Digital BW, The Print

I am newcomer to digital printing and do not have much experience
with it. Unfortunately, I have never seen a good (or bad!) digital
print. I am overseas right now.

I was wandering what exactly is metamerism? How do you define it? And
does it manifest itself differently in color than in black and white?
I just read on this forum that dye inks don't have metamerism, but
pigment inks do. Why? Does metamerism  exist in traditional wet
printing techniques? I have never come across this term until diving
into the digital printing world.

I hope these basic questions won't bother you who are beyond defining
metamerism, but who rather trying to get around it!

Thanks for your help!

Jeff Hargrove
jeff@...
http://www.jeffhargrove.net

Hello,
I had an interesting experience today when I took some BW Ultrachrome K2 photos to show a friend who works under fluorescent lighting.

Some BW shots were done on QTR and they looked exactly the same under any lighting. But the Photoshop warm quadtone turned decidedly magenta. I'd chosen that printing method for a change because I liked that particular tone - and it was fine until I got under fluorescent tubes. It does limit the places you can display such prints.

Just an observation ...
Richard
--
http://smallfield.vze.com
http://photos.smallfield.vze.com (Photos web site)
http://warkworth.vze.com/ (Warkworth photo essay)


   "Nothing in the world can take the place of persistence. 
   Talent will not; nothing is more common than unsuccessful 
   men with talent. Genius will not; unrewarded genius is almost 
   a proverb. Education will not; the world is full of educated 
   failures. Persistence and determination alone are omnipotent."
   --Calvin Coolidge

Metamerism has become an un-expectedly significant issue in my experimenting with glossy neutral printing. I never had to worry about it with the 100% carbon printing approaches. 

In addition to the dyes for cards, one significant goal for me is to develop an appropriate glossy medium that can hang near my Eboni-6 matte prints and have them both look good under varying lighting conditions.  

The energy-efficient CF and LED lights are major contributors to the problems I'm having reaching this goal.  

While I knew the dyes reacted relatively poorly to fluorescent lights, the extent of this metamerism with the older fluorescent lights (CRI 75) surprised me.  They can have a major green/blue/cyan shift. 

I have also been surprised by the extent of the metamerism I see in prints made with neutralized carbon plus LC & LM.  There seems to be a significant reddish shift under the new LED lights.  Of course, the more color in the print, the greater the problem seems to be for B&W.


Is there an easy way to objectively measure and describe the degree of metamerism?




Paul
www.PaulRoark.com

Hi Paul,

What you are calling metamerism is actually color constancy. That disclaimer aside, a measuring device won't pick this up as it has a single, full spectrum light source. A camera will pick it up, but it's necessary to photograph a good, spectrally neutral gray ramp along with the image, in order to have a known set of neutrals to go withe the unknown variables. Cameras measure in RGB which is less than ideal, but you can get approximate Labvalues out of them in Photoshop, after normalizing each image under your assorted light sources to have the same luminance for white and black patches, and color balanced to the gray ramp. 

Since the normalization process noted above, and the target noted above, are exact matches to the SpyderCheckr, I might as well admit that, and just say: shoot your sample print, dead flat, in the same plane as the Checkr, and light both as evenly as possible. The follow the normalization directions for the Chreckr in Lifhtroom or Adobe Camera Raw, without cropping, and you can then export to Photoshop, where the eye dropper tool will supply Lab values for your ink gray ramps. 

If you don't have a SpyderCheckr, send me an email off list and we will work on remedying that. 

C. D. Tobie
Global Product Technology Mngr.
Imaging Color Solutions
Datacolor.com
CDTobie@...

On Apr 17, 2012, at 1:26 PM, "Paul" <roark.paul@...> wrote:

