What I learned from Gollywop -- and what I wonder

Started Mar 26, 2013 | Discussions thread
gollywop Veteran Member • Posts: 8,245
Re: if i wanted to test/compare ETTR....

Paul Auclair wrote:

against conventional method(s)(ie.a light meter reading or gray card reading) how would i do it or better yet...what should i be looking for?

lets take an 18% gray card for example lit 100% with strobe. very easy and controllable example.

lets say the light meter indicates f5.6 at 250 ISO 200 and the resulting image has an expected peak right in the middle of the histogram.

now i use ETTR and dial aperture and/or shutter value(s) to get +2 EV and the resulting peak is far right (no clipping HLs).

so now i get the RAW ETTR and pull exposure back two stops in PP.

if comparo is done right so far the eye dropper will say both converted RAWs are 128/128/128/ RGB but what am i looking for as far as better IQ and why ETTR is better?

Paul, the experiment that you suggest is not likely to exemplify the benefits of ETTR particularly well unless the scene with the gray card also includes some shadows, i.e., unless the histogram also has some important information in the lower regions to the left of the peak.

A better experiment would be, say, a snow scene or a sunny beach scene where the overall lightness of the brighter parts trick your meter into underexposing the shot -- the snow, for example, comes up grayish instead of white, and there is a stop or two of "flat line" on the right hand side of the histogram.

You would then want to retake this shot with adequate EC (or increased exposure) to move the histogram to the right-hand edge. Now process the two shots so they both have the same appropriate brightness. This may entail increasing the processing exposure for the underexposed shot and decreasing that somewhat for the ETTR shot (which may have turned out to make the snow too bright).

The main difference you will see is in the shadow regions, which will be better defined, have better structure, and have less noise than in the underexposed shot, but there will be better s/n throughout. In the brighter regions, this will be most apparent in broader homogenous spans such as blue skies.

Going the other way, consider an indoor shot (without flash) with a bright sunlit window. The metered shot will tend to bring up the darkish room and, as a result, blow out the window highlights. This is fine if you're really interested in a great shot of the interior and don't care about a nicely structured rendering of the window and its contents.

But, if you want the window portion as well, you would again ETTR so as to preserve the highlights in the window region, just avoiding the blinkies. The resulting image will now show a very dark interior, but, within reason, you will be able to pull up the shadowed interior in processing.

Here is an example:

Scene as shot ETTR:

Scene after processing:


Or, here's another, first as shot ETTR (with UniWB, which is the reason for the green tint)

And after processing.

-- hide signature --


Post (hide subjects) Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow