Canon EOS WB shift

The Canon EOS cameras have a "white balance shift" setting. This provides for setting a "color correction" of up to ±9 units along two axes, designated "amber-blue" and "green-magenta".

I thought it would be interesting to see just what these two axes are and how they relate to the axes of known color spaces.

I tested by having my EOS 20D regard a WhiBal gray card exposed to a consistent ambient illumination. I used the CWB function to get a basic WB correction for the illumination. Then I took test exposures with the WB shift set to all nine possible combinations of -5, 0, and +5 units on each of the two axes (treating the "amber" direction and "green" directions as positive, the convention used in reporting the WB shift setting in the Exif metadata).

I plotted the results on several different coordinate planes, along with vectors showing the "directions" of the chromaticity of the three color space primaries (R, G, and B) and the complementary hues C, Y, and M. The result for the "0,0" setting was used as the reference, arbitrarily plotted at the origin of the coordinate system.

The results on the CIE u-v plane were the most telling. There, as a matter of fact, the six "primary" vectors are almost equally-spaced angularly.

Here we see that plot. The axes are labeled "du" and "dv" to reflect that this is a "differential" plot; the actual origin of the u-v plane is far off this chart.



Note that since I read the resulting chromaticities of the color-corrected images on an 8-bit RGB basis, the plot is susceptible to errors from quantizing that can approach 0.001 u-v unit.

The light violet lines connect the extreme points of the constellation of data points, a primitive way to estimate the orientation of the two axes without benefit of formal linear regression analysis.

The two black double-ended arrows show the orientations of these two axes.

We see that the two axes are not quite orthogonal (at right angles) on this plane. The "green-magenta" axis is seen to be closely aligned with the green and magenta primary vectors. The "amber-blue" axis falls between two pairs of primary vectors; thus we see that the labels "amber" and "blue" are somewhat arbitrary. (The two directions are actually an "orange" hue and a "cyan-blue" hue.)

On the larger front, we note that evidently the chromaticity shift produced by this setting "piles onto" the shift established by the "other" white balance setting, regardless of the choice. That is, whatever color shift would be produced by the current WB setting (including a decision made by the camera's AWB algorithm), the "WB shift" adds to that.

The Canon manuals generally indicate that, with respect to settings along the amber-blue axis, one unit of shift corresponds to 5 mireds of color temperature correction.

The mired is a unit of inverse color temperature (now preferably called the "mirek", or even more preferably, the "micro reciprocal Kelvin")

It properly only quantifies chromaticity shift along the "blackbody locus", that is. among the chromaticities produced by a black body radiator at various temperatures. The "amber-blue" axis of the Canon WB shift lies generally in that direction. (Since the locus is a curve on the u-v plane, no axis can truly follow it along its length.)

Similarly, the green-magenta axis, being orthogonal to the amber-blue axis on the CIE u-v plane, is generally the direction along which we measure the departure of a chromaticity from the blackbody locus (the direction reflected by the "tint" parameter in Photoshop).

is there any value to the green/magenta bias/cct

i have recently had a conversation with the support people at canon about the issue of value associated with the wb bias of green/magenta on the wb shift function. the blue/amber has a 5 mired value per shift on the blue/amber axis. why is there no known value for the green/magenta shift on the opposed axis?

the support people can't get an answer from the developers at canon, so i pose the question to those who may have used this function. is there anyone out there who has any idea what the value is for the cct of the green/magenta shift?

Hi, Michael,

Michael Desilets said:

i have recently had a conversation with the support people at canon about the issue of value associated with the wb bias of green/magenta on the wb shift function. the blue/amber has a 5 mired value per shift on the blue/amber axis. why is there no known value for the green/magenta shift on the opposed axis?

Click to expand...

The mired (today often called the mirek) is a measure of a shift in coordinated color temperature. (Recall that definition is in terms of the unit of coordinated color temperature, the Kelvin.)

A chromaticity shift essentially parallel to the "Planckian locus" on the CIE chromaticity diagram represents a change in correlated color temperature, and thus the amount of such a shift can be properly described in mireds.

The amber-blue axis of the Canon white balance color correction bias adjustment is generally in a direction parallel to the Planckian locus, and thus changes along that axis are reasonably quantified in mireds.

