Which color temp lights to use to view prints.

[denalilap]

I’ve been trying to improve the match between my monitor and the prints. I have a Benq sw2700pt (calibrated using i1display pro, pallete master, followed by DisplayCal). I’ve been viewing prints using an Ott light, with unsatisfying results. I recently installed track lights to illuminate photos on wall. I looked at prints under Solux 3500k lights and the match is exceedingly better. Still not all the way there but much better.

The above prompted me to go back to the Solux site and read:

The Guggenheim Museum uses SoLux 4700K for critical color work on art. Most photographers choose SoLux 4700K and 5000K light for critical color proofing, however, other well respected photographers like John Paul Caponigro recommend using SoLux 3500K for proofing, the light many galleries use to exhibit. Therefore, we now offer the full range of SoLux in our light kits.

Thus, my question... which color temp bulbs should one get and which temp for a given purpose.

[digitaldog]

Short answer; doesn't matter. The numbers define a large range of possible colors. Pick a Solux you visually prefer and yes many agree with JP by going a lot lower than CCT 5000K.

[faberryman]

Short answer; doesn't matter. The numbers define a large range of possible colors. Pick a Solux you visually prefer and yes many agree with JP by going a lot lower than CCT 5000K.

You really don't have any control over what the lighting will be in galleries and other venues. So get the 3500 or the 5000 and work on consistency of you prints.

[nemophoto]

I use a JUST Normlicht lightbox for viewing prints. I believe it's 5000K (been a while since I looked at the specs). To a degree, Andrew is correct, it doesn't matter, though I find for me it does serve a purpose. I at least know if the color balance is correct in a controlled situation. A while back I had a client who was getting crap pre-press from a pre-press house. Highlights were blown, color was off, etc. They tried to blame me (of course). I printed match prints for a while and FedEx'd them off with each job. I said that if your proofs don't match these, then your pre-press house is screwing you. Eight years later, I still shoot for the client, do the digital imaging and the pre-press house is history.

The moral of the story is, a calibrated lightbox is good to have when you are dealing with color critical issues and for your own peace of mind. Once it's out of your hands, anything is possible.

[denalilap]

My question was too vague... What I really should have asked is:

If you were calibrating your monitor to D50 (for example) which light temp would give  you the best match to what’s on the monitor. Same for D65.

[rasworth]

denalilap,

IMO it doesn't pay to become overly concerned with numbers/specs, rather one should pay attention to the process, and here is my recommendation:

1. Set up a viewing environment, good illuminants (Solux qualifies), and reasonably bright.

2. Print a reference image thru a properly color managed system, on your printer.  I use PrinterEvaluationImage_V002_ProPhoto.tif.

3. Now the hard part.  Your system display should be in a somewhat dim (not dark) environment, with the print viewing illumination nearby but not shining on the screen.  Calibrate and profile your display until you achieve a reasonable visual match between it and the print, with the evaluation image loaded into Photoshop in soft proof mode thru the same printer profile used to produce the reference image.  Concentrate mostly on brightness and tone, you'll drive yourself nuts trying to exactly match all colors.

That's it, other than perhaps reading Andrew Rodney's tutorial on such.

Or you can obsess over degrees Kelvin, illuminant brand, etc.  Nothing wrong with paying attention to the details, but remember the goal - produce an image on your display that will enable to you create a good print on the first pass (ok, I confess, sometimes I re-print).

Richard Southworth

Added by edit - I have my monitor calibrated to D50 chromaticity, and very dim, 50 L.  This isn't supposed to work but I'm weird.

[digitaldog]

Definitely SoLux 4700K or 5000K; adjust your calibration color temp and brightness to get the best match.

Nope; Definitely NOT!  First of all, the two (CCT 4700K and 5000K) are the same, the 5000K runs hotter and dies sooner, just so you end up with a silly number. Next, as you've been told in the past, the numbers are both meaningless, define a range of possible colors are more often than not, don't look as appealing as lower values.
The OP should contact Solux and ask, as my clients like Art Wolfe did, if they would ship several differing bulb values, then after actually examining the illuminant, pick which they desire. Like JP, Art vastly preferred the CCT 3500K bulbs and selected them for his entire gallery in Seattle.

[digitaldog]

My question was too vague... What I really should have asked is:

If you were calibrating your monitor to D50 (for example) which light temp would give  you the best match to what’s on the monitor. Same for D65.

The only object that actually produces D50 is 93 million miles from any display. 
These Standard Illuminants were produced from hundreds of actual measurements around the world, then averaged.
To answer your question, the real need to match a display to an illuminant is for viewing the print next to the display when soft proofing as outlined in my video below. Now you can use the same Solux system that you'll light the prints elsewhere or maybe you will not; doesn't matter. Your eyes adapt to the illuminant. Assuming it's a half decent Light source. You'll adjust the WP of the display calibration to produce a visual match. So it doesn't matter if the Solux bulbs are CCT 3500 or 4700 (and do not even think about wasting your money on those labelled 5000K).


