The Color Parrot - Report of test program

I have recently completed a test program intended to discern various aspects of the behavior and performance of the Color Parrot white balance diffuser.

An extensive report on my findings is available directly here:

...

Hi, Jerry,

This applies if you are actually using a "scene" for the WB reference frame. The camera makes its determination of incident light chromaticity from the center of the sensor frame (nominally the area corresponding to the circle in the finder), and that of course corresponds to the light received from objects near the camera axis.

When we use a diffuser, the camera still regards only the part of the "image" that falls on the center of the sensor frame. Let's consider what that light now represents.

Because the lens in this case is gravely out-of focus, this does not exactly correspond to the light from the central portion of the rear of the diffuser, but it probably does correspond somewhat. I think this is what you were talking about. (I plan to make some tests today to get some insight into this.)

This does not in any way correspond to the light arriving at the diffuser from objects near the camera axis (which some reader may think is what you are referring to).

I think one of your points is that the transmission through the center spot of the ColorParrot might have a different chromatic neutrality at its center than elsewhere. That is certainly possible (especially if the transmissivity of that overlay layer is not quite neutral, as is apparently the case for the other layers). That could then result in different chromatic response when used on Canon EOS cameras compared to other cameras whose WB measurement was not restricted to the center of the sensor frame.

In any case, I do not yet know how the center spot affects the matter of the "directivity" of the camera-with-diffuser. I would guess not much if at all.

Best regards,

Doug

This is the procedure listed for proper use of the Color Parrot on a 1Ds:

Canon 1DS

Yes, this is my point.

Yes in part, reflected light from objects near the camera's central axis will weigh heavily in computing white balance.

Exactly my point. If a centrally located subject is wearing a red dress and the background is white, it will affect the white balance reading through the diffuser

See the above response.

The custom white balance function of the Canon EOS 1Ds was not designed for transmissive light, but instead reflected light from a "white" source recorded with the center 2% spot.

Jerry

Hi, Jerry,

I thought so.

Not very much at all if measuring though a diffuser, even one like the Color Parry whose directivity pattern is narrower than that of, for example, an ExpoDisk.

That is because the light that strikes the front of the diffuser, arriving from nearly on-axis, still illuminates the entire face of the diffuser, and shows up over the entire rear of the the diffuser.

Its distribution over the rear surface, in luminance and chromaticity, is presumably (as you have conjectured) affected slightly by the "overlay spot", but the basic "diffuse receptor" behavior is not changed.

Only if the diffuser has a very narrow directivity pattern. (I assume you are speaking of a situation in which the subject does not fill a large angular subtense in front of the photographer. Of course, it did in the special test series I posted her recently.)

Note that this is not the same as the subject filling a large part of the frame in the actual shot, as when using a lens of substantial focal length. The "field of view" of the camera with the lens has almost no effect on the directivity pattern of the camera plus diffuser (See the comparisons in figure 2 of my report).

In any case, with a diffuser such as the ExpoDisk, or the Color Parrot , and a setup where the subject only subtends a modest angle, that will not occur to any consequential extent.

Note Figure 2 in my report. These directivity patterns show the relative emphasis of the camera with diffuser to light (and thus to the chromaticity of the light, as a part of the overall illumination on the diffuser face) arriving from different angles off-axis.

Suppose the subject in width subtended an angle of 20° from the camera position. I haven't done the numerical integration yet, but we can see that the influence of the chromaticity of that light on the overall "acceptance" of light by the diffuser would be very minor.

That's why, in my recent test shots, I needed to take the measurement with the Color Parrot from very near the subject to have the phenomenon show up in a very substantial way (so as to clearly illustrate the nature of the phenomenon ).

Indeed. We use these cameras with a diffuser as an "adaptation" of that situation.

Best regards,

Doug

I have just issued a further update of my report on testing of the Color Parrot.

In this issue, I have added to the chart of measured neutrality of the three white balance tools (figure 1 on page 12) a point showing the neutrality of our ExpoDisc as inferred from the manufacturer's numbers on the "QC card" included with that unit.

This shows good agreement with our measured result.

The same link will access the updated article. It is included here for convenience:

http://Pumpkin.Annex.home.att.net/ar...olorParrot.pdf

Those interested in interpreting the "QC card" numbers for their own ExpoDisc specimens should note that the three numbers (for "Red", "Green", and "Blue") are (so far as I can tell by reverse engineering) on a relative density basis. That is, the "relative luminance" for the respective "color channel" response is calculated this way:

Yrel=10^(-Drel)

where Drel is the relative density number stated on the card for that "color channel".

Best regards,

Doug