Doug Kerr
Well-known member
I haven't talked about white balance measurement with the ColorRight tool for a while. So lets do some of that.
The tool, in one of its functionalities, is essentially a white balance measurement diffuser. The traditional use of such a device is to allow us to use our camera to determine the chromaticity of the incident illumination for a shot setup. The camera can then use that determination to guide in-camera white-balance color correction.
The specific role of a measurement diffuser is to "gather" the components of light that comprise the overall illumination, potentially arriving from different directions,, mix them together, and present the mélange to the camera lens as a relatively-uniform luminous disk, which the lens "images" (very badly focused, but that is no disadvantage) on the sensor.
The camera, from the sensor output (often by later analysis of an image captured from the sensor) ascertains the chromaticity of the incident illumination upon the diffuser. This is then taken to be the chromaticity of the incident illumination upon the subject (since it is in the same place).
With a "classical" diffuser, the relative "weight" given in the mix to light components arriving from different angles is the same as the "weight" they will have in contributing to the illumination of the surface (which implies a "cosine" response by the diffuser).
Not surprisingly, since what we are intersted in is the chromaticity of the incident illumination upon the subject, we ideally should go to the subject to measure it. The classical diffuser measurement technique has us take the camera, with the diffuser affixed, to the subject location, orient it in an appropriate way (and I am not going to discuss here what that might be , since its not germane to where I am going), and make the measurement.
While the manufacturer of the ColorRight device, from time to time, recognizes this technique, he more strongly emphasizes the supposed unique ability of this diffuser to be successfully used in a much more convenient way: the diffuser is affixed to the lens, with the camera located where it will be for the actual shot, and pointed in the direction it will be pointed for the actual shot, and the measurement is taken.
Now it might at first seem that the result of this technique would not be very meaningful. After all, we want to know the temperature in Pittsburgh, but to avoid the inconvenience of travel, we will use a thermometer where we are, in Philadelphia. Really!
But in fact many testimonials and demonstrations suggest that using the ColorRight diffuser in this way often produces quite workable white balance color correction results (often better-looking to many tastes than with correction done via the reviled Automatic White Balance feature of the camera). And we have even experienced this to some degree in testing here.
How can this be?
The answer seems to be this. In many shot setups, the illumination on the camera (in its position for the shot) will be similar to the illumination on the subject. This would be easy to understand in many outdoor shots, and also in many indoor situations, for example, in a restaurant or ballroom, or even a living room. So we in effect perform a classical diffuser-based white balance measurement of the illumination on the camera, and have the camera treat that as the chromaticity of the illumination on the subject.
Now what unique properties does the ColorRight tool have, vis-à-vis other measurement diffusers, that might allow it to exploit this situation batter than other measurement diffusers? It has a somewhat "narrower" pattern of weighting of the incident light components as a function of angle of incidence. The manufacturer suggests that this is the key to the device's success, in that it causes it to better concentrate on light emerging from the subject.
But that would hardly seem to be an advantage. The light reflected from the subject has a chromaticity highly influenced by the reflective color of the subject. And it is the chromaticity of the incident light we really want the camera to determine. (The light reflected by the subject, of course, is part of the incident light on the camera, just as the light reflected from the photographer's shirt is part of the incident light on the subject.)
So that characteristic is probably counter-productive.
Interestingly enough, the manufacturer attributes the achievement of this "desirable" property to a unique design feature of the device: the diffuser disk is masked to a relatively small functional diameter. How would that narrow the acceptance pattern of the diffuser? It wouldn't. (Tests here have confirmed this.)
So where does the "narrower" pattern come from, if the masking doesn't produce it? It is just the inherent acceptance pattern of any simple diffuser. The wider "cosine" acceptance pattern of classical measurement diffusers is achieved only through the careful application of special techniques, such as an array of tiny lenses in front of the diffuser disk proper.
So what is the real advantage of the narrower pattern of the ColorRight device? Perhaps it is only that the device doesn't need any such auxiliary optical components, thus reducing the complexity of manufacture.
And so then what is the point of masking all but a small circle in the center of the diffuser disk? I leave that up to the reader to conclude.
Of course, there may be a much better explanation of the operation of the tool than this, grounded in some clear theoretical concept. We continue to await the revelation of this by the manufacturer. Perhaps we will read it in the patent when it issues.
