Print Evaluation Lighting

[Kit-V]

In my quest to dial in the calibration of my display, I can appreciate the role that ambient lighting plays when evaluating the final print to the on-screen soft proofed image. There seems to be considerable discussion regarding the "optimal" color temperature of the ambient lighting for print evaluation. Which gives rise to my question: Should the print be evaluated under a full-spectrum daylight bulb (i.e. 5,000 - 6,000K) or under the lighting that the prints will eventually be displayed & viewed? Of course, the second part of my question might be difficult to determine.

Thank you for help.

Kit

[digitaldog]

There is so much to address in your question, just start here first:
http://digitaldog.net/files/15TheRightLightpart1.pdf
http://digitaldog.net/files/16TheRightLightpart2.pdf
Yes, full spectrum, meaning your options are tiny (Solux) other than that superb actual full spectrum daylight produce 93 million miles from any print.
You want to soft proof and see a match of display and print (as close as possible)? Well that print needs to be next to that display. After that, any half decent print that doesn't suffer metameric failure will look fine anywhere else viewed under decent (not a 4 watt nightlight bulb) illuminant: your eye adapts to it.

[Kit-V]

Andrew: Thank you for the response & the links. I like Joseph Holmes' homemade lighting system. Assuming that one uses the components that he recommends, it seems like the light fixture can be mounted in alternative positions (other than the ceiling), as long as the fixture is approximately 48" from the print.

Since I really do not require a professional print viewing station, I am also curious about using an Ott-Lite (5,300K) lamp for occasional print evaluation. Although the color temperature seems correct, my concern would be the brightness compared to my screen image. Your thoughts?

Kit

[JRSmit]

After that, any half decent print that doesn't suffer metameric failure will look fine anywhere else viewed under decent (not a 4 watt nightlight bulb) illuminant: your eye adapts to it.

Van only agree, once ok, also check Arie Andries Kruithof,  i attached his graph showing the adaptability of our visual system.

[Kit-V]

Van only agree, once ok, also check Arie Andries Kruithof,  i attached his graph showing the adaptability of our visual system.

Could you please clarify? Thank you.

[deanwork]

What is wrong with 5000 k fluorescents in a shop light?

Could you please clarify? Thank you.

[digitaldog]

What is wrong with 5000 k fluorescents in a shop light?

Again: http://digitaldog.net/files/15TheRightLightpart1.pdf

[PeterAit]

In my quest to dial in the calibration of my display, I can appreciate the role that ambient lighting plays when evaluating the final print to the on-screen soft proofed image. There seems to be considerable discussion regarding the "optimal" color temperature of the ambient lighting for print evaluation. Which gives rise to my question: Should the print be evaluated under a full-spectrum daylight bulb (i.e. 5,000 - 6,000K) or under the lighting that the prints will eventually be displayed & viewed? Of course, the second part of my question might be difficult to determine.

Thank you for help.

Kit

It's essential to think about who will look at your prints, and where. If it's just for display in your home or studio then you have already answered your own question. If a gallery or web store then -- how can you know? You cannot. So don't fuss about it. Time is much better spent going out and taking photos rather than fussing over the fine points of color management.

[JRSmit]

Could you please clarify? Thank you.

Once your print under controlled high cri lighting looks like the image on your display , in my experience the print will also look good under different lighting levels provided the color temperature is as shown in the graph of Kruithof. Of course with reduced lighting levels our visual capability drop, but color wise it will stay ok.
Caveat is the spectral properties of lights, Kruithof only had incandescent lighting at that point in time (Worldwar 2)
Led lighting had a extremely poor spectrum f.I.

[digitaldog]

Once your print under controlled high cri lighting looks like the image on your display , in my experience the print will also look good under different lighting levels provided the color temperature is as shown in the graph of Kruithof. Of course with reduced lighting levels our visual capability drop, but color wise it will stay ok.