A chromaticity shift perpendicular to the "Planckian locus" on the chromaticity diagram represents no change in correlated color temperature, and thus the amount of such a shift in mireds would always be zero.

The green-magenta axis of the Canon color correction adjustment is in a direction generally perpendicular to the Planckian locus, and thus changes along that axis would basically have a mired measure of zero. (They would have a small mired value, but it would not be meaningful.)

So the real answer to your question is that along the green-magenta axis there is no change in mireds (at least the mired value would not be meaningful). And I guess that's why the Canon guys do not know the answer!

Best regards,

Doug

Hi Doug,

This is a great essay. Although I already knew a lot about color spaces and WB, this essay has helped me understand certain issues much more clearly than I did before. The information presented is generally applicable regardless of the fact that it is about the WB shift of the Canon cameras. For example, it gives a good insight into why WB controls in raw converters have G/M as their tint axis.
At first, I have tried to decipher the graph without reading your text and I have come a long way. This means that the graph is drawn very well and speaks for itself. Your text is, of course, essential. Thanks so much for helping us out with this knowledge, much appreciated.

Cheers,

PS: I have just seen this essay although it was posted almost 20 months ago. Why were there no reactions at that time? Go figure!

Doug_Kerr said:

A chromaticity shift essentially parallel to the "Planckian locus" on the CIE chromaticity diagram represents a change in correlated color temperature, and thus the amount of such a shift can be properly described in mireds.

The Planckian locus is the path of all points representing the chromaticity of a "blackbody radiator" at different temperatures. Only points on that curve have actual color temperatures. For chromaticities near, but not on, the curve, we can state a "correlated color temperature", which is the temperature of the nearest point on the curve. When we hear of the "color temperature" of a light source, it is usually the correlated color temperature that is actually meant.
​

The amber-blue axis of the Canon white balance color correction bias adjustment is generally in a direction parallel to the Planckian locus, and thus changes along that axis are reasonably quantified in mireds.

A chromaticity shift perpendicular to the "Planckian locus" on the chromaticity diagram represents no change in correlated color temperature, and thus the amount of such a shift in mireds would always be zero.

The green-magenta axis of the Canon color correction adjustment is in a direction generally perpendicular to the Planckian locus, and thus changes along that axis would basically have a mired measure of zero. (They would have a small mired value, but it would not be meaningful.)

Click to expand...

Hi Doug,

My compliments, you've worded that very well.

For those who want to get into a lot more detail, I refer to the excellent site of Bruce Lindbloom. He also offers ANSI C program code to calculate the Coordinated Color Temperature from an XYZ color coordinate (using Robertson's algorithm), and formulas to convert to and from other coordinates, such as Luv or xyY (click on the "Math" button on his site to get an overview of the available conversions).

Cheers,
Bart

Rephrased...

Thanks guys for the quick attention to this matter. Perhaps, I was not clear in expressing my initial question to the group.

To rephrase:

If I am in a situation where I needed a plus green filter, could I adjust by using the green/magenta bias to the wb/shift program on my Canon camera?

And, by how many steps/shifts along the green/magenta bias (not the blue/amber bias), would equal a Cinegel Tough ½ Plusgreen filter value of CC15 Green?

I apologize for the previously phrased question.

Regards,

Michael Desilets said:

Thanks guys for the quick attention to this matter. Perhaps, I was not clear in expressing my initial question to the group.

To rephrase:

If I am in a situation where I needed a plus green filter, could I adjust by using the green/magenta bias to the wb/shift program on my Canon camera?

Click to expand...

There's a difference between shooting through a filter, and changing the whitebalance after the fact. Can't you just include a WhiBal or other gray object in a test shot and do the colorbalancing afterwards? It'll be much more accurate as well.

And, by how many steps/shifts along the green/magenta bias (not the blue/amber bias), would equal a Cinegel Tough ½ Plusgreen filter value of CC15 Green?

Click to expand...

I'd have to look that up and calculate, but it seems like an exercise that's not too useful. It will stiil be postprocessing when you don't alter the colorbalance of the light that will fall on your sensor.

Cheers,
Bart

Michael Desilets said:

Thanks guys for the quick attention to this matter. Perhaps, I was not clear in expressing my initial question to the group.

To rephrase:

If I am in a situation where I needed a plus green filter, could I adjust by using the green/magenta bias to the wb/shift program on my Canon camera?

