Metamerism

[domenico Foschi]

This seems like the unspoken word in this forum.
Am I the only one being frustrated by this phenomenon?
Lately I got an order from an interior designer whose client requested some inkjet prints, and just because metamerism was so evident, I was very seriously considering stopping digital prints until something better would come out.
The client loved the images, but I hate to live with the apprehension that my images will look different according to different light sources.
I am not talking of B/W, but when I print files that have been sepia toned in PS, then I start seeing green, and magenta depending by the light source,...and more I look more intense it gets, obviously.
In the Meyerowitz ad on HP that has been circulating lately, he says that the new generation of HP printers have reduced this effect.
Is HP the only one?

[jetcode]

no metamerism with the ipf5100 - I checked and I know exactly what metamerism looks like - the ipf5100/6100 prints are drop dead gorgeous especially combined with one of the baryta papers - the ipf5100 software comes with the ability to warm or cool an image using color inks - no metamerism that I can detect though I have not printed anything as warm as you print - in this case I would go for maximum warm effect in software (you can adjust a fairly wide range of tone) and use a warmer paper base such as the Ilford Gold Silk

[Gordon Moat]

Every generation of printers addresses this problem. Even in high end very large printers using solvent inks or UV cured inks, there can be this problem. However, the important aspect is the perception of the final print. Just as some eyes can see more lp/mm than others, there are people with better colour acuity than others. In other words, you might be noticing it more than your clients/buyers notice it.

A separate issue is prints changing with aging. One unavoidable occurrence is changes to the paper, mostly due to OBAs, which have no known stabilizers. That could impart a yellowing over time. Also, inks will age at different amounts, so older prints could potential show a colour cast, or a shift in colour. Different than metamerism, though likely just as great a concern.

Ciao!

Gordon Moat Photography

[tim atherton]

A couple of things

Just about all "b&w" (toned or otherwise) exhibit some degree of metamerism, whether they are darkroom prints or inkjet prints.

As a photo archivist I have noticed some types of traditional prints which exhibit quite a significant degree of metamerism (some types of tones albumen/pop typed papers in particular). Although with most traditional prints the degree of metamerism isn't usually so great.

(it's also quite significant with certain Cibachrome colour prints as well)

The biggest difference on this issue between most traditional and most inkjet prints is that the metamerism in the former generally tends towards changes which (for whatever psychological reason) are generally felt to be "pleasant". That is, even when the metamerism is quite significant, the hues/colour usually change in a way that is agreeable to the eye. As such, the changes aren't often noticed.

Whereas with many inkjet ink/paper combinations, the metamerism - even when it doesn't change to a significant degree - more often seems to move in a direction which isn't nearly so agreeable to the eye. What I call "Magentarism" (which is the problem I find most disagreeable...) when you move from daylight to certain types of tungsten or fluorescent light.

That said, with the recent paper/ink combinations I've been using - Epson K3 pigment inks in a 3800 on paper like Crane Silver Rag, Crane Museo Portfolio Rag or Harman FB AI - I have found the metamerism is nowhere near as pronounced as it was with most other paper/ink combinations. And I find the changes - with the sort of printing I prefer - to tend much more towards the pleasant rather than unpleasant direction.

(I haven't done as much colour printing recently, but what I have done has also tended to display changes which are of the more agreeable type, when they are noticeable to my eye)

That said, I've usually preferred a slightly warmer than cooler print for b&w in both the darkroom and via inkjet. So I tend to print things just slightly warm - using either the Epson ABW or QTR. With that, I find that although there is a small amount of metamerism, it is of the sort that is more acceptable to the eye when you move to different light sources (even in the most annoying type of light - artificial mixed with daylight).

Now I recall that you produce quite strongly toned images?, so achieving the same in your case may be a bit more tricky...

(btw the Silver Rag has the advantage of no OBA's)


tim

[Ted Harris]

Dominico,

You didn't say which printer you are using. If it is an Epson then your best bet may be to spray your prints with one of the available sprays for this purpose.

Both Canon and HP have greatly reduced/almost eliminated metamerism in most cases by reformulating and further neutralizing the black/gray inks. Canon has attacked the problem by almost totally eliminating the use of any magenta coloration in their gray ink sets. HP's inks are also more neutral than Epsons and HP also uses a gloss enhancer which further reduces the problem.

