E-M5 sensor nonlinearities

Started Jan 5, 2013 | Discussions thread
ForumParentFirstPreviousNextNext unread
Flat view
gollywop
Veteran MemberPosts: 5,441
Like?
E-M5 sensor nonlinearities
Jan 5, 2013

I just did an experiment with the E-M5 to see if there was any evidence of color shifts with reduced processor exposure that might occur due to possible sensor nonlinearities near saturation. All exposure evaluations referred to in the following were done using RawDigger.

I set up my CCPP and took a series of shots (M mode, f/5.6, ISO 200) with shutter speeds that went in 1/3 EV steps from severely underexposed to a complete blow-out of all 24 squares. I picked one that was a good overall exposure (down about 1/3 EV from ETTR) to act as a base, and I set its WB using D3. I then set the WB of all the other images to the same. The D5 square remained unblown right up to the penultimate shot, so I adjusted the level of this square on each exposure to equal its value on my base image.

The white square, D1, was, of course, the first to blow. I then examined several images starting with the one that was one exposure level less, in which the white square was 1/6 EV below clipping. Toggling through them, I could see no discernible differences in any of the colors, but, just to be sure, I used the eye dropper to measure and compare the RGB trio for about a dozen individual colors. They were all the same within a point or two. This, of course, is not surprising: although the while square was on the edge of clipping, all of the colors squares were still quite safely below clipping in all channels.

Next, I kept going up exposure until I got to the first color square that showed clipping. Naturally this was yellow, C4. Again, starting one exposure below, where the green channel of yellow was about 1/6 EV below clipping, I examined the RGB trio for the yellow square for several lesser exposures. They were all the same.

I then looked at the next square to blow, B5, and the results were the same.

I then went to the primary colors. Green was the first to blow, and again the next shot down was about 1/6 EV below clipping. Again there was no change in color as the exposure lessened. The same was true with Red, which was the next primary to clip.

Blue was the last primary to clip, and at this point the entire chart was essentially blown. I did see a bit of a color shift in the blue channel (the green value of the trio dropped about 8 points) as the exposure went down one notch from just-under-clipping, but it then stabilized.

It appears that exposing to within 1/6 EV of saturation results in virtually no color shifts with reduced processor exposure, a result that would seem to rule out any effects due to possible sensor nonlinearities. The very slight (but noticeable) change in the blue primary is problematic because of the extreme overexposure of the entire image. Even if it exists under better conditions, its practical significance is close to nil. True ETTR with the E-M5, it would appear, can be done with impunity.

-- hide signature --

gollywop

Nikon D1 Nikon D3
If you believe there are incorrect tags, please send us this post using our feedback form.
ForumParentFirstPreviousNextNext unread
Flat view
ForumParentFirstPreviousNextNext unread
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow