Eyeone's 6500K is actually D65

[Serge Cashman]

Oh well... It looks like Eye-One Match's 6500K target is actually a D65 (0.3127, 0.3290).  And 5500K is D55 (0.3324, 0.3474). So I guess the Huey targets are all D numbers as well. Go figure.  Colorvision, Basiccolor, Coloreyes etc use other naming conventions for K and D (K being the blackbody coordinates while D being on the daylight locus)...

[digitaldog]

Oh well... It looks like Eye-One Match's 6500K target is actually a D65 (0.3127, 0.3290).  And 5500K is D55 (0.3324, 0.3474). So I guess the Huey targets are all D numbers as well. Go figure.  Colorvision, Basiccolor, Coloreyes etc use other naming conventions for K and D (K being the blackbody coordinates while D being on the daylight locus)...
[a href=\"index.php?act=findpost&pid=75178\"][{POST_SNAPBACK}][/a]

This is why I dislike a setting using a correlated color temperature (6500K) because we really don't know the chromatisities values since it's kind of a "vague" value. Standard Illuminants are clearly defined (not that selecting either produces that on your display <g>). But at least you know what the software is shooting for.

For the huey market, the users wouldn't know a standard illuminant from a standard light fixture. They're trying hard to hide all that under the hood.

[Tim Lookingbill]

Still can't figure out why Photoshop combines labeling of white points in their CustomRGB dialog box as both 6500K/D65, 5000K/D50, etc.

[Serge Cashman]

This is why I dislike a setting using a correlated color temperature (6500K) because we really don't know the chromatisities values since it's kind of a "vague" value...

OK - finally I understand it. It's vague because software vendors use K values arbitrarily. Some use it for daylight coordinates and some use it for blackbody coordinates. I agree that D numbers (D65 and so on) make more sense as target descriptions because they eliminate the ambivalence..

Not that I care about Huey but if we can say what it's targets are - why not? They are the D targets AFAIK.

[32BT]

OK - finally I understand it. It's vague because software vendors use K values arbitrarily. Some use it for daylight coordinates and some use it for blackbody coordinates. I agree that D numbers (D65 and so on) make more sense as target descriptions because they eliminate the ambivalence..

Huh? I know you understand the concept, but apparently you're ready to dive in deeper.

How many D numbers do you believe exist?

What colors do they represent?

How are these colors derived from the D number?

[Serge Cashman]

Huh? I know you understand the concept, but apparently you're ready to dive in deeper.

How many D numbers do you believe exist?

What colors do they represent?

How are these colors derived from the D number?
[{POST_SNAPBACK}][/a]

Well, I sort of read the theory, it was discussed here before...  I think I understand how it works in principle. I understand there's an infinite number of chromaticities on a daylight locus, planckian lockus or any isothermal line.

I'm not sure what do you mean by "What colors do they represent". I assume the chromaticities used as targets during calibration are used because they can be percieved reasonably neutral (or "white" at highest luminance). If there are any practical differences between D chromaticities and blackbody chromaticities for white point targets of calibration software - I don't know. I don't think there are any, but I just relized that the K numbers are used ambiguously.

[a href=\"http://doug.kerr.home.att.net/pumpkin/Color_Temperature.pdf]http://doug.kerr.home.att.net/pumpkin/Color_Temperature.pdf[/url]

[Tim Lookingbill]

Well, since 6500K will always have an arbitrary appearance monitor to monitor and D65 has a defined appearance, if anyone can come up with an RGB combo that resembles D65 according to its spectral distribution graph, please share.

I would think you'ld start out with 255 in the blue channel to hold luminance and decend in a sort of 20 degree downward slope with red channel having the lowest number.

That's my guess as to what D65 looks like.

I couldn't gather any real world useful insight with that color temperature pdf as others I've come across on the web-way over my head. And they didn't even use colors to illustrate it. What's the point?

[digitaldog]

I'm not sure where this is all going. I can tell you that if you heated your display such that it glowed hot enough and like the theoretical blackbody, you could end up with 6500K but I suspect you'd screw up the unit's electronics big time ;-).

The only "thing" I know of that can produce D65 is unfortunately also a very hot object and a good 93 million miles away from where you're editing your images.

