DPP: Exercise 8: Your Camera's Dynamic Range

The previous two exercises looked into perhaps the biggest issues in digital capture:  highlight clipping and noise.  These correspond to the camera's extremes of the tonal palette, the light and the dark, and it is this range that is examined here.

The brief calls for shooting a scene with a range of contrast in bright daylight .  It also calls for a white card, but this was omitted since the scene had a couple of reasonable sized white patches (the bar on the No Entry sign and the painted curb).

The scene was shot in AP, multi-metering mode, with no noise reduction.  Extreme brightness required using exposure compensation to dial down 1EV.   Playback showed some clipping on the white curb, but as the previous exercise indicated the camera overcompensated in its display of highlight clipping, I thought this exposure would be adequate and not show any clipping once imported into LR.

That proved not to be the case, with the curb showing 100% in all three channels.  This is a bit confusing and requires more investigation.

To determine the scene's dynamic range, the camera was switched to center metering mode, the telephoto engaged to maximum, and a meter reading made on each area of the scene where shutter speeds are indicated in the photo above.

As can be see, the curb tops out the camera's ability at 1/2000s, and the overall dynamic range presents as 8.3 stops.  According to data here, the camera's range in RAW is 12.3 stops (in tests using a Stouffer 4110 stepchart).

Here is the image as it appears in LR showing clipping at both ends.  Shadow clipping is minor, equivalent to a couple of spots at the end of the tunnel and in the dark rectangles left.

I shot the same with the Nikon D5100.  Playback showed some highlight clipping, but I ignored this suspecting the camera of reporting conservatively.  That proved not to be the case.  Not only were highlights clipped, so were shadows.  In-camera playback, though, showed no shadow clipping.

Metering results show a dynamic range of 7.7 stops.  As with the Sony, I had to step down one EV.  Test results here claim a range of 12.5 stops.

I posted to the OCA photography forum regarding the differences between the test results at Imaging Resource and my own results.  Perhaps the best reply received was:

It's like physics practicals in school, the conclusion is all about why one didn't get the expected theoretical result.

#

Oh, by the way, here's a link to a handy app for calculating the number of stops between any two settings:  http://imaginatorium.org/stuff/stops.htm

#