Doug Kerr
Well-known member
I have just posted to my technical information site, "The Pumpkin" a new tutorial article,"White Balance Diffusers in Digital Photography", available here:
http://doug.kerr.home.att.net/pumpkin/index.htm#WB_Diffusers
Let me take a moment to describe what impelled me to prepare this article at this time.
The emergence of a new "white balance tool", the Color Parrot, by Drew Strickland (proprietor of the Pro Photo Home forums), has lead to an earnest round of discussion on that forum ("fomented a catfight" would perhaps be a more candid description).
The Color Parrot is what we may call a "white balance diffuser". It is an accessory to be mounted on the front of a camera's lens, enabling the camera to measure "something"‡ that is of use in the camera's applying color correction ("white balance correction") to the images under the camera's "custom white balance" feature.
‡ I say it that way since it is not clear what quantity it is intended to measure.
Much of the discussion arose from the fact that many of us were surprised that Strickland recommends, as the best technique for using the Color Parrot, making the measurement from the camera position (that is, where the camera will be for the shot), with the camera (Color Parrot equipped for the moment) aimed as it would be for the shot (i.e., "toward the subject").
It is widely (but certainly not universally) accepted that the information needed by the camera to conduct the color correction (in the theoretically classical way) is the chromaticity of the ambient light illumination on the subject (to say it in a more complicated way that we usually do, necessary for technical precision, I'm afraid).
It is also widely (but not universally accepted) that to assuredly do this, we must make a measurement at the subject's location, using an "instrument" (such as our camera equipped with a white balance diffuser) that accepts the incident light the same way that light influences the illumination of the subject during actual "shots".
In the face of that concept, it is hard to see how a measurement from the camera position could glean the necessary information (unless, of course, the overall incident light falling on the camera position were the same, insofar as the chromaticity of the illumination it provided) as that falling on the subject, which is certainly nearly true in some situations).
I asked Strickland to help me bridge the gap between my concept of these principles and the modus operandi of the Color Parrot in my (now infamous) post, "How do it know?". He gently chastised me for muddling up what was supposed to be a joyous celebration of the birth of a new, and convenient, tool by asking scientific questions. He emphasized that scientific understanding, while certainly worthwhile, was not so important here as the fact that the thing works very well (he says) and that people are ordering it.
I continued my quest for insight. After several days, and a lot of statements by Strickland, I still had not received any (to me) satisfying (or even understandable) concepts of how this device accomplished what I believed to be the necessary result in that matter.
One property that Strickland said was responsible for the superior performance of the Color Parrot was that it was "more targeted" than other well-known white balance diffusers, such as the ExpoDisc. I asked exactly what he meant by that, but he did not explain it in any technical terms I could recognize. But the context of the discussion suggests that he means that the "acceptance sensitivity pattern" of the Color Parrot is narrower than that of other diffusers. The object of this is evidently to make the instrument more concentrate its examination on the light arriving from places near the camera axis (that is, in the typical case, from the subject). Of course, I am still baffled as to how that would contribute to the instrument being able to make what I have always thought is the relevant determination (the chromaticity of the net illumination on the subject).
Eventually, Strickland attempted to foreclose the fruitless discussion by saying, in effect, that "There were six guys, who probably didn't intend to buy a Color Parrot anyway, who have asked a lot of questions, and over one hundred who have ordered one. Who do you think is right?" This question, certainly worthy of an election year, ranks right up there, in my mind, with such classics as, "The moon is smaller than the Earth, but is it farther away?"
In any case, not wanting to spend the rest of eternity in the colorimetric purgatory of Drew Strickland's "B" list, I diverted the month's lunch money to ordering a Color Parrot and an ExpoDisc (the latter being something that I had always wanted anyway).
Of course, as you would expect of me, I will probably not much use these in the service of serious photography. Rather, I will examine them, test them (as best I can with my limited laboratory facilities), reverse-engineer them (Ah, magic crystals, you can run but you can't hide), and see what they do in a few actual photographic situations. Perhaps in fact I will discover a scientific principle of which I was previously unaware, or had accidentally ignored, that allows me to see how the Color Parrot do the voodoo that Strickland says it do so well.
Looking ahead to my report on this project, I felt that I would need to, by way of introduction, explain some of the scientific principles I felt were relevant. But I realized that this would lead to a 12-page preamble to an 8-page report.