And, by how many steps/shifts along the green/magenta bias (not the blue/amber bias), would equal a Cinegel Tough ½ Plusgreen filter value of CC15 Green?

I apologize for the previously phrased question.

Regards,

Click to expand...

If I understand you correctly Michael (of which I am not certain):
1) There is a shot which should have been taken using a green filter.
2) You took the shot without using the filter.
3) In the raw conversion or post processing, you want to simulate the green filter's effect by making the tint greener.

The other scenario being:
A) There is a shot which was taken using a green filter.
B) In the raw conversion or post processing, you want to nullify the the green filter's effect by making the tint more magenta.

So which scenario are we talking about? 1-2-3 or A-B? Or a third one?

If we are talking about scenario A-B, Bart's advice of using a WhiBal or a gray card is quite sensible.

If we are talking about scenario 1-2-3, there are simulated photo filters in photoshop which add any tint to your image after it has been converted normally via the raw converter. There are many ways to achieve this in the post.

Sorry if I misunderstood this totally .


Cheers,

Cem_Usakligil said:

If I understand you correctly Michael (of which I am not certain):
1) There is a shot which should have taken using a green filter.
2) You took the shot without using the filter.
3) In the raw conversion or post processing, you want to simulate the green filter's effect by making the tint greener.

The other scenario being:
A) There is a shot which was taken using a green filter.
B) In the raw conversion or post processing, you want to nullify the the green filter's effect by making the tint more magenta.

So which scenario are we talking about? 1-2-3 or A-B? Or a third one?

Click to expand...

Cem,

The other scenario is that one wishes to take a picture with a green filter, but achieve the same reusult through color balance adjustments. What settings correspond to what filters?

In practice would one take a WB shot, then adjust for for the wanted green filter effect and in processing use the WB shot taken before the effect was added?

Asher



Asher

Asher Kelman said:

...The other scenario is that one wishes to take a picture with a green filter, but achieve the same reusult through color balance adjustments. What settings correspond to what filters?

In practice would one take a WB shot, then adjust for for the wanted green filter effect and in processing use the WB shot taken before the effect was added?

Click to expand...

Hi Asher,

This is my scenario 1-2-3, please re-read what I wrote. In this scenario, I would set the WB as normal during raw conversion and then afterwards add the green filter effect in photoshop.

Cheers,

UPDATE: Now that I've finally understood what Michael wanted, I also think that your message was actually about the in camera WB correction. If so, then I stand corrected, that was my third scenario indeed

This example should clear up the use for the green balance shift theory that I have.

Under fluorescent lighting requiring a Cinegel Tough ½ Plusgreen filter value of CC15 Green to my strobe to balance the light, could I conceivably use the green/magenta bias (knowing the value to the shift) to bring the scene back to a neutral/white balance?

It is just an idea and was wondering if there was a way to work this in camera rather than PP.


I really appreciate all the attention to this matter.

Michael Desilets said:

This example should clear up the use for the green balance shift theory that I have.

Under fluorescent lighting requiring a Cinegel Tough ½ Plusgreen filter value of CC15 Green to my strobe to balance the light, could I conceivably use the green/magenta bias (knowing the value to the shift) to bring the scene back to a neutral/white balance?

It is just an idea and was wondering if there was a way to work this in camera rather than PP.


I really appreciate all the attention to this matter.

Click to expand...

This is rather straightforward to achieve in the camera. Just use a gray card or a white sheet to photograph under those conditions you've mentioned (i.e. fluorescent light and flash without the Plusgreen filter) and use that image to set the custom WB setting of the camera. Afterwards, use the custom WB for that shoot. That's pretty much it.

HTH,

Thank you all for your attention and diligent effort in understanding such a difficult topic.

I look forward to implementing your suggestions and sharing the results.

Regards,

Cem_Usakligil said:

This is rather straightforward to achieve in the camera. Just use a gray card or a white sheet to photograph under those conditions you've mentioned (i.e. fluorescent light and flash without the Plusgreen filter) and use that image to set the custom WB setting of the camera. Afterwards, use the custom WB for that shoot. That's pretty much it.

HTH,

Click to expand...

