Monitor Colo(u)r Temperature Recommendations

[billtils]

I have been using Colormunki's recommended D65 for monitor calibration and recalibration. While waiting for delivery of the Pro100s I browsed through some of the the Canon manuals, and their "Digital Photo Printing" publication says to use D50 (with what appears to be a recommendation to use D50 fluro tubes for ambient lighting).

Any advice/comments?

[DanK]

Bill,

I am certainly not an expert on this, but a few thoughts:

For what seems to me to be a clear explanation, see this article:

This is her conclusion:

So which color temperature setting is best? As with the gamma setting, it depends on what kind of work you do. For many years, the standard color temperature setting for graphic arts work was 5000°K (also known as D50). This is closest to neutral white and simulates common lighting conditions for reading printed materials. Therefore, I feel this is the ideal color temperature to select if you do mostly work for print.

If you create mostly web graphics or other images viewed on screen, choose 6500°K (also known as D65). This is the default color temperature of the sRGB color space and is used by mass market computer monitors, most of which are uncalibrated. It also displays images with a bluish color cast that is familiar to consumers who watch lots of TV (e.g., most Americans).

That said, I assumed that the defaults of both of my i1 devices would be for print (I recently had to upgrade to the Display Pro to be compatible with Windows 10), and I have never changed the default color temperature. One reason I assumed that is that calibration turned every monitor I calibrated less blue, sometimes a lot less blue. But my assumption may have been wrong.

I calibrate my monitors under the lighting I will use for editing, which is a set of halogens that I think are 3000K, but I don't recall for certain. I have done this for years, and it has worked fine: my prints seem to have natural colors. However, you have made me doubt what I did, and next month, when I have more time, I will read the i1 documentation and perhaps recalibrate my monitor.

Manfred is far more careful than I am about this, so he can probably provide a much more informative answer.

Dan

[xpatUSA]

Manfred is far more careful than I am about this, so he can probably provide a much more informative answer.

Dan

I, on the other hand, am quite careless. ;-)

Just for information, my NEC P242W is as it came out of the box - sRGB with no corrections. I don't print, so I'm less concerned about absolute color accuracy. The ambient lighting is a shaded LED strip at left and rear about 4 ft away, warm white, CCT unknown but it only casts about 8 lux on the screen. The screen measures 53 cd/m^2 ... quite enough for my tired old eyes.

Please don't shoot me; if I ever got into printing, I would get a calibrator and dutifully buy a D50 lamp.

[Manfred M]

For photography, the D65 standard is generally what is used, especially when it comes to creating photographic prints. My understanding is that while this illuminant predates modern photography, the sRGB white point is based on it, so this is part of the reason for its widespread use.

Your explanation of the D50 is close Dan and it is still very much in use in the graphic arts business. My understanding it that this is also the standard used in offset printing (i.e. the printing press); so I'm not sure why this came about. In the CMYK process used by this industry the white point is defined by the medium; paper. There is no white ink. White paper, even when chemically bleached and uses optical bleaching agents (OBA) to make it whiter, will naturally be warmer than what we can generate on our computer screens. A warmer white point for press work seems to make sense logically.


Dan you mention your lights for evaluating prints. The approach has shifted over time as well. At one point (predating the PC) the recommendation was to use a standardized daylight setting; around 5500K for photo evaluation and a carefully calibrated lightsource was used to view images. The MacBeth Company (now owned by x-Rite) was the largest supplier of these viewing booths. In those days there were colour analyzers used in the wet darkroom and proofs would be sent by the commercial photographer to the client who would examine them under the same type of light source as the photographer. This was the "common ground" that work was judged under.

Today we no longer do this and the recommendation is to evaluate the print under the lighting conditions it will be viewed under. If this is not possible, then the recommendation is to view the print under full spectrum lights, the type that is used in galleries and museums, to evaluate the print. The human visual system adapts to the colour temperature of the light source quite quickly, so the absolute colour temperature is less important than ensuring that all of the wavelengths are emitted by the evaluation light source. A couple of the high end photographic printers that I know have recommended Solux lights to me as the best available print viewing lights.

http://www.solux.net/cgi-bin/tlistor...ges/index.html

[xpatUSA]

For photography, the D65 standard is generally what is used, especially when it comes to creating photographic prints. My understanding is that while this illuminant predates modern photography, the sRGB white point is based on it, so this is part of the reason for its widespread use.

Your explanation of the D50 is close Dan and it is still very much in use in the graphic arts business.