> Metamerism has become an un-expectedly significant issue in my experimenting with glossy neutral printing. I never had to worry about it with the 100% carbon printing approaches. 
> 
> In addition to the dyes for cards, one significant goal for me is to develop an appropriate glossy medium that can hang near my Eboni-6 matte prints and have them both look good under varying lighting conditions. 
> 
> The energy-efficient CF and LED lights are major contributors to the problems I'm having reaching this goal. 
> 
> While I knew the dyes reacted relatively poorly to fluorescent lights, the extent of this metamerism with the older fluorescent lights (CRI 75) surprised me. They can have a major green/blue/cyan shift. 
> 
> I have also been surprised by the extent of the metamerism I see in prints made with neutralized carbon plus LC & LM. There seems to be a significant reddish shift under the new LED lights. Of course, the more color in the print, the greater the problem seems to be for B&W.
> 
> Is there an easy way to objectively measure and describe the degree of metamerism?
> 
> Paul
> www.PaulRoark.com 
> 
> 


[Non-text portions of this message have been removed]

Hi Paul. Along with C D's comments and suggestions, I fear you are looking at spectral deficiencies in the light sources themselves. I'm not sure how you can overcome this. It's just an extended example of printing to match the display conditions directly because one can't have it all. While you might be able to adjust ink formulations to maintain visual (but not instrument measured) neutrality with a particular CF lamp brand or even a range of them, how can you plan for an unknown or later version of such or a different manufacturing run. As for LEDs the spectral distribution is up and down all over the place and very lumpy indeed.Even worse is the fact that different manufacturing runs of LEDs will vary quite significantly so any change at all in lamps or image placement on display could have serious impacts even if you could compensate for a single circumstance.It's extremely doubtful that affordable lighting would be very consistent from lamp to lamp over time either.Reference LEDs must be closely temperature and current controlled if expected to be constant in spectral output.
If you have access to an emission spectrophotometer you can measure the lamps and see what I'm referring to in their spectral curves.

Unfortunately our eyes are extremely sensitive to minute color deviations with truly neutral objects. The best ,if one can call a compromise that, case might be to attempt to adjust the inks such that the lack of constancy bends in a pleasing direction and trust that the physiological aspects of human vision will handle things as well as possible.

There's a lot of work in your task I believe.

Regards,
Duane


--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Paul" <roark.paul@...> wrote:

>
> Metamerism has become an un-expectedly significant issue in my experimenting with glossy neutral printing. I never had to worry about it with the 100% carbon printing approaches. 
> 
> In addition to the dyes for cards, one significant goal for me is to develop an appropriate glossy medium that can hang near my Eboni-6 matte prints and have them both look good under varying lighting conditions.  
> 
> The energy-efficient CF and LED lights are major contributors to the problems I'm having reaching this goal.  
> 
> While I knew the dyes reacted relatively poorly to fluorescent lights, the extent of this metamerism with the older fluorescent lights (CRI 75) surprised me.  They can have a major green/blue/cyan shift. 
> 
> I have also been surprised by the extent of the metamerism I see in prints made with neutralized carbon plus LC & LM.  There seems to be a significant reddish shift under the new LED lights.  Of course, the more color in the print, the greater the problem seems to be for B&W.
> 
> 
> Is there an easy way to objectively measure and describe the degree of metamerism?
> 
> 
> 
> 
> Paul
> www.PaulRoark.com
>

Paul:
Welcome to the "real world"! ;-)

Many of those LED lights are really solid state fluorescent lights without the tube. To make them "warmer" to match incandescent light bulbs, they are using a phosphor mix similar to fluorescent tubes. (native LEDs have a very cool color) 

I have noticed some of the "spiral" CF lights tend to exaggerate red balance also. Still hard to beat incandescent bulbs  for full spectrum balance.

My initiation into this cunundrum came when trying to match a B&W print done on color photo paper that had been damaged. I could match it under incandescent, or daylight, but not both. In the end, made the print using the B&W mode of the Z3100, which stayed neutral under all lighting conditions, but did not match the original chromagenic paper B&W print. (is that an oxymoron? ;-))

John Nollendorfs

"dlruckus" <dlruckus@...> wrote:
>
> ... I fear you are looking at spectral deficiencies in the light sources themselves.


Yes, that is certainly the case.

> I'm not sure how you can overcome this.

Perhaps not completely, but the 100% carbon images seem to hold up very well under all the lighting I've tried.  The light's deficiencies are not particularly noticeable with a medium that does not itself consist of a "neutral" that is actually an R, G and B image that is balanced to look neutral.  It's that balancing of the colors that is impossible under the varying, un-even light sources.