Cem,

The two scenarios do not give the same mathematics. When one balances the lights with filters, the values of the numbers that are low will be recorded with a certain accuracy depending on how many photons arrive and what the noise level is at various stages. The exposure will have to be longer to account for the light loss of the filter for the wavelengths one wants conserve. This maximized the photon counts for the lower numbers. One, in practice is strengthening that accuracy of counting photons in the image you will finally get. If this is done in PS, then one is boosting the value of lower numbers to achieve the required color balance. That is not as good as having those levels to begin with. So it's seems to me to be always best to expose with the spectral distribution one wants for the type of image one is making.

If the image is bright, the difference might not be great. However, with a full range of tonalities, likely the image will be better where corrections are made at the time of exposure.

Asher

Michael Desilets said:

Under fluorescent lighting requiring a Cinegel Tough ½ Plusgreen filter value of CC15 Green to my strobe to balance the light, could I conceivably use the green/magenta bias (knowing the value to the shift) to bring the scene back to a neutral/white balance?

Click to expand...

Given the generally problematic emission spectrum of fluorescent lighting, it is best to modify that at the source. Because of practical reasons (having to fit all fluorescents with daylight sleeves) you instead chose to modify the strobes so they match the fluorescents. If you didn't, you'd have mixed lighting which is a pain to colorbalance afterwards. An internal camera adjustment cannot correct for mixed lighting.

Since you now have an overall greenish lighting (fluorescents and filtered strobes), it is relatively easy to white balance by creating a "Custom White Balance" in your camera and shoot with that. That's much easier than adding an offset to another colorbalance. The overall correction for the greenish lighting should be pretty good for spectrally uniform materials, but for critical colors you may still suffer some color inconsistency or metamerism effects. It is better to have a daylight type of lighting from the onset.

Alternatively you could delay the whole color balancing (from green to neutral) to postprocessing, because there is little difference between the camera applying a (custom) color balance, or the Raw converter on your computer. They both alter the same Raw data, they are both doing postprocessing. The only benefit from doing it in camera is that your embedded thumbnails have decent colors.

Cheers,
Bart

Cem_Usakligil said:

Hi Doug,

This is a great essay. Although I already knew a lot about color spaces and WB, this essay has helped me understand certain issues much more clearly than I did before. The information presented is generally applicable regardless of the fact that it is about the WB shift of the Canon cameras. For example, it gives a good insight into why WB controls in raw converters have G/M as their tint axis.

Click to expand...

I agree with the above statement. As a newbie, it's good for me to read stuff like this. My background is mostly in family portraits and wedding photographer. It's good for me to read things that are slightly over my head, and learn new things about color and features. Thanks for the information.

Hi, Gina,

Gina Stephen said:

I agree with the above statement. As a newbie, it's good for me to read stuff like this. My background is mostly in family portraits and wedding photographer. It's good for me to read things that are slightly over my head, and learn new things about color and features. Thanks for the information.

Click to expand...

Thanks for writing.

Best regards,

Doug

Thanks for the warm welcome Doug! As I said, some of this stuff flies above my head, but I think that is a good thing! I come from the family of a wedding photographer and a science nerd. I want to learn more about the science of our art so I can do a better job with my pictures!

Hi everyone. I'm new to this forum, and have been very impressed with the quality of content. I stumbled upon it while researching how to properly use a Sekonic C-500 Color Meter with my Canon camera. I am contemplating getting the C-500, and I want to make sure that it will work for me before I spend any money.

First, Doug, the picture you included in the original post appears to be missing from the hosting site. Would it be possible to re-post the image?

Let me start with my current situation, and why I think the C-500 will help me, and then I will ask the questions that I am trying to understand. I do a lot of indoor gymnasium sports shooting with ambient light. I shoot RAW, but I want the WB setting to be at a good starting point so that I don't have to do much post process color correction. So, to do this, I first take a reading of light with a light meter so that I can set my manual exposure. Then I set my custom white balance function with an X-Rite ColorChecker Passport card. This process generates some great results as a starting point, and it saves me a lot of time in post process.

Sometimes it is awkward and intrusive to shoot the ColorChecker card when there are other public activities going on (I like to take that shot in the middle of the gym floor where the lighting is consistent with the later shots). I started to explore color meters, and was surprised by how expensive they were. But, for that price, I thought that the quality of result (color accuracy) should be superior to my current approach.

So I began to see what I could learn about how to use a C-500 with my Canon camera, and was surprised how little information there was from either company, or from any real people.

