that paper and your printer. I hope this helps.

Good luck,
Gary

pompon44 · « **Reply #5 on:** April 17, 2011, 10:35:41 am »

Hi guys,

Thanks a lot for the help.

For the record, I believe I did use the correct profile (named IGSPP11_EPP3880_PSPPn), and I indeed tried both rendering intents (the "recommended" one of relative colorimetric first, and then the perceptual) before asking my question here. The rendering intents changed the contrast of the image but not so much the colors for that specific case.

After a good night, and a small visit to Photoshop and soft-proofing, I'm now convinced that none of the two papers (Epson semigloss and Ilford Galerie Smooth Pearl) reproduce correctly my screen (at least not in sky), i.e. in both cases I obviously see the effect of the paper color, both in soft-proof and in print. Turns out that the Epson paper is "warmer" and that's probably why the change in the blues did not bothered me at first glance, as I tend to prefer warmer looks.

So indeed I'll have to play a bit with soft-proofing to see what I can do for this sky ;-)

Again, thanks for your help,

Mark D Segal · « **Reply #6 on:** April 17, 2011, 11:04:16 am »

BTW, fundamental question I forgot to ask: is your display calibrated and profiled? If so, how - what hardware and what parameters?

pompon44 · « **Reply #7 on:** April 17, 2011, 11:19:44 am »

My display (an iMac 24" from end of 2008) has been calibrated (a couple of months ago, admittedly...) with a Spider 2.

Concerning the parameters, as far as I remember I did not fiddled with those proposed by the software... but I'll have to cross-check that.

pompon44 · « **Reply #8 on:** April 17, 2011, 05:09:16 pm »

Checked what the settings were for my Spyder 2 (and not my spider as I wrote in my previous post ;-) ) :

Gamma 2.2
White-point 6500 K

Does that make sense ?

Mark D Segal · « **Reply #9 on:** April 17, 2011, 05:20:36 pm »

6500 may be too cool for the papers you are using and the light you would probably be viewing them under (say a range of halogen to daylight - about 3600~5000 respectively). You may wish to try recalibrating and profiling on the basis of 5000~5500. I'm not sure how well a Spyder 2 calibrates and profiles an iMac screen. There are folks around here who would know that better than me. The main point here is that if your display is showing things too cool, you may be warming them up too much before you send them to print, but that's a hypothesis, and it should play out for all the papers you use, but may show as a problem more on one paper than another. No harm doing a new display profile say at 5000, activate soft-proof with the correct paper profile loaded, readjust your image for that combination of stuff, print it and see how it comes out. All that said, if the problem is too much magenta in the blues, re-setting the yellow-blue aspect of display management may not solve it. I still think getting a custom profile made for your paper and printer can't do any harm, except the cost, and just may solve the problem.

Ernst Dinkla · « **Reply #10 on:** April 18, 2011, 03:16:20 am »

Print the Atkinson evaluation target (with the rock arch and sky showing through it) without any editing done. Perceptual and Relative Color metric. 5000K viewing light and change the distance to the viewing light. The less light/more distance the more the magenta creeps in. Another variable is printing from different color spaces, sRGB more easily translates to magentish skies than printing from ProPhoto. Must have something to do with color mapping. Then of course there is the variety of profile creators. I tested the Ilford profile for the Z3200-PS maybe a year ago against the HP Z3200 profile creators Color Center and APS on Ilford Galerie Smooth Pearl II. There were slight differences but the other factors mentioned here were more significant.

I think you know that there was an older version of that paper and didn't mix up profiles between them, unlikely that there is a 3880 Ilford profile for the old paper version.

Magenta skies is an ever returning topic in printing.

met vriendelijke groeten, Ernst

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/

pompon44 · « **Reply #11 on:** April 18, 2011, 03:49:54 am »

Hi Ernst,

Indeed, yesterday evening I printed the test image from http://www.outbackprint.com/printinginsights/pi049/essay.html (which, as far as I understand, derives from Atkinson's one) on Epson Semigloss (perceptual) and Ilford Galerie Smooth Pearl (the new one), relative and perceptual.

Relative and perceptual prints on Ilford were on par, as afar as colors go (the relative print was more "contrasty" and generally more pleasing to my eyes, but the colors were about the same as the other one).