So, I'm really back to where I started and what we know in the context of the lightfastness also - 100% carbon is the ideal, and moving away from this, the more carbon and less color we use the better off we are; nothing really new here.  What surprised me was, first, the large visual difference an apparently small difference in CRI made for the dyes, and, second, the extent of carbon + color image sensitivity under the LEDs.

So we know if the Epson printers with more colors are incorporating them into their ABW mode printing?  I wonder if spreading the toning function to more, different color inks helps.


> It's just an extended example of printing to match the display conditions directly because one can't have it all.


And printing to match my gallery's lighting as well as my home lighting are the top criteria for me, but I also don't want a customer to get a print home an find it looks green inder their lighting.

With respect to my gallery, I took what I though were the best compromises with the dyes and the MIS glossy carbon plus Lm and Lc to the gallery for an inspection under that (poor) fluorescent lighting.  The dyes just did not cut it.  Thus ends the Epson 4000-dye experiment.  Luckily, the MIS glossy carbon plus Epson Lm and Lc looked very good.  So, that is currently looking like my best path forward for neutral glossy printing for gallery display.  (The dyes will remain for cards and other high gloss output, but only the Epson 1400 will be used.)


> ... As for LEDs the spectral distribution is up and down all over the place and very lumpy indeed.

They are worse that I thought.  I wonder if those using the Epson ABW mode printing have noticed problems with the new LED lights. 


>... It's extremely doubtful that affordable lighting would be very consistent from lamp to lamp over time either....


I think this all reinforces my belief that 100% carbon on no-OBA paper is the safest medium for fine art.  But that doesn't give me the glossy neutral I'm after.  So, I'll continue to follow what appears to be the most promising path to that goal (as well as the 100% carbon matte prints, of course).

 
> Unfortunately our eyes are extremely sensitive to minute color deviations with truly neutral objects. The best, if one can call a compromise that, case might be to attempt to adjust the inks such that the lack of constancy bends in a pleasing direction and trust that the physiological aspects of human vision will handle things as well as possible.


Hitting that best compromise seems to be the trick with the color + carbon inksets. It may be rather venue specific, which may not bode well inflexible systems.

At the moment, I'm rather happy with what I've seen with the MIS LK plus LM and LC.  It's not in the same league as the 100% carbon, but I was able to profile it well enough to look good in the venues and under the lights I have.  I'm inclined to follow that path for neutral  glossy display printing at the moment.

Paul
www.PaulRoark.com

Paul,

It becomes a journey between Scylla and Charybdis if you want to avoid 
gloss difference, bronzing, color inconstancy and lay no restrictions on 
the viewing conditions.

It now looks like all the dyes types have issues with color constancy in 
monochrome prints, more than black pigments have. LEDs and CFLs are not 
without spectral pikes. The question is whether the higher transparency 
of dyes and by that the kind of reflective filtering plays a role 
compared to carbon pigments or whether the black dye colorant spectral 
reflection is irregular. You could check a spectral plot of the dye made 
with a normal (non UV-cut) spectrometer. The smallest wavelength 
interval measurement for an Eye1 Basic would be possible with ArgyllCMS. 
You might need a spectroradiometer to locate smaller deviations in 
spectral reflectance, absorption, transmission. I think that is the 
basic approach to tackle color inconstancy at the print side. You have 
to find a neutral medium to print the colorants on. Yes it would be nice 
to check the carbon pigments too then and your ink mediums.

You did not mention the paper you used. OBAs remain an issue anyway but 
I guess that the different filtering of dyes compared to the more opaque 
carbon pigment enhances the OBA effects. If carbon pigment was totally 
opaque the color mixing of greys (black + paper white) could actually be 
called additive. Transparent black dyes + paper white still have a 
subtractive color mixing base and all the irregularities in light 
transmission + reflection of both are added to one another. The more 
continuous and flat the spectral reflectance of the paper white is, the 
less the colorants different transparencies affect the reflectance of 
grey area. That said if the dye itself is not showing an irregular 
spectral reflectance, including fluorescence. Carbon pigment will not 
show fluorescence and will suppress possible fluorescence in the other 
components by absorbing UV and what is emitted in energy from 
fluorescents. TiO2 whitening agent in the paper could reduce UV excited 
fluorescence in other components of the print too as it absorbs UV and 
emits the energy beyond the visible spectrum.