Piecing together what I have gleaned, here is how I think it would work, and I would love to get any informed feedback and commentary. The C-500, when reading ambient light, provides the color temperature in Kelvin, and the Green/Magenta shift filter compensation in units they call CC Index values. For my camera, I believe that I would put it in Custom Temperature mode, and inputting the Kelvin reading directly. Then for the CC Index value, I would go into the White Balance Correction screen. Since the Kelvin reading gets the Blue/Amber level right, that would be set at 0, and the Green/Magenta data would be converted to a corresponding value on the vertical axis only. (Does this make sense? I think it does, but I may be missing something.)

The conversion of CC Index value to the steps (0 to 9) on the Canon screen is the part that has no documentation anywhere. The only thing ai could find is a dpreview.com post from five years ago, where kbentsen did his own test and found that 1 Canon step was equal to 4.3 CC Index values.

Assuming that kbentsen is right, let me use this example: I get a reading from the Sekonic C-500 that shows 4600K, and 18G. This means that I should use a corresponding level of green filter on my lights. But I am in a gym and can't control the lights, so I want to make that green adjustment in the camera. So, I enter 4600 in the Custom Temperature setting, and I set the WB Shift 4 steps towards Green (18 / 4.3).

My questions:
1) does my understanding make sense?
2) does anyone have a C-500 and use it like this
3) would this be more accurate than my current ColorChecker card Custom White Balance approach?

Thanks in advance for any insights!
Nick

...one more question

...related to my last post...

I am thinking about renting a Sekonic C-500 and just running some tests myself to see if I can answer my own questions.

Any recommendations on how I should design and conduct a meaningful test?

Essentially I want to know how to use the readings from the C-500 used in ambient light mode with my Canon camera in a way that will give me better White Balance results than I could with just a grey card and the Custom White Balance function.

Thanks,
Nick

Hi Nick,

Welcome to OPF, I hope you will like it here. I don't have the answers you are looking for. I am sure that Doug will eventually chime in but I think that he is on the move for the next few days. So don't get discouraged if it takes a while.

Nick Pudar said:

Hi everyone. I'm new to this forum, and have been very impressed with the quality of content. I stumbled upon it while researching how to properly use a Sekonic C-500 Color Meter with my Canon camera. I am contemplating getting the C-500, and I want to make sure that it will work for me before I spend any money.

Click to expand...

Hi Nick,

Welcome to OPF.

The Sekonic C500 Color Meter is unique in the sense that it not only has a traditional (film based) filter set for measuring the spectrum, but also a more Digital filter set. That should help in getting things a bit more predictable when used with a DSLR or Medium Format Digtal Back.

However, these meters are more suited for balancing the incident and ambient light colors (by the use of light modifier filters and/or camera/lens filters), than as a guide for dialing in a WB setting for postprocessing.

The reason for that is two-fold. For one the Color Temperature and Tint controls between different Raw converters / photo-editors will give different value readouts when you click on something with known spectrally neutral reflection. So input of a given Color temperature will result in different color rendering responses. The other reason is that it is much easier (and cheaper) to use something like a WhiBal or a ColorChecker card to get a neutral White Balance with a mouse click in the Rawprocessor, assuming it is used correctly (i.e at the position of the main subject, and perpendicular to the camera).

I do a lot of indoor gymnasium sports shooting with ambient light. I shoot RAW, but I want the WB setting to be at a good starting point so that I don't have to do much post process color correction. So, to do this, I first take a reading of light with a light meter so that I can set my manual exposure. Then I set my custom white balance function with an X-Rite ColorChecker Passport card. This process generates some great results as a starting point, and it saves me a lot of time in post process.

Click to expand...

That's the best way to do it, IMHO, as long as you shoot Raw image files. It also takes into account a certain amount of reflection from the floor color, as long as you measure the WB at approx. eye- or chest level.

Sometimes it is awkward and intrusive to shoot the ColorChecker card when there are other public activities going on (I like to take that shot in the middle of the gym floor where the lighting is consistent with the later shots).

Click to expand...

This won't be any different with a Color meter, that too must be used just like an incident light meter, at the subject position.

So the best you can do is set a Color Temperature and Tint in you Camera if you want the JPEG preview to be approximately correct (based on experience, the balance between fluorescents and other light sources, and the reflection from the floor color), or use a camera preset like Fluorescent or Tungsten, although the Autosetting is often not too bad (it will never be accurate with such mixed lighting anyway).