I think Dan was talking about monitor viewing, for which D65 is the white point as you say (Rec. 709, Rec. 2020 and Adobe), but ICC says this about lighting for the monitor:

An ICC v2 sRGB profile is provided below, with XYZ black point scaling to zero. This profile has the preferred rendering intent in the header set to 'perceptual'. The profile contains the standard linearized Bradford D65 to D50 chromatic adaptation tag (this tag was often not present in older sRGB profiles), and the media white point tag is set to D50 (as is required for ICC v4 display profiles and recommended for v2 profiles). This avoids the inappropriate color casts that older sRGB v2 ICC profiles sometimes produced when the absolute colorimetric intent was used.

(my bold), which appears to confirm what Dan said and is about as close as it gets.

[NorthernFocus]

Like many things I think it is useful to understand where "standards" come from in order to make intelligent decisions about their use. The article that Dan K referenced is great in that regard. Of course as with most things, a little knowledge can be dangerous. Or at least confusing. Clearly if one is serious about producing "correct" colors, a different calibration profile should be used during editing based on the intended output medium/audience. To complicate things further one is confronted with the closely related issue of choosing a WB setting when editing an image.

Then again for the non graphic artist there is always the question of whether the target audience can tell the difference

[Manfred M]

Sorry for the confusing reply Ted. Both Dan and I were writing about computer screen settings. I generally look at an end-to-end process and colour managed workflow with the appropriate settings at each step of the way, so my work flow is from capture to print. This is the same workflow I follow, even if the image ends up not being printed.

D65 is the setting that I use in my work flow and is based on the advice of knowledgeable, practising commercial photographers. It is also the recommended default that x-Rite recommends with its calibration devices.

[xpatUSA]

Like many things I think it is useful to understand where "standards" come from in order to make intelligent decisions about their use. The article that Dan K referenced is great in that regard. Of course as with most things, a little knowledge can be dangerous. Or at least confusing. Clearly if one is serious about producing "correct" colors, a different calibration profile should be used during editing based on the intended output medium/audience.

Confusing indeed, Dan! Fortunately for us these days, most stuff is color-managed. Which means that I can happily save a file as sRGB and expect that someone with an Adobe-gamut monitor will see correct colors. That also means that I could save as Adobe or even ProPhoto (!) and still expect everybody else's stuff to show correct colors.

That makes me wonder why a different calibration profile is necessary for the intended output medium/audience. Perhaps I misunderstood.

To complicate things further one is confronted with the closely-related issue of choosing a WB setting when editing an image.

Then again for the non graphic artist there is always the question of whether the target audience can tell the difference

Indeed ...

[DanK]

One other thought. When using an sRGB monitor, I haven't worried about getting things to look right on uncalibrated monitors. I worry about prints and the appearance on calibrated monitors. However, I plan to get a wide-gamut monitor, and that adds another wrinkle. Clearly, an image edited in a wider gamut could look lousy on an sRGB monitor, calibrated or not. So, my plan is to edit for calibrated sRBG monitors when just posting online but editing a second copy for printing. I'm curious whether others who have wide-gamut monitors do this.

[xpatUSA]

Sorry for the confusing reply Ted. Both Dan and I were writing about computer screen settings ...

Thanks for the clarification, Manfred. I thought that Dan K mentioned the lighting around his monitor (which ICC expects to be D50?). Makes me wonder what ICC means by "media white point" ! Off to google about that now ...

... [edit] that was a waste of time - ICC literature is almost as bad as ISO for intelligibility.

[billtils]

Enjoyed the discussion on colour temps, but it doesn't quite explain why Canon (who make the printer ...) and Colormunki (who make the monitor calibrator) differ, unless the answer is that in this respect Canon can be regarded as working within the print industry.

[george013]

Enjoyed the discussion on colour temps, but it doesn't quite explain why Canon (who make the printer ...) and Colormunki (who make the monitor calibrator) differ, unless the answer is that in this respect Canon can be regarded as working within the print industry.

It's just not just hard science I think. A lot of feeling and a choice.

George

[xpatUSA]

Enjoyed the discussion on colour temps, but it doesn't quite explain why Canon (who make the printer ...) and Colormunki (who make the monitor calibrator) differ, unless the answer is that in this respect Canon can be regarded as working within the print industry.

This article helped me. It concludes that D65 is correct for a monitor setting, as opposed to D50 which is not correct.

http://www.color-image.com/2012/02/m...soft-proofing/

Basis being that a white patch on a D50 monitor will look yellow and get no adaptation by the eye, whereas a white patch in a print under D50 light will look white because of adaptation by the eye. The difference being that the print is illuminated but the monitor is luminous.

Makes sense?

[DanK]

Enjoyed the discussion on colour temps, but it doesn't quite explain why Canon (who make the printer ...) and Colormunki (who make the monitor calibrator) differ, unless the answer is that in this respect Canon can be regarded as working within the print industry.