White LEDs can be based on UV or Blue LEDs immersed in (yellow) 
phosphors. For normal lighting the RGB based White LEDs are too 
expensive. Next to spectral spikes there are other issues.  Some of that 
is discussed in a document that I have no link for, so do a google for: 
LpR10_Instrument_Systems_web.pdf  The Wiki pages on LEDs are good too.

Better wait for Quantum Dot LEDs if efficient, full spectrum lighting is 
needed. Till then halogens are best.


-- 
Met vriendelijke groet, Ernst

Dinkla Grafische Techniek
Quad, pi\ufffdzografie, gicl\ufffde
www.pigment-print.com

Ernst Dinkla <e.dinkla@...> wrote:
>
> ...
> It now looks like all the dyes types have issues with 
> color constancy in monochrome prints, more than black
> pigments have.

My observations are consistent with that.  Still, I'd say for many if not most uses, the dyes will still look better on high gloss media.

> LEDs and CFLs are not without spectral pikes. 

The dyes seem to do better with LEDs and carbon pigments seem to do better with the fluorescent lights.  

Most of the CFs here now carry an "energy star" label which requires a CRI of 80 or higher.  See http://www.energystar.gov/index.cfm?c=fixture_guide.pr_fixtures_guide_lightquality .  The problem fluorescent lights I ran into with the gallery are older style, with a 75 CRI.  So the extent of the problem I ran into may be much less applicable to most people.

> ... whether the black dye colorant spectral reflection is irregular.


I think that is a significant part of the problem.  There is a green/cyan bias to the black inks, the peak of which may be too close to a similar peak in the fluorescent light spectrum.

> You did not mention the paper you used. OBAs remain an issue ...

I tried a number of papers.  The lower the OBA amount, the less the problem but even on Museo the dyes shifted colors too badly.

The glossy print that finally looked good enough to hang next to the Eboni-6 prints was the MIS glossy carbon plus LM and LC on Museo.


> Better wait for Quantum Dot LEDs if efficient, full spectrum 
> lighting is needed. Till then halogens are best.
> 

A combination of government energy efficiency regulations and economics is forcing the issue here.  I think "hot wire" technology is out of the question.  The current LEDs appear to be better than the CFs already.  We will just have to watch our profiling to be sure we don't end up with reddish prints.

Paul
www.PaulRoark.com

On 2012-04-18, at 10:48 AM, "Paul" <roark.paul@...> wrote:

> The glossy print that finally looked good enough to hang next to the Eboni-6 prints was the MIS glossy carbon plus LM and LC on Museo.

Paul, what would prevent you from considering the diluted HP PK ink set for these scenarios?

Terry.

Terry Ritz <t.ritz@...> wrote:
>
> ... what would prevent you from considering the diluted HP PK ink set for these scenarios?


The HP gray or HP PK diluted with the generic base has serious bronzing and gloss differential.  The HpGO mix (HP PK diluted with MIS glop/amber base) that I was so enthused about in the past turned out to look very bad under the LEDs.  Whether we like them or not, these LEDs and CFs are going to be everywhere.  So, I think we need to adjust our printing approaches to what seems to be an inevitable viewing environment.

Paul
www.PaulRoark.com

On 04/18/2012 04:48 PM, Paul wrote:


> Most of the CFs here now carry an "energy star" label which requires a
> CRI of 80 or higher. See
> http://www.energystar.gov/index.cfm?c=fixture_guide.pr_fixtures_guide_lightquality
> . The problem fluorescent lights I ran into with the gallery are older
> style, with a 75 CRI. So the extent of the problem I ran into may be
> much less applicable to most people.

CRI numbers are like Lab values for paper white, they tell very little 
about the spectral distribution and flatter manufacturer's specifications.