But clearly (*) the Ilford prints have a magenta cast in the sky below the arch, while the Epson has not. In addition, the color squares in that image (below the grayscale image) show pretty "dull" red and orange in the Ilford, and not super clearly separated greens (the 3 next to the yellow in the top row).

So, my question now probably is : can this be fixed with a better profile or should I just move on and abandon that paper ?

Related to this question is another one : should I invest into a profiler (thinking of the ColorMunki as that's as expensive as I'm ready to go ;-) ) ? I mean, if I have this kind of problem with the first (well, actually second, counting the Epson Semigloss as first) paper I test, will it happen over and over again with other papers ? And would the ColorMunki solve it ?

Sorry for all the questions...

Thanks,

(*) I should mention though that I do not have a controlled viewing environment. I checked that a few minutes ago (7h30 GMT) under daylight (shade of a sunny day, morning)...

Ernst Dinkla · « **Reply #12 on:** April 18, 2011, 07:04:54 am »

Hard to say, all the prints I made on the Ilford are equally satisfying when compared to Z3200 prints on HP ID Premium RC Photo Papers. The only comparable RC paper of Epson that I used is the Epson Proofing White Semimatte and that one shows a lot more saturation on all hues and an equal blue + no magentisch sky. Custom Z3200 profile. But that paper is not a mainstream Epson RC paper, it doesn't have FBAs (at least not the spectral output bump) and possibly has a better coating. Epson uses it to show (off) the Epson x900 gamuts, it may tell more about that paper than about the printers.

Try some other Epson driver media presets for the Ilford than recommended by Ilford. If ColorBase works on the 3880 a correction of a slight deviation in the factory calibration could work too. You might know someone with a good spectrometer for that purpose.

I have my reservations about all spectro or colorimeters below the Eye 1 Basic level, mainly in relation to FBA effects. The ColorMunki doesn't measure below 430-420Nm, the colorimeters are not much better there. While colorimeters may have been adapted reasonably well to CcMmYK dye and pigment inks I doubt the same is true for ink sets with extra hues/pigments, OG or RGB inks. At least I didn't get a satisfying answer when I discussed that subject with a Datavision representative some years back.

Your viewing light conditions, UV more or less included, sufficient light level?

met vriendelijke groeten, Ernst

New: Spectral plots of +250 inkjet papers:

http://www.pigment-print.com/spectralplots/spectrumviz_1.htm

Mark D Segal · « **Reply #13 on:** April 18, 2011, 08:29:53 am »

Quote from: pompon44 on April 18, 2011, 03:49:54 am

So, my question now probably is : can this be fixed with a better profile or should I just move on and abandon that paper ?

Related to this question is another one : should I invest into a profiler (thinking of the ColorMunki as that's as expensive as I'm ready to go ;-) ) ? I mean, if I have this kind of problem with the first (well, actually second, counting the Epson Semigloss as first) paper I test, will it happen over and over again with other papers ? And would the ColorMunki solve it ?

I would not recommend you abandon the paper if you like it - on the condition that you can get the colours right. To make sure whether or not it is a profiling problem, it would be sensible to get one custom profile made and see whether that solves the problem. If it does, then you have your answer. As for buying a profiling kit such as ColorMunki, it depends on how many different papers and printers you intend to work with. If you plan to settle on one or two papers for which you have good profiles and you intend to use the same printer for a while, you can put off that expenditure, but if you are willing to invest in it, you have a lot more flexibility to experiment with lots of different stuff and not incur the cost of custom profiles every time. It's a trade-off which varies according to your personal conditions and requirements.

Alan Goldhammer · « **Reply #14 on:** April 18, 2011, 08:37:32 am »

I have a ColorMunki for profiling and it does a good job on the blue sky in the Atkinson image (I also use the Outback test print that Jack Flesher created). I don't print on the Ilford Smoot Pearl paper except for 4x6 snapshots where I use the Ilford profile; the prints are fine but I've not taken the time to evaluate the colors as they are just snapshots. I have profiled Ilford Gold Fiber Silk with ColorMunki and the results are fine. As Mark Segal said, you need to evaluate the number of papers you want to print on before getting a ColorMunki (of course you can also use it to profile your monitor as well). It is cost effective but if you are only going to use a couple of papers, having a profile made might be the way to go.

pompon44 · « **Reply #15 on:** April 18, 2011, 12:43:04 pm »

Again, thanks for your help.