As Tim says, you won't be able to totally eliminate it but you should be able to reduce it to a point where it is only noticeable at a very small number of angles and not usually at all when viewed normally. Paper choices are also part of the equation and you will be able to best tell which papers most meet your needs. I think you will find that any of the baryta based papers will help, especially the Ilford and Hahnemuhle which have more 'tooth' than the Harman.

[jetcode]

the B/W prints I get out of the ipf5100 are as good or better then the Septone system I have - daylight observation

[John O'Connell]

As an old textile dyer, I am familiar with metamerism and find the inkjet printer issues kind of humorous. "Curing" metamerism requires formulating a dye or pigment set to maintain a similar color across three light sources---"daylight," fluourescent, and incandescent. It is notoriously more difficult to do this with pigments, and it's usually the addition of a dye or pigment that shifts violet under one light as opposed to another that causes the most serious problems. (Diazo dyes, i.e. Ilfochrome, were also notoriously different than other dyes back in the day.) And the substrate has a serious effect on the matter---particularly in the inherent reflectance of the surface and the use of optical brighteners.

So because it seems difficult to test print a Macbeth chart or a reference photograph and then examine it under a lightbox prior to every print, I'm not sure what an individual art user can do, but I do know that it is easily possible to adjust for the end-user's intended light source, as long as you can find out the light source---though the increasing use of CFLs in homes makes the problem worse, as incandescent-to-fluourescent metamerism can be a real problem for some dye sets.

[Tyler Boley]

you did not mention what printer you are using. The difference in metamerism failure between say, the UCK2 printers like the 2200, and the UCK3 printers, like the 3800, 4800, etc., is considerable simply because the ink build in the neutral axis has far less color ink and far more LLK ink. Color profiles we make for use with a RIP for more monochromatic tinted printing have neutral inks all the way up the scale (100% GCR), makes a big difference. We've printed the same files on both systems (UCK2 and UCK3 100% GCR) and had them match under one light, and look radically different from each other under another. The very definition of metamerism failure.
I believe the new HPs do the same thing even with the standard driver, not sure what the Canons are doing.
At any rate, you are at a disadvantage of using a color printer to make a monochromatic print, sort of like doing a sepia print with type C paper, so either compromises are accepted, or more unusual setups might need to be employed.
Basically, as artists we tend to ask more from these products than their designers might have anticipated. Tim's excellent information reveals how we have been living with this for a long time.
Tyler

[Bruce Watson]

This seems like the unspoken word in this forum.
Am I the only one being frustrated by this phenomenon?
Lately I got an order from an interior designer whose client requested some inkjet prints, and just because metamerism was so evident, I was very seriously considering stopping digital prints until something better would come out.
The client loved the images, but I hate to live with the apprehension that my images will look different according to different light sources.
I am not talking of B/W, but when I print files that have been sepia toned in PS, then I start seeing green, and magenta depending by the light source,...and more I look more intense it gets, obviously.
In the Meyerowitz ad on HP that has been circulating lately, he says that the new generation of HP printers have reduced this effect.
Is HP the only one?

The real problem here is the laws of physics. The various color inks are the color they are because they reflect certain wave lengths of light and absorb the others. Clearly, as the light sources change, their spectrum changes -- they have more of certain wave lengths and less of others. This is the root cause of metamerism -- if there no light for the magenta inks to reflect, the print is almost forced to go cyan and/or yellow (that is, green).

If possible, use a fixed tone monochrome inkset (for example, one of the inksets that Cone makes). This will get you the minimum metamerism IIRC. If you must, use a variable tone inkset (for example, one of the inksets that MIS makes). The variable tone inksets use color inks for toners however, and may have somewhat more metamerism. Last choice is to make B&W prints using color inksets.

Finally, specify the light source that will give best results for the prints you make. This will give you something to fall back on if clients complain. But tell them all -- no florescent lights! Florescents are evil for print viewing IMHO.

[domenico Foschi]

Thank you all.
The images were printed with Epson Inks.

The info in this thread has been very helpful: Metamerism for inkjets is a complicated issue but not unsolvable.
I will investigate the effects of coating the paper as Ted Harris suggests, and I will talk to my Lab on how we can workout this problem for the benefit of both parties.
I am very pleased with the quality that a good scan can deliver even at in big enlargements , but the colors shift is a real let down.

Metamerism in silver prints is probably acceptable because ALL the tonalities change at the same rate, when instead an inkjet print is affected mainly in the midtones.
Am I wrong in my perception?