>  > You did not mention the paper you used. OBAs remain an issue ...
>
> I tried a number of papers. The lower the OBA amount, the less the
> problem but even on Museo the dyes shifted colors too badly.

No OBA in them as far as I measured their catalogue.


> A combination of government energy efficiency regulations and economics
> is forcing the issue here. I think "hot wire" technology is out of the
> question. The current LEDs appear to be better than the CFs already. We
> will just have to watch our profiling to be sure we don't end up with
> reddish prints.

As written the Quantum Dot LEDs promises all we would like to see.
There is also the yellowing issue, I guess cataract removal will be 
needed for all the veterans here to enjoy that light even more and no 
need for any OBA papers after that :-)



-- 
Met vriendelijke groet, Ernst

Dinkla Grafische Techniek
Quad, pi\ufffdzografie, gicl\ufffde
www.pigment-print.com

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, Ernst Dinkla <e.dinkla@...> wrote:

> On 04/18/2012 04:48 PM, Paul wrote:

> CRI numbers are like Lab values for paper white, they tell very little
> about the spectral distribution and flatter manufacturer's specifications.

And yet despite known limitations, both CRI values and L*a*b* ratings (using
UV-included and UV excluded measurements) have great practical merit. I've never
seen an FL or other type of lamp with CRI rating under 90 that provides a better
color viewing environment than Full spectrum fluorescents with 90+ CRI ratings.
Similarly, the Lab data I provide on the AaI&A website for media white points,
especially the delta b* values which compare the difference between UV included
and UV-excluded b*), correlate very well with Ernst's database of spectral
plots on media white performance. Both reporting methods reveal high, medium,
low, or no OBA content levels, for example.

cheers,
Mark
http://www.aardenburg-imaging.com

Obviously be aware  that not all LED's are created equal. The market is being flooded by a lot of "cheap" Chinese products. Unless you can get full disclosure of CRI, and or a spectral graph, be wary of buying these bulbs. Test them first if viewing photographs is your intention. Also different batches are likely to vary. 

But I was very pleased with the Z3100's B&W mode printing results in images that don't seem to vary under different light sources. But add just a bit of color, to change the tone. . .  

John Nollendorfs

On 04/18/2012 09:33 PM, Mark wrote:
>
>
> --- In DigitalBlackandWhiteThePrint@yahoogroups.com
> <mailto:DigitalBlackandWhiteThePrint%40yahoogroups.com>, Ernst Dinkla
> <e.dinkla@...> wrote:

>  > CRI numbers are like Lab values for paper white, they tell very little
>  > about the spectral distribution and flatter manufacturer's
> specifications.
>
> And yet despite known limitations, both CRI values and L*a*b* ratings (using
> UV-included and UV excluded measurements) have great practical merit.
> I've never
> seen an FL or other type of lamp with CRI rating under 90 that provides
> a better
> color viewing environment than Full spectrum fluorescents with 90+ CRI
> ratings.
> Similarly, the Lab data I provide on the AaI&A website for media white
> points,
> especially the delta b* values which compare the difference between UV
> included
> and UV-excluded b*), correlate very well with Ernst's database of spectral
> plots on media white performance. Both reporting methods reveal high,
> medium,
> low, or no OBA content levels, for example.
>
> cheers,
> Mark
> http://www.aardenburg-imaging.com

Mark,

A bit harsh, my statement. At least for the Lab paper white values, I 
should have referred to paper manufacturer's Tappi and ISO values for 
brightness which are normally used to indicate the reflectance of paper 
in the specifications. Lab tells a lot more and is often not included in 
the specs. However on this thread's subject of "metamerism" a spectral 
plot says more than both Tappi and Lab numbers.

For CRI there are well known flaws. Your Aardenburg-Imaging print 
testing method increased the number of colors compared to other testing 
methods. On good grounds. CRI could do with a similar upgrade and use a 
wider range of pigments.