Accordingly, I decided to prepare the background discussion as a free-standing article, to which I could refer the readers of my report when it is prepared. It is that piece that I introduce in this note.
http://doug.kerr.home.att.net/pumpkin/index.htm#WB_Diffusers
Let me take a moment to describe what impelled me to prepare this article at this time.
The emergence of a new "white balance tool", the Color Parrot, by Drew Strickland (proprietor of the Pro Photo Home forums), has lead to an earnest round of discussion on that forum ("fomented a catfight" would perhaps be a more candid description).
The Color Parrot is what we may call a "white balance diffuser". It is an accessory to be mounted on the front of a camera's lens, enabling the camera to measure "something"‡ that is of use in the camera's applying color correction ("white balance correction") to the images under the camera's "custom white balance" feature.
‡ I say it that way since it is not clear what quantity it is intended to measure.
Much of the discussion arose from the fact that many of us were surprised that Strickland recommends, as the best technique for using the Color Parrot, making the measurement from the camera position (that is, where the camera will be for the shot), with the camera (Color Parrot equipped for the moment) aimed as it would be for the shot (i.e., "toward the subject").
It is widely (but certainly not universally) accepted that the information needed by the camera to conduct the color correction (in the theoretically classical way) is the chromaticity of the ambient light illumination on the subject (to say it in a more complicated way that we usually do, necessary for technical precision, I'm afraid).
It is also widely (but not universally accepted) that to assuredly do this, we must make a measurement at the subject's location, using an "instrument" (such as our camera equipped with a white balance diffuser) that accepts the incident light the same way that light influences the illumination of the subject during actual "shots".
In the face of that concept, it is hard to see how a measurement from the camera position could glean the necessary information (unless, of course, the overall incident light falling on the camera position were the same, insofar as the chromaticity of the illumination it provided) as that falling on the subject, which is certainly nearly true in some situations).
I asked Strickland to help me bridge the gap between my concept of these principles and the modus operandi of the Color Parrot in my (now infamous) post, "How do it know?". He gently chastised me for muddling up what was supposed to be a joyous celebration of the birth of a new, and convenient, tool by asking scientific questions. He emphasized that scientific understanding, while certainly worthwhile, was not so important here as the fact that the thing works very well (he says) and that people are ordering it.
I continued my quest for insight. After several days, and a lot of statements by Strickland, I still had not received any (to me) satisfying (or even understandable) concepts of how this device accomplished what I believed to be the necessary result in that matter.
One property that Strickland said was responsible for the superior performance of the Color Parrot was that it was "more targeted" than other well-known white balance diffusers, such as the ExpoDisc. I asked exactly what he meant by that, but he did not explain it in any technical terms I could recognize. But the context of the discussion suggests that he means that the "acceptance sensitivity pattern" of the Color Parrot is narrower than that of other diffusers. The object of this is evidently to make the instrument more concentrate its examination on the light arriving from places near the camera axis (that is, in the typical case, from the subject). Of course, I am still baffled as to how that would contribute to the instrument being able to make what I have always thought is the relevant determination (the chromaticity of the net illumination on the subject).
Eventually, Strickland attempted to foreclose the fruitless discussion by saying, in effect, that "There were six guys, who probably didn't intend to buy a Color Parrot anyway, who have asked a lot of questions, and over one hundred who have ordered one. Who do you think is right?" This question, certainly worthy of an election year, ranks right up there, in my mind, with such classics as, "The moon is smaller than the Earth, but is it farther away?"
In any case, not wanting to spend the rest of eternity in the colorimetric purgatory of Drew Strickland's "B" list, I diverted the month's lunch money to ordering a Color Parrot and an ExpoDisc (the latter being something that I had always wanted anyway).
Of course, as you would expect of me, I will probably not much use these in the service of serious photography. Rather, I will examine them, test them (as best I can with my limited laboratory facilities), reverse-engineer them (Ah, magic crystals, you can run but you can't hide), and see what they do in a few actual photographic situations. Perhaps in fact I will discover a scientific principle of which I was previously unaware, or had accidentally ignored, that allows me to see how the Color Parrot do the voodoo that Strickland says it do so well.
Looking ahead to my report on this project, I felt that I would need to, by way of introduction, explain some of the scientific principles I felt were relevant. But I realized that this would lead to a 12-page preamble to an 8-page report.
Accordingly, I decided to prepare the background discussion as a free-standing article, to which I could refer the readers of my report when it is prepared. It is that piece that I introduce in this note.