Pages: [1]   Go Down

Author Topic: 6500kelvin  (Read 5879 times)

kingscurate

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 61
6500kelvin
« on: September 01, 2011, 01:08:26 pm »

Monitors are usually set to a kelvin temp of 6500(daylight???). Lightroom uses 5500K as daylight temp. Could someone shed any light if there is a difference or if it matters


Cheers
Logged
I aint a pro

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: 6500kelvin
« Reply #1 on: September 01, 2011, 03:03:53 pm »

Monitors should be set to whatever kelvin temp matches a print next to that display, properly illuminated (see:http://www.luminous-landscape.com/tutorials/why_are_my_prints_too_dark.shtml)

The numbers, be it for a display or what LR may say is daylight are immaterial, its the color appearance you desire that counts. YMMV.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Waeshael

  • Newbie
  • *
  • Offline Offline
  • Posts: 20
  • Medical camera designer since 1976
    • God, dogs and photography
Re: 6500kelvin
« Reply #2 on: September 11, 2011, 08:14:52 pm »

If you are a web browser and you set your monitor to 6500 K then you will see images pretty much the way they come from the photographer. If you set the monitor to 5500K, then all the images will have a warm brownish tinge to them. But if you are a printing house or graphics art person, the standard lighting around the world is for lights of 5500K. So if you look at your monitor at 5500 K and you look at your print at 5500K the colors will be a closer match, and adjustments in the editor will be easier to make. And people who print for a living use a different computer for browsing the web.

When I print - which isn't often these days - I look at the prints under a 5500K lightbox, and set the monitor to 5500K. It only takes a short while to get used to the color as the eyes adjust quickly. But some colorimeters restrict the color temperature that you can calibrate to, so if you intend to both browse and print you need a colorimeter that will allow you to choose both 5500 and 6500 K settings so that you can switch back and forth from the monitor preferences panel.

The temp 5500K was chosen as being closest to the color of noon daylight - I don't know where in the world they measured this but I suspect the French had something to do with it. In South Carolina in the summer it is definitely a cooler light.

Of course if your viewers look at your prints under these new fangled low energy bulbs, it doesn't matter what color temp you use - it's going to look like ....

cheers

Waeshael
Logged
Digital technology is fun.

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: 6500kelvin
« Reply #3 on: September 11, 2011, 08:42:46 pm »

But if you are a printing house or graphics art person, the standard lighting around the world is for lights of 5500K. So if you look at your monitor at 5500 K and you look at your print at 5500K the colors will be a closer match, and adjustments in the editor will be easier to make.

Ah actually not so. For years upon years, users who calibrate to a CCT 5500K display WP find a severe mismatch between that and their prints viewed under CCT 5500K. The CCT (correlated color temperature) is the main reason for this but other factors exist to explain the mismatch.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Waeshael

  • Newbie
  • *
  • Offline Offline
  • Posts: 20
  • Medical camera designer since 1976
    • God, dogs and photography
Re: 6500kelvin
« Reply #4 on: September 11, 2011, 09:02:16 pm »

In my own printing, I have been lucky enough to get very good color matches, so perhaps my lighting is more suitable for my particular situation.

Of course, reflected light and transmitted light have inherent differences. My experience is in the medical imaging industry. I was for a number of years responsible for the quality of print media and medical images for a medical imaging Company.

thanks for the comment. - You didn't say what these people do, today, who find a sever color mismatch in their own business. I wonder if you would follow up on this?
Logged
Digital technology is fun.

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: 6500kelvin
« Reply #5 on: September 11, 2011, 09:08:16 pm »

My experience was in print and prepress, few found 5500K WP for display calibration provided anything than a dingy yellow appearance and mismatch to the so called 5500K Fluorescent light booths used. It was usually 6500K +/- tuned to produce a visual match to some viewing booth.

Quote
- You didn't say what these people do, today, who find a sever color mismatch in their own business. I wonder if you would follow up on this?

http://www.luminous-landscape.com/tutorials/why_are_my_prints_too_dark.shtml
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Waeshael

  • Newbie
  • *
  • Offline Offline
  • Posts: 20
  • Medical camera designer since 1976
    • God, dogs and photography
Re: 6500kelvin
« Reply #6 on: September 11, 2011, 10:30:20 pm »

Ah, how things have changed. I stand corrected.