For as long as it exists I trust www.olino.org on its test methods for 
lamps and they have good explanations on the methods and standards. In 
the article on CRI:
http://www.olino.org/us/articles/2009/11/30/a-close-look-at-the-color-rendering-index-cri-or-ra
there is a statement at the bottom, from a CIE report:

 >>The result of the three investigations show that there is a low to no 
correlation at all between the computed CRI and the visually determined 
differences in color. The TC comes to the following remark:
the CIE CRI is in general not applicable for a ranking on color 
rendition when white led sources are part of the observed illuminants.
The conclusion of the TC is a recommendation to develop a new color 
rendition index (or maybe a set of CRIs). In the beginning existing in 
parallel to the current one (which needs to give information obout the 
color comparison. Also this new CRI needs to be valid for all kinds of 
light sources.<<

which is obvious if you just check the spectral plots for similar CRI 
values in the article.

That site has to my knowledge the best tests on LEDs etc. Dutch origins 
I believe so something to be proud off. Hard to navigate in its pages 
though.

-- 
Met vriendelijke groet, Ernst

Dinkla Grafische Techniek
Quad, pi\ufffdzografie, gicl\ufffde
www.pigment-print.com

Ernst Dinkla <e.dinkla@...> wrote:

> ... 
> As written the Quantum Dot LEDs promises all we would like to see.
> ...

I hope you're right, but the Wiki article is not encouraging with respect to a smooth spectral distribution: "Quantum-dot-based LEDs are characterized by pure and saturated emission colors with narrow bandwidth ..."  http://en.wikipedia.org/wiki/Quantum_dot_display

But, I'm sure there are lots of people working to perfect these light sources.

Paul
www.PaulRoark.com

On 04/19/2012 04:47 PM, Paul wrote:
> Ernst Dinkla <e.dinkla@...> wrote:
>
>  > ...
>  > As written the Quantum Dot LEDs promises all we would like to see.
>  > ...
>
> I hope you're right, but the Wiki article is not encouraging with
> respect to a smooth spectral distribution: "Quantum-dot-based LEDs are
> characterized by pure and saturated emission colors with narrow
> bandwidth ..." http://en.wikipedia.org/wiki/Quantum_dot_display
>
> But, I'm sure there are lots of people working to perfect these light
> sources.


Check what QD Vision already has. There is a wider choice of colors to 
mix with different QD sizes and the conversion to visible light gives 
less energy loss than the phosphors so far could. The conversion is not 
into a narrow bandwidth. Cadmium free components too.

But a more interesting development is this discovery:
http://www.vanderbilt.edu/exploration/stories/quantumdotled.html
were the QDs are so small they excite a broader spectrum white light 
right away.

-- 
Met vriendelijke groet, Ernst

Dinkla Grafische Techniek
Quad, pi\ufffdzografie, gicl\ufffde
www.pigment-print.com

Ernst Dinkla <e.dinkla@...> wrote:
>
> ...
> 
> Check what QD Vision already has.

http://www.qdvision.com/solid-state-lighting  

It all sounds promising.  I suspect they are just too expensive to be in my local Costco.  Then again, I have no idea what technology the ones I've tested use.



> http://www.vanderbilt.edu/exploration/stories/quantumdotled.html

From 2005, but still  interesting.  I particularly like the way the student slapped  one of these together.

Paul
www.PaulRoark.com

On 04/19/2012 08:15 PM, Paul wrote:

>  > http://www.vanderbilt.edu/exploration/stories/quantumdotled.html
>
>  >From 2005, but still interesting. I particularly like the way the
> student slapped one of these together.
>
> Paul
> www.PaulRoark.com

There must be a can of Hydrocote in your garage. The difficult part is 
the 30 atoms Quantum Dots. They can make them in an hour.

Sandra Rosenthal's group published more articles on the subject since, I 
am sure this is not a dead end. Even more interesting is that there is a 
chance the same effect is achieved in powering them electrically instead 
of powering them optically. Quantum Dots are the subject of a lot of 
research right now, production is scaled up, Nanoco in the UK does that 
and recently a Dutch university spin-off made a deal with a US 
manufacturer for another process.

-- 
Met vriendelijke groet, Ernst

Dinkla Grafische Techniek
Quad, pi\ufffdzografie, gicl\ufffde
www.pigment-print.com