Cheers,
Bart

Cem, Bart, thanks for the welcome!

Bart_van_der_Wolf said:

However, these meters are more suited for balancing the incident and ambient light colors (by the use of light modifier filters and/or camera/lens filters), than as a guide for dialing in a WB setting for postprocessing.

The reason for that is two-fold. For one the Color Temperature and Tint controls between different Raw converters / photo-editors will give different value readouts when you click on something with known spectrally neutral reflection. So input of a given Color temperature will result in different color rendering responses. The other reason is that it is much easier (and cheaper) to use something like a WhiBal or a ColorChecker card to get a neutral White Balance with a mouse click in the Rawprocessor, assuming it is used correctly (i.e at the position of the main subject, and perpendicular to the camera).

Click to expand...

Bart, I appreciate the reminder about the Raw processor affects. That leads me to a different question, which strays slightly from my original post. (I am always amazed by the fact that the more I know, the more there is to learn!)

I currently use the ColorChecker card for my white balance to then set the camera with its Custom White Balance function. I also take a shot of the Gretag Macbeth color chart of the passport for later color correction. I import the photos into my Lightroom Catalog, and I use the X-Rite Plugin to adjust the colors for my camera sensor.

I do not reset the white balance by using the eyedropper on the WhiBal card, since I believe the camera Custom White Balance already took care if that.

My new question is about how the Gretag Macbeth adjustment for my sensor may be messing things up, or whether it is not a factor in the White Balance part of the process.

Anyway, back to my original question. I'm still wondering if the C-500 could (with appropriate testing and translation tables) be used as I am hoping with a quality result.

Bart_van_der_Wolf said:

This won't be any different with a Color meter, that too must be used just like an incident light meter, at the subject position.

Click to expand...

Yes, you are right, but it is much easier to walk onto the floor and take a quick reading with the meter than to have to have someone stand and hold the chart for me to take a picture of. Most time, I don't have someone else to hold the card, so I perch it on the floor (at a slight angle to get less of the floor coloring reflection). I realize this is not good, but it is the best I can do in most cases.

I've been thinking a bit more about how I would set up a test with a rented C-500.
I would go to my daughter's high school gym, armed with a large, high quality grey card, a full-size Gretag Macbeth color chart, and my X-Rite ColorChecker Passport. I would replicate (with good annotation) my current process with the Passport, and also with the full size cards. I would have the pictures taken at appropriate height from the floor, and I would also do the floor-perch Passport card to later compare how off that really is. With the camera's Custom White Balance set with each set of cards, I would then take my reference photos of the Grey Card.

I would then, for comparison, use the C-500 to dial in the Custom Temperature and White Balance Shift in the camera. Again, I would then take other reference photo of the Grey Card.

Next, I would import everything into Lightroom. I would open all the photos of the Reference Grey Cards in Photoshop, convert them to Lab color space, and with the eyedroper tool, see where the grey card reads on the a and b channels. A good result should read a 0 on both of those channels at all points of the grey card.

There are some other testing aspects that I have not thought through to convert the C-500 CC Index values into appropriately translated WB Shift steps on the camera. I do realize that I have to solve this before I can do the above gymnasium test.

Does this test approach seem reasonable?

Hi, Nick,

Welcome to the forum.

Nick Pudar said:

Hi everyone. I'm new to this forum, and have been very impressed with the quality of content. I stumbled upon it while researching how to properly use a Sekonic C-500 Color Meter with my Canon camera. I am contemplating getting the C-500, and I want to make sure that it will work for me before I spend any money.

First, Doug, the picture you included in the original post appears to be missing from the hosting site. Would it be possible to re-post the image?

Click to expand...

Yes, I changed my web host.

I'm on-the-road until Sunday (2013.04.28), but when I get home I should be able to locate that figure and repost it.

I haven't thought much about this topic for a while, and I will have to revive my thoughts! But a quick look at your message suggests to me that you are on the right track.

Let me review it a little later today (after I really wake up!) and get back to you further.

Again, it is nice to have you with us.

Best regards,

Doug

Just a small niggle regarding the use of a reflective-neutral card as a target in connection with white balance color correction.

In theory, at least, the card should be at the subject location, and oriented parallel to the surface of the subject for which you are most keen to have the proper color correction.

