Doug Kerr
Well-known member
This is to summarize in this section my recently-gained understanding of the actual function of the hemispherical "dome" found on most incident light exposure meters. I have discussed this in some tedious detail in the "theory" section.
We see such a dome in its usual form in this picture of a classical incident light exposure meter, which is, in this version, arguably the most beautiful exposure meter ever made, the Norwood Director Model B, as made by American Bolex (ca. 1948):
Norwood Director Model B
************
The object of incident light exposure metering is to determine the photographic exposure (combination of shutter speed and aperture) that will give each scene element an exposure on the sensor, relative to the saturation exposure, proportional to the reflectance of the element. Thus a white cat on a snowdrift will come out looking like a white cat on a snowdrift, not a gray cat on an ash heap (which could happen with an exposure determined by reflected light exposure metering.
We determine that photographic exposure my direct measurement of the illuminance of the incident light on the subject, using an incident light exposure meter.
This will only work "perfectly" if each subject element receives the same illuminance from the incident light. But we cannot be sure of this, because:
• The different subject surface elements face in differing directions, and
• The "potency" of (each portion of) the incident light may vary with the direction of its arrival.
A realistic "compromise" way to deal with this would be to base our exposure on the average luminance on all subject surfaces—well, all surfaces that the camera can see (obviously the illuminance on the far side of a subject has nothing to do with the matter of interest).
The hemispherical dome on an incident light exposure meter is a "proxy" for the overall surface of the subject (the part of it the camera can see). It is a reasonable match for the shape of a human face, always a preoccupation of exposure metering.
The totality of the incident light (from whatever directions) landing on the dome (and this is what the meter responds to) is proportional to the average illuminance on the dome and, by similitude, the average illuminance on the (visible to the camera areas of) a subject's face.
Thus, the use of a meter with a hemispherical dome fulfills our strategy that the most useful thing we can do is to base our exposure on the average luminance on the subject surface (the part of it visible to the camera).
Best regards,
Doug
We see such a dome in its usual form in this picture of a classical incident light exposure meter, which is, in this version, arguably the most beautiful exposure meter ever made, the Norwood Director Model B, as made by American Bolex (ca. 1948):
Norwood Director Model B
Photo © James Ollinger
************
The object of incident light exposure metering is to determine the photographic exposure (combination of shutter speed and aperture) that will give each scene element an exposure on the sensor, relative to the saturation exposure, proportional to the reflectance of the element. Thus a white cat on a snowdrift will come out looking like a white cat on a snowdrift, not a gray cat on an ash heap (which could happen with an exposure determined by reflected light exposure metering.
We determine that photographic exposure my direct measurement of the illuminance of the incident light on the subject, using an incident light exposure meter.
This will only work "perfectly" if each subject element receives the same illuminance from the incident light. But we cannot be sure of this, because:
• The different subject surface elements face in differing directions, and
• The "potency" of (each portion of) the incident light may vary with the direction of its arrival.
A realistic "compromise" way to deal with this would be to base our exposure on the average luminance on all subject surfaces—well, all surfaces that the camera can see (obviously the illuminance on the far side of a subject has nothing to do with the matter of interest).
The hemispherical dome on an incident light exposure meter is a "proxy" for the overall surface of the subject (the part of it the camera can see). It is a reasonable match for the shape of a human face, always a preoccupation of exposure metering.
The totality of the incident light (from whatever directions) landing on the dome (and this is what the meter responds to) is proportional to the average illuminance on the dome and, by similitude, the average illuminance on the (visible to the camera areas of) a subject's face.
Thus, the use of a meter with a hemispherical dome fulfills our strategy that the most useful thing we can do is to base our exposure on the average luminance on the subject surface (the part of it visible to the camera).
Best regards,
Doug