Yes but let's avoid if possible, CRI; it really needs to be ignored! it's a kludge to make a light source appear to be closer to daylight for marketing and light manufacturers. CRI was developed in large part to aid in the sales of Fluorescent tubes. There are BCRA tiles used to compare under a reference light source but only eight-ten. That's too small a set of tiles. That make it easy to create a spectrum that will render the 8-10 tiles and doesn't tell us that the light source is full spectrum. It doesn't tell us how the other colors will render. My understanding is there are two reference sources; Tungsten for warm bulbs and D50 for cool ones: Above and below 4000K. That means that a normal tungsten bulb and perfect daylight both have a CRI of 100! As such, a high CRI is a decent gauge of how well a light will preform in your home but not such a great indicator of how well it will work for photography and proofing. Both a Solux 48 and a "full spectrum" tube from home depot may have a CRI of 97. I can assure you the Home Depot bulb has a giant mercury spike and some spectral dead spots. 

A better metric is called CQS (15 very colorful patches). That doesn't tell us about the spectrum which is even a better way to evaluate the illuminant. 

[JRSmit]

I Agee Andrew, I just used high cri as a quick (and dirty) shortcut to point to a high quality lightsource, also given some previous posts.

[digitaldog]

I Agee Andrew, I just used high cri as a quick (and dirty) shortcut to point to a high quality lightsource, also given some previous posts.

It's dirty and it doesn't provide a short cut to a high quality light source, it does provide a short cut to a marketing driven value.

[headmj]

I have found that currently made high quality "daylight" LED bulbs have provided me with an adequate reviewing light.  Using this my prints have appeared as expected in other environments.  As expected very low illumination does not work.  Detail quickly disappears.

[Alistair]

There is so much to address in your question, just start here first:
http://digitaldog.net/files/15TheRightLightpart1.pdf
http://digitaldog.net/files/16TheRightLightpart2.pdf
Yes, full spectrum, meaning your options are tiny (Solux) other than that superb actual full spectrum daylight produce 93 million miles from any print.
You want to soft proof and see a match of display and print (as close as possible)? Well that print needs to be next to that display. After that, any half decent print that doesn't suffer metameric failure will look fine anywhere else viewed under decent (not a 4 watt nightlight bulb) illuminant: your eye adapts to it.

Apologies for butting in but this bit piqued my interest: "any half decent print that doesn't suffer metameric failure will look fine anywhere else viewed under decent".

What inkjet print is not subject to metameric failure to a greater or lesser extent?

[Ernst Dinkla]

Apologies for butting in but this bit piqued my interest: "any half decent print that doesn't suffer metameric failure will look fine anywhere else viewed under decent".

What inkjet print is not subject to metameric failure to a greater or lesser extent?

Good question!

I can only add some conditions that will keep better 'color constancy' in prints under decent illuminations.  Avoid OBA content in papers and inks in the first place. Even some B&W monochrome dye inks can show fluorescence. If a print does not shift much going from 2700-4000K continuous spectrum light indoors to the shade/overcast outdoor light then there is some color constancy in it, my eyes the observer. I like my HP Vivera pigment inks on OBA free papers that way. And less the ones on OBA papers. Without color samples with excellent color constancy next to the print this test is not representing a metameric matching/failure test.

Sunlight can be called 'decent" but it has UV content that can trigger OBAs and most of us here do not make outdoor prints but aim at indoor use where the Kelvin number can go from 2700 to 5500 and the intensity from 45cd/m² to maybe 1000cd/m2. For the luminance shift the Kruithof curve has sense. But our eyes improve in color judging up to 2000cd/m², which does not represent the average print display illumination or is practical for a viewing light. I use a 4000K halogen lamp and an 3800K Pharox 400 LED lamp as a viewing light. Philips Fluorescent tubes TL-D 90 950 at the ceiling in the print space.

I also have a halogen lamp here for the condition 45cd/m² + 2700K + a sample of UV blocking glass. Odd museum condition where the digital images of the originals were made with that glass in front and I only could check the repro print to the original with that glass in front and at that luminance level.

My left eye delivers a warmer image than the right, cataract operated, one.

Olino.org is a very good sight for lamp data.  It works independently and has several lamp types tested in an extensive way.  Explains all the test criteria and has a service for testing lamps that should not be too expensive for manufacturers.

One of the older threads on the subject: https://forum.luminous-landscape.com/index.php?topic=116755.0

White LEDs must have improved since.


Met vriendelijke groet, Ernst Dinkla

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

"For every complex problem there is an answer that is clear, simple, and wrong." H. L. Mencken

[digitaldog]

Apologies for butting in but this bit piqued my interest: "any half decent print that doesn't suffer metameric failure will look fine anywhere else viewed under decent".