See:

Why are my prints too dark?
A video update to a written piece on subject from 2013

In this 24 minute video, I'll cover:
Are your prints really too dark?
Display calibration and WYSIWYG
Proper print viewing conditions
Trouble shooting to get a match
Avoiding kludges that don't solve the problem

High resolution: http://digitaldog.net/files/Why_are_my_prints_too_dark.mp4
Low resolution: https://youtu.be/iS6sjZmxjY4

[digitaldog]

denalilap,

IMO it doesn't pay to become overly concerned with numbers/specs, rather one should pay attention to the process, and here is my recommendation:

1. Set up a viewing environment, good illuminants (Solux qualifies), and reasonably bright.

2. Print a reference image thru a properly color managed system, on your printer.  I use PrinterEvaluationImage_V002_ProPhoto.tif.

3. Now the hard part.  Your system display should be in a somewhat dim (not dark) environment, with the print viewing illumination nearby but not shining on the screen.  Calibrate and profile your display until you achieve a reasonable visual match between it and the print, with the evaluation image loaded into Photoshop in soft proof mode thru the same printer profile used to produce the reference image.  Concentrate mostly on brightness and tone, you'll drive yourself nuts trying to exactly match all colors.

That's it, other than perhaps reading Andrew Rodney's tutorial on such.

Or you can obsess over degrees Kelvin, illuminant brand, etc.  Nothing wrong with paying attention to the details, but remember the goal - produce an image on your display that will enable to you create a good print on the first pass (ok, I confess, sometimes I re-print).

Richard Southworth

+1

[denalilap]

Thanks for all the responses. The difference in viewing between the Ott and the Solux is what got me thinking about the temp of the illumination. I’ll stick with the Solux (since I like the way it illuminates my office) and go back to tweaking the calibration for a better match.

Thanks

[digitaldog]

What happens after that under different lighting conditions is way, way less critical thanks to a phenomenon called "Color Constancy" or "discounting the illuminant"; this is very well explained in Real World Color Management by Fraser, Murphy and Bunting, a book I cannot recommend enough.

It is of course well explained and dismisses what you wrote, the opposite of what I did in terms of the numbers:
page 48
A final point about color consistency and color management:color constancy presents and argument that the color temperature of lighting isn't as important as some people (Franz?) think. it's still is important but not at the expense of everything else. For example, when calibrating your monitor, to a D50 or D65 white point. Many people choose D50 in order to match the exact white point color of viewing environments which is usually a D50 lighting booth. But we, alone with many other practitioners recommend D65 because we think you'll be happier looking at a bright D65 monitor rather than a dingy yellow D50 one, even if D50 is colorimetrically closer to D50 lighting.

Now this is important:
Matching brightness levels between two viewing environments may be as important as, or even more important than matching the color temperature - color constancy does a lot to adapt to slight differences in color temperature.


Adapt to the illuminant; as I outlined! Ignore the numbers because someone incorrectly believes this large range of values is closer to something else (it's not). Again, this fine book dismisses so much of what you wrote a few years ago about Solux over on PhotoNet. URL of that silliness provided earlier. And what you wrote today. 

[digitaldog]

Thanks for all the responses. The difference in viewing between the Ott and the Solux is what got me thinking about the temp of the illumination. I’ll stick with the Solux (since I like the way it illuminates my office) and go back to tweaking the calibration for a better match.

Thanks

The Ott light sucks like so many illuminants once you examine it's spectrum. Which is really important and something that requires special software and hardware.
http://digitaldog.net/files/15TheRightLightpart1.pdf

[digitaldog]

Blah, blah, blah...

“There are two ways to be fooled. One is to believe what isn't true; the other is to refuse to believe what is true.” -Søren Kierkega\
We both agree that Bruce and Chris's book is excellent, but only one of us, as illustrated, understands it.

[digitaldog]

The 5000K runs hotter, has a higher color temp and lasts not as long, but lasts for years in my setup.

Got actual proof? Didn't think so. And why pay for a bulb that burn's out faster just for a number that's meaningless? No reason.

Not a silly number as I prefer the results with the 5000K over the 4700K.

Got any proof other's should? Didn't think so.

If lower values look more appealing all by themselves is an issue of personal taste/preference.

As multiple people have reported.

The OP clearly stated that he wants to match the monitor to the print viewed in the digital darkroom and in my experience the 4700K or 5000K give a way better match than lower values; 3500K for instance forces me to lower the color temp of the monitor so much that I end up with a dingy yellow looking monitor.

Utter nonsense. As other's have reported. Myself, Art Wolfe, JP; no such issues.

What lighting to use for displaying your prints for others to see is a totally different issue and is very subjective; Guggenheim likes 4700K, Wolfe and Caponigro like 3500K.

And you stated, incorrectly again I'll add:

Definitely SoLux 4700K or 5000K; adjust your calibration color temp and brightness to get the best match.