If I go the custom profile route before getting a ColorMunki (if ever), then I'll be faced to the problem of selecting the maker of that custom profile... (knowing that I'm from France, I'm assuming I should try to find "local" ?). Because I suspect not all profile makers are created equal ? (well, they may be created equal, like we all are ;-), but they may have, let's say, various levels of competence ?)

Regards,

Mark D Segal · « **Reply #16 on:** April 18, 2011, 12:54:08 pm »

Quote from: pompon44 on April 18, 2011, 12:43:04 pm

Again, thanks for your help.

If I go the custom profile route before getting a ColorMunki (if ever), then I'll be faced to the problem of selecting the maker of that custom profile... (knowing that I'm from France, I'm assuming I should try to find "local" ?). Because I suspect not all profile makers are created equal ? (well, they may be created equal, like we all are ;-), but they may have, let's say, various levels of competence ?)

Regards,

Here's the guy you want to got to in France http://www.cmp-color.fr/Imprimante.html (Christophe Metairie) in Bayonne. He's very experienced and very good.

pompon44 · « **Reply #17 on:** April 18, 2011, 05:20:10 pm »

Thanks for the link.
Turns out Bayonne is about 15 kms of my hometown ;-) (but I'm now 600 kms north of that, unfortunately...)
I'll contact the guy.
Regards,

pompon44 · « **Reply #18 on:** April 21, 2011, 05:53:41 pm »

Hi,

Some followup.

I did contact Christophe Metairie, but the price of the profile (50-60 € depending on number of target patches, as far as I understand), made me think twice. So I made some further checks (e.g. changing the media type from semigloss to glossy or luster) to see if I should simply abandon that paper...

Turns out that in addition to having a slight magenta cast, the prints also exhibit what seems to be called "scum dots" (aka light gray/cyan border around the image, it's subtle but definitely there. More easy to see under daylight than under halogen light, but it's there !).

Problem is that, as far as I can tell, the profile (IGSPP11_EPP3880_PSPPn.icc) is a v2.0.0 one... So what can I do ? I'm printing from LR 3.3 under MacOSX 10.6.7 to an Epson 3880... (I even tried to reapply the 10.6.7 update as a combo update, with no change at all in the print :-( )

To be fair with that paper, I also tried the others I had from sample packs, and I observe the scum dots on at least Hahnemühle FineArt Photo Rag Ultra Smooth 305gsm (subtle) and on Canson Infinity Baryta Photo 310 gsm (very pronounced cyan border here !)

So, what's the solution here ?

Mark D Segal · « **Reply #19 on:** April 21, 2011, 07:18:00 pm »

Quote from: pompon44 on April 21, 2011, 05:53:41 pm

Hi,

Some followup.

I did contact Christophe Metairie, but the price of the profile (50-60 € depending on number of target patches, as far as I understand), made me think twice. So I made some further checks (e.g. changing the media type from semigloss to glossy or luster) to see if I should simply abandon that paper...

Turns out that in addition to having a slight magenta cast, the prints also exhibit what seems to be called "scum dots" (aka light gray/cyan border around the image, it's subtle but definitely there. More easy to see under daylight than under halogen light, but it's there !).

Problem is that, as far as I can tell, the profile (IGSPP11_EPP3880_PSPPn.icc) is a v2.0.0 one... So what can I do ? I'm printing from LR 3.3 under MacOSX 10.6.7 to an Epson 3880... (I even tried to reapply the 10.6.7 update as a combo update, with no change at all in the print :-( )

To be fair with that paper, I also tried the others I had from sample packs, and I observe the scum dots on at least Hahnemühle FineArt Photo Rag Ultra Smooth 305gsm (subtle) and on Canson Infinity Baryta Photo 310 gsm (very pronounced cyan border here !)

So, what's the solution here ?

Andrew Rodney (Digitaldog) has written a fair bit about the scum-dot issue including on this website, so you may wish to do a search and see what he advises. My understanding is that it becomes a problem with V4 profiles, not V2 profiles, so this may come as a surprise. You can get in touch with him - I expect he would be interested.

Author Topic: Magenta-ish skies with Ilford Galerie Smooth Pearl (compared to Epson Semigloss) (Read 8202 times)