Is your light meter linear?

Alan Ross likes to ask this question.  If the response of your meter is linear, you will get the same image value whether you are metering a subject of high or low luminosity.  Some meters do better than others in this regard.  One way to test yours is to shoot targets of varying luminosity and compare the meter readings with the resulting image values.  Ideally, they should all be identical; in real life, they will vary a little bit.  If you have made an exposure-response curve for your system as in the previous pages, you can compare the meter-test image values to your systemís response curve, and determine the margin of error for your meter.

Here are the L* values for my test series.  I metered with the Pentax V and shot with the Nikon D700 at ISO 200, using the white and gray sides of a gray card in sunlight, shade, and window light.

linearity

Zooming in:

zoomed

Here are the minumum and maximum image values marked on a section of the exposure-response curve.  The readings vary by about 2/3 of a stop above and below zone V, but have a definite trendline that could be used to make adjusments on-the-fly.  (Something like:  below 6, add 1/2 stop; above 15, subtract 1/2 stop.)

range



I repeated the test, this time using the Nikonís meter, set to aperture priority:

zoomed

range

Looks like the Nikonís meter is a little better after all . . . .