No, it's not definitely; you've now changed your tune that it's subjective and that different people definitely do NOT want CCT 4700K or 5000K.

[digitaldog]

Blah, blah, blah...

Your sad and repetitive answer to facts.

[digitaldog]

More facts.

Mark D. Fairchild's Color Appearance Models, page 29
Color constancy refers to the everyday perception that the colors of objects remain unchanged across significant changes in illumination color and luminance level. Color constancy is served by the mechanisms of chromatic adaptation and memory color and can easily be shown to be very poor when careful observations are made.
Discounting the illuminant refers to an observer's ability to automatically interpert the illumination conditions and perceive the colors of objects after discounting the influences of illumination color.

Page 156:
6.12 Color Constancy?
Color constancy is another phenomenon that is often discussed. Typically color constancy is defined as the apparent invariance in the color appearance of objects upon changes in illumination. This definition is somewhat misleading. The main reason for this is because color constancy does not exist in humans! The data presented in the previous sections of this chapter and the discussion of chromatic adaptation in Chapter 8 should make this point abundantly clear.

[Alan Klein]

The 5000K runs hotter, has a higher color temp and lasts not as long, but lasts for years in my setup. Not a silly number as I prefer the results with the 5000K over the 4700K. If lower values look more appealing all by themselves is an issue of personal taste/preference. The OP clearly stated that he wants to match the monitor to the print viewed in the digital darkroom and in my experience the 4700K or 5000K give a way better match than lower values; 3500K for instance forces me to lower the color temp of the monitor so much that I end up with a dingy yellow looking monitor.

What lighting to use for displaying your prints for others to see is a totally different issue and is very subjective; Guggenheim likes 4700K, Wolfe and Caponigro like 3500K.

It seems to me that 3500 is liked better by people.  It's why I lit my home in warmer light with lower Kelvin values.  4700 or 5000 is just too cold.  While 3500 may not bring out "exact" color in prints, people are just happier in a warmer environment. People adjust to any color shifts brought on by a change in Kelvin.   

So if you're selling something like pictures in a gallery, you want prospective clients to feel comfortable, warm and cozy and not let them think they're in a medical clinic or workshop. 

[Doug Gray]

It seems to me that 3500 is liked better by people.  It's why I lit my home in warmer light with lower Kelvin values.  4700 or 5000 is just too cold.  While 3500 may not bring out "exact" color in prints, people are just happier in a warmer environment. People adjust to any color shifts brought on by a change in Kelvin.   

So if you're selling something like pictures in a gallery, you want prospective clients to feel comfortable, warm and cozy and not let them think they're in a medical clinic or workshop.

Yeah, me too.

But it sucks for comparing prints to a monitor. For that matter, ever try profiling your monitor to 3500K? It looks pretty horrible! Yet lighting a room at 3500K or even 2800K is pleasing. Why, one might ask?

The effect is beyond adaptation. It is, to a large degree, psychological. We get accustomed throughout our life to expect certain "whites' and "colors" to be certain ways. We all have grown up looking at monitors/TVs that are set to 5500K to 7500K. And when we look out the window we see the same, higher CCT level view. But scenes from the window just look brighter, not so much bluer. Even when our homes are illuminated with 2800K.

If you have a white sheet of paper on your desk and measure it's color in CIE XYZ and duplicate that exact same color and size of paper on your monitor it will not look the same as the one on your desk. The monitor's image will look like a yellowish brown and very different than the paper on your desk.

But if you pick up the paper and bring it close to the monitor, there is a sudden, freaky, shift when it gets within a few inches of the monitor and suddenly the colors will blend. Every time I do that experiment or show it to others, it is just mind boggling.

This phenomena is also described in the Wiley book: "Color Appearance Models" in Chapter 8.3

Perception of color is highly complex combination of adaptation based on the brightest surrounding light colors, as well as "known" colors, such as a white sheet of paper. And other knowns such as our expectations of what white looks like on a monitor.

There are ways of suppressing this. For instance camouflaging the monitor such that it isn't apparent we are seeing a screen and setting the light levels to be the same as what room surface reflecting light would produce.

[Rand47]

This is kind of an “aside” to this conversation.  I have a little Fiilex V70 lamp that cycles 3000, 4000, 5000 and 6500k light.  I have both a GTI booth at 5000k and a gray magnetic 4x8’ wall illuminated at 4700k Solux track lights, for my print evaluation use.  Once I have a final print made, I’ll often show someone what it looks like at 3000, 4000, 5000, and 6500k, and at differing luminance, just as an “education” bit to illustrate to them that “where and under what lighting” they display the print will make a significant difference in how it looks. 

Rand

[denalilap]

OP here, I watched Andrew's video (see his earlier post) and the key take-away is that the monitor color temp and luminance needs to be tweaked to match the print (using a reference image). Must webpages I've seen that explain calibration tend to recommend specific values and never mention that they need to be tweaked. Andrew's video was very helpful.