The reason is that you want the card illuminated the same way "that surface of the subject" is by the various light sources.

Of course, which way it is oriented is only of any consequence if the subject is illuminated by different light sources, with their light arriving from different directions, and with their light having different chromaticities.

Of course, for a non-planar subject, which surface will we be most keen to have "correctly balanced"? Well, we must decide, Is it the model's left cheek, or right cheek, or forehead.

If we can't decide, then our only choice is to hold the card at whatever angle is handy and take "pot luck".

And of course in many cases of mixed light, we would not want "ideal" white balance color correction for all surfaces of the subject. If the subject's left side is illuminated by firelight, for example, we may well want that side to look "reddish-yellowish" (which ideal color correction would overcome).

Rules like "have the card face the camera" or "have the card face a direction halfway between the azimuths to the two principal light sources" may work, but only by accident.

Some may find this article interesting:

http://dougkerr.net/Pumpkin/articles/White_Balance.pdf

Best regards,

Doug

Just in the interest of greatest precision (and the lack of it here causes no misunderstanding), note that only light whose chromaticity falls on the Planckian locus (so-called blackbody radiation) has a color temperature.

For any other kind of light, the first property reported by a color meter of the type discussed in this thread, on Kelvin, is actually the correlated color temperature (CCT).

Again, there is no real misunderstanding here if we are not precise. Usually.

The second property reported is the distance (on a certain chromaticity plane) between the chromaticity of the light being measured and the chromaticity of the nearest point (reckoned on that plane) on the Planckian locus. And it is the color temperature of that chromaticity (yes it has one) that is called the correlated color temperature of the chromaticity of the light being measured.

I am not sure what the unit is for that quantity (the Planckian offset) as reported by the meter under discussion. (I probably knew once!). The basic scientific unit is the "u-v unit" - the unit of the scale of the axes of the CIE u-v chromaticity plane.

Some may find this article of interest:

http://dougkerr.net/Pumpkin/articles/Color_Temperature.pdf

Best regards,

Doug

Hi Nick,

Chiming in a little late here, but have you considered trying an incidence collector filter, such as an ExpoDisc (or a cheaper alternative) to determine your custom white balance? Certainly much easier to carry around than cards/charts, and you don't need any assistance to hold anything. So a darned side quicker to boot.
However, it's still not going to be the magic bullet in mixed light source situations; they will continue to be problematic.

John.

John,
Thanks, but I do have an ExpoDisc which I used before I got the ColorChecker Passport. It does a reasonable job, but as you said, it can't really compensate for the mixed lighting in the gym, (or the excessive floor coloring).
Nick

Hi, Nick,

Nick Pudar said:

John,
Thanks, but I do have an ExpoDisc which I used before I got the ColorChecker Passport. It does a reasonable job, but as you said, it can't really compensate for the mixed lighting in the gym, (or the excessive floor coloring).

Click to expand...

Nor can a neutral target, nor in fact, white balance color correction in general.

Best regards,

Doug

Doug,
I read the articles you linked to, as well as many more in your photography collection. Really great stuff!! I have a much better understanding of the content. Thank you.

I'm going to most likely rent a Sekonic C-500 to do my own tests as described in an earlier post of mine in this thread. As I mentioned, the important first step is to try to figure out what the transfer function is from the Sekonic CC Index (Wratten Filter designation) and the Canon WB Shift +/- 9 steps in the Green/Magenta axis.

Here is my thinking on that, and I'd appreciate any commentary or advice.
First, I bought some large lots of Wratten Filters on eBay from someone who obviously has no more use with them. There is a large collection of some used, and some unopened, in varying strengths and colors.

I'm thinking of gelling my strobes with varying levels of Green and Magenta and taking shots of a large high quality grey card. I will then remove the gels and take photos of the grey card at all 19 levels of the G/M axis of WB Shift. For all of these shots described above, I will also take a reading from the C-500 to see what the CC Index shift off of the CCT is being measured. I will then figure out how to plot all of this data to see where the equivalency levels are. Hopefully it is some sort of linear response so that future translations are easy to calculate. Once I have this mapping figured out, then I can do my other tests to see how the quality of using the Sekonic is.

Anyway, that's what I'm thinking for now.

Any thoughts would be appreciated.

Best regards,
Nick