Now what's the question again???

[Serge Cashman]

For all practical purposes D65 is 0.313, 0.329 CIExy.

6500K in this context is normally  0.314, 0.324 CIExy .

This is quite obvious ,say, in Basiccolor. It is less obvious but documented in Colorvision products. E1 uses the D coordinates for their K targets (which obviously do have the same corresponding color temperature).

[Tim Lookingbill]

Andrew,

We're (I'm) talking about color cast simulation on a monitor as serge points out in his original post about the EyeOne. He expressed it in XY coordinates. I'm expressing it in how it looks.

I see pure white objects outdoors change in color cast from midday bluish, not cyanish white, and orangey pink white with the sun at 20 degrees off horizon.

Which one is D65?

Pretty simple question and why it never gets answered is beyond my comprehension.

[digitaldog]

We're (I'm) talking about color cast simulation on a monitor as serge points out in his original post about the EyeOne. He expressed it in XY coordinates. I'm expressing it in how it looks.

I see pure white objects outdoors change in color cast from midday bluish, not cyanish white, and orangey pink white with the sun at 20 degrees off horizon.

Which one is D65?

Pretty simple question and why it never gets answered is beyond my comprehension.
[a href=\"index.php?act=findpost&pid=76452\"][{POST_SNAPBACK}][/a]

First off, how do you control Chromatic Adaptation visually?

The Standard D illuminants were to my knowledge made from actual measurements by the CIE many years ago. It might take a bit of searching but I recall they made these actual daylight measurements in a certain location at a certain time of day.

[Tim Lookingbill]

Referencing the spectral distribution graph of D65 pretty much gave me a clue as to the look, but was just wanting a confirmation.

I see through quite a few P&S digicam shots of outdoor scenes posted on the web accurate looking midday white point readouts having this RGB number ratio that matches pretty close to the graph.

White points in these straight out of the camera unedited digicam shots rarely have R=G=B readouts.

[32BT]

The Standard D illuminants were to my knowledge made from actual measurements by the CIE many years ago. It might take a bit of searching but I recall they made these actual daylight measurements in a certain location at a certain time of day.

Apparently they took over 600 daylight spectra and plotted their x and y co-ordinates. This results in scattered dots. After taking the mean of the dots you end up with a curve called the daylight locus which runs slightly above and roughly parallel to the planckian locus.

Note however that this curve only describes the relation between the x and y co-ordinates!

[jani]

Apparently they took over 600 daylight spectra and plotted their x and y co-ordinates. This results in scattered dots. After taking the mean of the dots you end up with a curve called the daylight locus which runs slightly above and roughly parallel to the planckian locus.

I just can't help wondering what effect global dimming has on these measurements.

[bjanes]

I'm not sure where this is all going. I can tell you that if you heated your display such that it glowed hot enough and like the theoretical blackbody, you could end up with 6500K but I suspect you'd screw up the unit's electronics big time ;-).

The only "thing" I know of that can produce D65 is unfortunately also a very hot object and a good 93 million miles away from where you're editing your images.

Now what's the question again???
[{POST_SNAPBACK}][/a]

I'm not certain what the question is either, but I did discover some interesting information about daylight. D65 is defined by a spectral power distribution related to average overcast skylight and sunlight mixed with total skylight as explained in this link:

[a href=\"http://www.kayelaby.npl.co.uk/general_physics/2_5/2_5_4.html]http://www.kayelaby.npl.co.uk/general_physics/2_5/2_5_4.html[/url]

However, the spectrum of the sun as viewed from space is similar to a 5780 K blackbody as explained in this link:

http://casa.colorado.edu/~ajsh/colour/Tspectrum.html

D65 is more blue than direct sunlight because of the contribution of skylight due to Rayleigh scattering.

http://hyperphysics.phy-astr.gsu.edu/hbase/atmos/blusky.html

Since D65 is the average of multiple spectral readings under various conditions, sunlight at any given moment can not be used as a reference, but D65 can be approximated by xenon lamps with filters:

http://www.zeiss.de/4125680F0053A38D/Conte...1256C2C00452AAB

I don't know if this contributes to the discussion, but I found the links interesting.

Bill Janes