Bill, I would simply ignore this. I'm guessing that Canon's saying this has something to do with its own print software, which I never use for color anyway. I have always let i1 do its thing, and the results are fine. That is, they are fine as long as I remember to handle color right: having software handle it with the correct ICC profile and setting the printer's firmware not to do any color matching.

So, my unsolicited advice is to forget about what you found in the Canon documentation unless you don't find the prints to be as they should be.

I think you will be very pleased with this printer. It can produce truly stunning prints.

[billtils]

Makes sense?

Indeed it does Ted.

[billtils]

... unless you don't find the prints to be as they should be. ... I think you will be very pleased with this printer. It can produce truly stunning prints.

Dan

"Ignore" sounds good! The prints are exactly as I'd like them, I am very pleased with it, and the family conclusion (not just me) is that they are indeed stunning.

[Manfred M]

Dan - have been running on a wide gamut screen for about 10 years now (I'm onto my second wide gamut screen, which is a bit over 1 year old), so it is 100% sRGB compliant and 99% AdobeRGB compliant. I also run a dual screen setup; one sRGB and one AdobeRGB. The screen I work on is the wide gamut one and the one I park all my menus, etc is an sRGB one. Both have been calibrated and profiled, so if I want, I can get a side-by-side view of what the sRGB and AdobeRGB images look.

I rarely have run into an issue where an image is not widely out of gamut in on the wide gamut screen, but is wildly out in sRGB, but it does happen. At that point, I have to make some choices on how to best address the issue.

I normally (but not always) work in the 16-bit ProPhoto colour space including when I print. I know other photographers that stick to AdobeRGB. I can give valid reasons for both approaches, but as with many things photographic (and in life), the "correct answer" is not terribly clear cut and it really depends a lot on the original image, the printer being used and the paper that is being used to print with. There can be "surprises" in the output where the screen rendering intent handles large areas that are out of gamut that are in gamut for the printer / ink / paper combination, but that happens so rarely that I would rather use the additional data that ProPhoto can capture and worry about the consequences at print time.

When it comes to printing, I use a copy of my final edited image and do additional adjustments to it. I generally have two copies of the image open; one as the original completed image in whatever colour space I am working in and the second one viewed using softproofing mode. I try to adjust the image (local vibrance and saturation are typically areas I work on) to compensate for what the print process will do to the image. I will also resize to print size and do my output sharpening at that time. If the image appears to be problematic or I am printing a large format, I will do a small test print before any final print and will tweak my print file accordingly, if there is something I don't like in the test print. I generally do not save the test print file as I can usually reproduce it in 5 or 10 minutes and usually only print a file once.

If I can't get the print image looking the way I want it, I will generally switch to a different paper. While I like the thick cotton rag fine art papers, they don't always get me the colours I want. As a general rule, higher gloss papers (lustre, baryta, etc.) tend to produce a wider gamut than the matte ones.

[DanK]

Manfred,

That's helpful. I recently got a new computer, and I haven't yet checked to see whether I can use separate calibrations for my two monitors. My old computer couldn't do that. I just calibrate the main one. However, I think some wide gamut monitors can be switched on the fly to sRGB, which would solve my problem.

I always work in ProPhoto. I see no disadvantages to doing that. However, almost all of my software renders the image on hte monitor in sRGB. (I assume it would use Adobe RGB if the monitor handled that.)

My default paper has been a luster paper, but I have also used matte and others, and the difference in gamut is often quite striking. I print 7 x 10 folded cards on a cold white matte card stock. Most of my cards have been flower photos, which means that a fair number have gone out of gamut when I printed on that medium. Usually I can get it more or less OK, but I have had a few that I couldn't sufficiently repair.

All this may change for me. I'll be semi-retired at the end of the year, and at that time, I intend to experiment with a wider range of papers.

Dan

[Manfred M]

With my Benq SW2700, I can jump between AdobeRGB, sRGB and B&W at the press of a button. It's a feature I rarely use; but when I have a problematic image it does help. The only downside is that the brightness varies in each of the modes, so I still do all my edits with the AdobeRGB settings.

[rudi]

.... It concludes that D65 is correct for a monitor setting, as opposed to D50 which is not correct......

One can find articles that states the opposite. It is not all that black or white….IMO. As often 'it depends' and comes down to personal preferences, and what works for you.
For normal use, my monitor was at 6000°K, but when printing on warmer papers I did set my monitor to 5000°K. When printing on whiter papers my monitor was set to 6500°K or 7000°K. And I found that very usefull at that time.
Nowadays I have settled on 5500° for most off my work, and is rarely changed (this may change overtime, who knows..) YMMV.