What inkjet print is not subject to metameric failure to a greater or lesser extent?

Many.

[Doug Gray]

Apologies for butting in but this bit piqued my interest: "any half decent print that doesn't suffer metameric failure will look fine anywhere else viewed under decent".

What inkjet print is not subject to metameric failure to a greater or lesser extent?

A problem with metameric failure is that it occurs naturally and there isn't a thing you can do about it. For instance you can have a great setup with D50 profiles, a D50 monitor and D50 display side by side with great matching of print to display. Now try this:

Print a matching ColorChecker image and compare to a real ColorChecker under the D50 display. With a good workflow they match.  Now look at them under incandescent light at 2700K. They don't quite match because the spectral reflectance of many of the ColorChecker patches differ from that of the printed one.

And it isn't just printed CC v actual CC.  You can have two objects that have the same color under one light but appear different under another. That's the broader metameric failure.  But which one is "right?"  And if you were to photograph and print them what should the print try to match?

It's complicated and there isn't a perfect solution because metameric failure is intrinsic in real life objects. Not just printing.

But by far the biggest problem with prints is OBAs. Just avoid OBA paper. The next biggest problem is bad display illuminants like most fluorescents.

[bjanes]

Both a Solux 48 and a "full spectrum" tube from home depot may have a CRI of 97. I can assure you the Home Depot bulb has a giant mercury spike and some spectral dead spots.

A better metric is called CQS (15 very colorful patches). That doesn't tell us about the spectrum which is even a better way to evaluate the illuminant.

For home use I have been using a floor lamp fitted with a 47K 36 degree Solux bulb and have had decent matches to my calibrated monitor. One can vary the illumination to some extent by changing the distance to the print. The illuminance at 24 inches is 230 lux, not very bright, and the luminance from an 18% gray card at this distance is 20 cd/m^2. At 12" lux is 700 the luminance of the 18% card is 60 cd/m^2.

Looking at the B&H web site, I see they have a Fiilex V70 LED lamp with adjustable color temp and a 100% to 25% dimmer. CRI is stated as 90. Output in lumens is not stated, but power consumption is ≤ 13 watts. For comparison, a 10.5 65 watt replacement BR30 LED that I have puts out a claimed 650 lumens and the lux at 24" is 560. Color temp is 2700K. Luminance of an 18% gray card at 24 inches is 56 cd/m2.

https://www.bhphotovideo.com/c/product/1119531-REG/fiilex_flxv071_v70_led_color_viewing.html

My monitor Luminance is 110 cd/m^2 as measured from a Photoshop screen filled with white (calibration was 120 cd/m^2). Filled with 18%, the luminance is 26 cd/m^2.

As I understand color management with soft proofing, one should try to match the luminance of the print to that of the monitor. It is preferable that the color temperatures should match but this is often not the case. Most viewing booths use D50 but a monitor calibrated to D50 is unattractive and most people calibrate to D65 and rely on color adaption of the human visual system to take care of temperature mismatch. This usually works if the monitor and print are not in the same view.

B&H user reviews of the Fiilex are favorable and they list it as a best seller. It is not inexpensive at US $195 but this is cheap compared with a viewing booth. Has anyone used this device or at least have any thoughts on it?

Regards,

Bill

[Kit-V]

Thank you, everyone, for your responses to my original query. This is all very illuminating (yes, pun intended). But I must admit that as a photographer, my interest in color management is secondary. Don't get me wrong....color management is a critical component in expressing one's creativity. But since the quest for a "perfect match" between screen & print can be akin to hitting a moving target, I will likely limit my efforts to display calibration/profiling & basic viewing under D50 lights. Then after a bit of tweaking of the display's luminance & white point, it's time to grab the camera & hit the road. Nevertheless, I admire those of you who relish learning the intricate science of color management. I am the beneficiary of your knowledge. Thank you.

[JRSmit]

I have a lot of photographers as customer. The main 'problem' across all these photogs is that their monitor is set too bright. And close second is the side effect of Apple users, over emphasis of saturation of the Retina displays. More and more are purchasing a Eizo monitor in addition for proper image development.
Other than that the usual oops errors like chromatic abberations, sensor dirt causing greyish spots in the image, under exposure ( thus noise and loss of shadow detail) , etc.
So in other wordt focus on your image creation, and find a printer to help you refining your technical image quality and color management aspects.