The tool, in one of its functionalities, is essentially a white balance measurement diffuser. The traditional use of such a device is to allow us to use our camera to determine the chromaticity of the incident illumination for a shot setup. The camera can then use that determination to guide in-camera white-balance color correction.
The specific role of a measurement diffuser is to "gather" the components of light that comprise the overall illumination, potentially arriving from different directions,, mix them together, and present the mélange to the camera lens as a relatively-uniform luminous disk, which the lens "images" (very badly focused, but that is no disadvantage) on the sensor.
The camera, from the sensor output (often by later analysis of an image captured from the sensor) ascertains the chromaticity of the incident illumination upon the diffuser. This is then taken to be the chromaticity of the incident illumination upon the subject (since it is in the same place).
With a "classical" diffuser, the relative "weight" given in the mix to light components arriving from different angles is the same as the "weight" they will have in contributing to the illumination of the surface (which implies a "cosine" response by the diffuser).
Not surprisingly, since what we are intersted in is the chromaticity of the incident illumination upon the subject, we ideally should go to the subject to measure it. The classical diffuser measurement technique has us take the camera, with the diffuser affixed, to the subject location, orient it in an appropriate way (and I am not going to discuss here what that might be , since its not germane to where I am going), and make the measurement.
While the manufacturer of the ColorRight device, from time to time, recognizes this technique, he more strongly emphasizes the supposed unique ability of this diffuser to be successfully used in a much more convenient way: the diffuser is affixed to the lens, with the camera located where it will be for the actual shot, and pointed in the direction it will be pointed for the actual shot, and the measurement is taken.
Now it might at first seem that the result of this technique would not be very meaningful. After all, we want to know the temperature in Pittsburgh, but to avoid the inconvenience of travel, we will use a thermometer where we are, in Philadelphia. Really!
But in fact many testimonials and demonstrations suggest that using the ColorRight diffuser in this way often produces quite workable white balance color correction results (often better-looking to many tastes than with correction done via the reviled Automatic White Balance feature of the camera). And we have even experienced this to some degree in testing here.
How can this be?
The answer seems to be this. In many shot setups, the illumination on the camera (in its position for the shot) will be similar to the illumination on the subject. This would be easy to understand in many outdoor shots, and also in many indoor situations, for example, in a restaurant or ballroom, or even a living room. So we in effect perform a classical diffuser-based white balance measurement of the illumination on the camera, and have the camera treat that as the chromaticity of the illumination on the subject.
Now what unique properties does the ColorRight tool have, vis-à-vis other measurement diffusers, that might allow it to exploit this situation batter than other measurement diffusers? It has a somewhat "narrower" pattern of weighting of the incident light components as a function of angle of incidence. The manufacturer suggests that this is the key to the device's success, in that it causes it to better concentrate on light emerging from the subject.
But that would hardly seem to be an advantage. The light reflected from the subject has a chromaticity highly influenced by the reflective color of the subject. And it is the chromaticity of the incident light we really want the camera to determine. (The light reflected by the subject, of course, is part of the incident light on the camera, just as the light reflected from the photographer's shirt is part of the incident light on the subject.)
So that characteristic is probably counter-productive.
Interestingly enough, the manufacturer attributes the achievement of this "desirable" property to a unique design feature of the device: the diffuser disk is masked to a relatively small functional diameter. How would that narrow the acceptance pattern of the diffuser? It wouldn't. (Tests here have confirmed this.)
So where does the "narrower" pattern come from, if the masking doesn't produce it? It is just the inherent acceptance pattern of any simple diffuser. The wider "cosine" acceptance pattern of classical measurement diffusers is achieved only through the careful application of special techniques, such as an array of tiny lenses in front of the diffuser disk proper.
So what is the real advantage of the narrower pattern of the ColorRight device? Perhaps it is only that the device doesn't need any such auxiliary optical components, thus reducing the complexity of manufacture.
And so then what is the point of masking all but a small circle in the center of the diffuser disk? I leave that up to the reader to conclude.
Of course, there may be a much better explanation of the operation of the tool than this, grounded in some clear theoretical concept. We continue to await the revelation of this by the manufacturer. Perhaps we will read it in the patent when it issues.