Interesting.
I suppose film (linear) would be the opposite of digital.
Seems to confirm that, eh?
If I understand this correctly it disagrees with the ETTR concept. But I probably don't understand it correctly since there is so much in the digital world I don't. Can anyone clarify? Thanks.
I'm not quite sure I agree with his entire concept of testing in this manner.
--Bob
Exposing to the right is simple, extend the exposure to the right side of the histogram without clipping the highlights. I tend to do this when I want to control noise although at times I have used it to improve on the dynamic range of the subject.
It is well known that overexposing digital is not the best thing to do. When exposing digital images attention to expose from an important bright area should be the goal. That gets good details from the bright areas without clipping. The dark areas in general are not problematic since opening up to 3 stops of light bring details to those areas although we have to be careful because noise could show its ugly face. It is easir to rescue details from underexposed areas than it is from the overexposed highlights.
This is what I know and it has worked for me. Mileage can vary.
I've said the same thing many times, but get used to the opposite reaction here on the Hog. I do underexpose as a rule with Canon gear...
Wanderer2 wrote:
If I understand this correctly it disagrees with the ETTR concept. But I probably don't understand it correctly since there is so much in the digital world I don't. Can anyone clarify? Thanks.
I don't think so. If there is any clipping, that's not ETTR. ETTR is exposing as much as possible without any clipping.
Exactly. ETTR is exposing to the right, not over exposing. It does take some actual testing to see where one's camera fails. Then, one will know how much additional exposure can be added. Metering the brightest portion of the scene is also critical. Select the wrong area and more than likely blown highlights will result.
--Bob
jackm1943 wrote:
I don't think so. If there is any clipping, that's not ETTR. ETTR is exposing as much as possible without any clipping.
Nothing new, but indeed very interesting to see it in a side-by-side-comparison!
rmalarz wrote:
I'm not quite sure I agree with his entire concept of testing in this manner.
--Bob
The methodology leaves a bit to be desired but the findings are consistent with what show in
Film vs. Digital Characteristic Curves.
What Michael Zhang left out is that you can vary the DR of film by under-developing (DR increases) or over-developing (DR decreases).
If you want to reply, then
register here. Registration is free and your account is created instantly, so you can post right away.