Thanks for the update.

Waeshael.
Logged
Digital technology is fun.

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: 6500kelvin
« Reply #7 on: September 12, 2011, 10:26:14 am »

Monitors should be set to whatever kelvin temp matches a print next to that display, properly illuminated (see:http://www.luminous-landscape.com/tutorials/why_are_my_prints_too_dark.shtml)

The numbers, be it for a display or what LR may say is daylight are immaterial, its the color appearance you desire that counts. YMMV.

YMMV. That is quite true. In the color management book by Fraser, Murphy, and Bunting, they recommend leaving the monitor at native WP or 6500K and adjusting the brightness of the monitor so that it matches that of the print. The eye can adapt quickly to changes of white point, but adaption to brightness is slower. They do state that matching the white point of the monitor to that of a sheet of paper in the viewing booth is an option in extremely critical environments, but they prefer to deal with paper white in the printer profile and concentrate on matching the brightness.

That book is rather dated, having been published in 2005. Have things changed with the newer LCD monitors with high bit look up tables? While many high end professionals have viewing booths, many (myself included) use a Solux lamp at a distance of about 18 inches from the print for proofing. The Fraser approach may be best for us.

Regards,

Bill
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: 6500kelvin
« Reply #8 on: September 12, 2011, 10:31:03 am »

On a less than high bit display, start with Native. Less banding is useful but a mismatch is less useful so its possible you may have to live with a bit of banding to get a better screen to print match by adjusting WP (and for those products that ONLY allow you to control CCT K, don’t provide a way to alter the other axis to affect the dreaded magenta/green cast, your SOL).

Got to nail the backlight (luminance) first, no question.

Higher bit depth makes the banding issue moot and better products provide full control over tweaking the WP numerically or visually (or both).
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: 6500kelvin
« Reply #9 on: September 12, 2011, 03:40:23 pm »

On a less than high bit display, start with Native. Less banding is useful but a mismatch is less useful so its possible you may have to live with a bit of banding to get a better screen to print match by adjusting WP (and for those products that ONLY allow you to control CCT K, don’t provide a way to alter the other axis to affect the dreaded magenta/green cast, your SOL).

Got to nail the backlight (luminance) first, no question.

Higher bit depth makes the banding issue moot and better products provide full control over tweaking the WP numerically or visually (or both).

Thanks for the reply and update. I was not aware that the SOL (Solux 4700K) lamp had a magenta/green cast. Can you elaborate or give some reference?

Regards,

Bill
Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: 6500kelvin
« Reply #10 on: September 12, 2011, 03:43:48 pm »

I was not aware that the SOL (Solux 4700K) lamp had a magenta/green cast. Can you elaborate or give some reference?

I don’t know they do (news to me).

The point I was making is some users, depending on the display and a slew of other possible conditions, can see a cast going magenta or green. Software that only provides a WP target will not allow that to be resolved. Better products provide a number of options for setting WP, including xy values which can address this issue.
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

MarkM

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 428
    • Alaska Photographer Mark Meyer
Re: 6500kelvin
« Reply #11 on: September 12, 2011, 05:51:29 pm »

Thanks for the reply and update. I was not aware that the SOL (Solux 4700K) lamp had a magenta/green cast. Can you elaborate or give some reference?

I don't think it makes much sense to talk about the cast of a light source without a reference to another white point. Our eyes will discount the illuminant and white will be white within pretty broad limits.

Solux does publish SPD data for their lamps. For instance here: http://www.solux.net/ies_files/SoLux%20Spectral%20Data.xls.pdf

It's not difficult to go from the SPD to xy chromaticity. In the case of the 4700K lamp you end up with x:0.3566619 y:0.3788944 which corresponds to a cct of 4708K (if my calculations are correct.) If we want a reference white point, the Plankian radiator at 4700k is as good as any, although still arbitrary. It should be around x:.35 y:.36 for 4700k. So, if the Solux performs to spec, I'd say it's pretty darned good.
Logged

Tim Lookingbill

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2436
Re: 6500kelvin
« Reply #12 on: September 13, 2011, 12:22:00 am »

The two LL articles below will show why the differences between a 6500K display can still match to a 5000K or neutral-ish print viewing light source.

http://www.luminous-landscape.com/tutorials/image_matching.shtml

http://www.luminous-landscape.com/reviews/accessories/pdv-3d.shtml

Basically the two tints aren't that pronounced enough to contaminate colors especially if the luminance of the display to the white of the paper matches.

Since the camera can't induce adaptation, the image samples in the links above show the two as looking different but in reality our eyes will correct for this. Neutral-ish is good enough because there isn't any device display or light source that can mimic daylight exactly especially with regard to the appearance of color cast. You'll notice in both images that the two displays slightly have a green bias which the camera is picking up on but our eyes don't see.
« Last Edit: September 13, 2011, 12:26:28 am by tlooknbill »
Logged

Ernst Dinkla

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4005
Re: 6500kelvin
« Reply #13 on: September 13, 2011, 04:53:50 am »

Both articles are quite old meanwhile. The consensus right now is that Solux halogen lamps with their continuous spectrum offer better viewing light conditions. Another change is to use lower Kelvin grade viewing lights more related to the display conditions of the prints later on, not the average of conditions the graphic industry aims at. Shifts from the original 5500K to 4700-4100K for viewing lights where display conditions may go as low as 3500-3000K. The full spectrum Solux lamps + inkjet inks with a better color constancy to different light sources "less metamerism" improved the relation between the proof print and the print exhibited (not for cheap galleries with fluorescent lighting though). Whether the relation viewing light - monitor improved has to be seen. The first LCD panels did not improve on what was available in CRTs. Gradually there has been an improvement in backlights, display control and calibration options, The task became harder too with both monitor and inkjet ink color gamut increases. I have some doubts that 6500K monitors and 4100K viewing lights are a good combination and try to keep a lower Kelvin grade for the monitor + a corresponding lower illumination. That is another thing to consider, if there is a difference in color temperature between monitor and viewing light should the light level not be adapted on that too? Following the Kruithof curve more or less? Right now the monitor light level is related to the ambient light level and the viewing light level is based on eyeballing to create the match to the monitor. And not everyone uses shields on both to reduce the (changing) ambient light influence. Some new articles on the subject that reflect those changes in time may be needed.


met vriendelijke groeten, Ernst

New: Spectral plots of +250 inkjet papers:

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


Logged

digitaldog

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 20630
  • Andrew Rodney
    • http://www.digitaldog.net/
Re: 6500kelvin
« Reply #14 on: September 13, 2011, 10:15:41 am »

Another change is to use lower Kelvin grade viewing lights more related to the display conditions of the prints later on, not the average of conditions the graphic industry aims at. Shifts from the original 5500K to 4700-4100K for viewing lights where display conditions may go as low as 3500-3000K.

Yes, it further illustrates the folly of looking solely at CCT Kelvin values and making any kind of definite decision based on the numbers without first using your eyes!
Logged
http://www.digitaldog.net/
Author "Color Management for Photographers".

Tim Lookingbill

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2436
Re: 6500kelvin
« Reply #15 on: September 13, 2011, 11:34:20 am »

Didn't link to those articles to read, only to show the differences in color cast between a 6500K calibrated display and a supposedly "5000K" viewing booth/light in the image samples.

There are a number of color cast tints associated with 6500K that don't contaminate 16 million or so colors available on a display and viewed in a photograph. It's about coming up with a neutral enough cast that maintains equal balance in saturation levels among all those colors without introducing hue errors which color management maintains quite well as long as the display's color characteristics are measured and a profile produced.

The Solux introduces some contamination in certain greens and blues but its spectral reflectance and spotlight characteristics closely mimics the characteristics of a sunbeam which affect the appearance of hue/saturation which happens to mimic the hue/saturation relationships in a calibrated/profiled display and viewed in a CM app.
Logged

madmanchan

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2115
    • Web
Re: 6500kelvin
« Reply #16 on: September 14, 2011, 04:48:59 pm »

Yes, be careful with only going by CCT values, because you can have very different lamps all mapping to the same CCT value but very different appearance.  I do really like the Solux bulbs.
Logged
Eric Chan
Pages: [1]   Go Up