same , if i want to still motion i use strobes 1/2000 sec :-)
March 29, 2023
422
-
-
After reading your comment I downloaded RawDigger as a trial.
Using RawDigger I looked at a couple of old raw files where I had pushed the camera's histogram as far as I could to the right either without clipping or about 1/2 a stop more.
On each of the raw files RawDigger was telling me I could have actually gone about 1 stop past where the camera's histogram started to indicate highlight clipping.
I then did 2 test shots of the same sunlit scene with some deep shadows under trees and decking.
The first shot pushed the camera histogram all the way to the right to the point of clipping. The second shot was with the same aperture and ISO but with 1 stop slower shutter speed. Reviewing the second shot in camera (based on jpeg output from the raw data) showed significant clipping in the few clouds in the sky as expected.
Looking at the the two raw files in RawDigger, RawDigger showed what I expected in that the histogram data for the first test shot stopped about 1 stop to the left of the raw data clipping point. For the second shot RawDigger showed the right side of the histogram was at or very close to the raw data clipping point as I expected.
Both shots were then edited to output a nice looking jpeg without any highlights clipped.
So this quick and very simple test confirmed what I already knew - that I can safely add about 1/2 stop more exposure* from where the camera histogram begins to show clipping without clipping the raw data.
Looking at the deep shadows in both shots at 100% showed that the difference in the amount of visible noise was negligible for all intents and purposes.
This also confirms for me that as I posted either earlier this thread or in another thread, the camera's histogram is a close approximation to the raw histogram for my purposes and final image goals.
* exposure - amount of light striking the sensor per unit area while the shutter is open
** optimal exposure - the maximum exposure* within dof and motion blur requirements without clipping important highlights.
*** under exposed - more exposure* could have been added with the DOF and blur constraints still being met without clipping important highlights. -
thats why on my a74 i have set the histogram to 110% setting so its much closer to the raw DR.
-
...or ISO.
-
Yes, that is true and is typically what I do if I have already set the optimal exposure**.
* exposure - amount of light striking the sensor per unit area while the shutter is open
** optimal exposure - the maximum exposure* within dof and motion blur requirements without clipping important highlights.
*** under exposed - more exposure* could have been added with the DOF and blur constraints still being met without clipping important highlights. -
That's going to vary with the lighting spectrum and the reflectivity spectrum of your subject.
-
We have two approaches on what to do if we can increase ISO after maximizing exposure and without clipping relevant highlights.
One approach has been to apply ETTR even at higher ISOs. I assume the reason is to reduce the read noise. However, based on P2P measurements, I do not see relevant noise improvements above the bump. Are there any examples where we see the effectiveness of that approach?
The other approach I prefer for ISOs above the dual conversion gain is to avoid raising ISO to avoid highlight clipping.
I would love to hear your opinions on why one approach is preferred. -
As you know. I only seriously pursue ETTR at base ISO. The small read noise differences at high conversion gain are, in my experience, totally inconsequential. However, I've not tested every camera on the market.
-
In the mean time is it possible to just put a set of basic explanations into a locked pinned thread with a link to the relevant discussion thread? Give the pinned post a title like photography basics or basic photography terms explained.
-
In the meantime, if any actual beginners should wander thru this series of posts, here are a few well written basic explanations from knowledgeable sources on relevant topics...
Adobe Whitepapers / Technical Papers
Understanding Digital Raw Capture
Raw Capture, Linear Gamma, and Exposure
Rendering the Print: the Art of Photography
Glossary
American Society of Media Photographers - Glossary of Terms
PrintWiki - from the Encyclopedia of Graphic Communications
Tutorials
-
"The common factor in all color filter array cameras is that, no matter what color filter arrangement is used, each element in the sensor captures only one color. The red-filtered elements produce a grayscale value proportional to the amount of red light reaching the sensor, the green-filtered elements produce a grayscale value proportional to the amount of green light reaching the sensor, and the blue-filtered elements produce a grayscale value proportional to the amount of blue light reaching the sensor." - dangerous misconception.
This article, also by late Bruce Fraser, contains a basic mistake. "Linear processed raw captures look very dark." is the matter of the assigning the wrong colour space, with wrong gamma. I would avoid recommending this article.
-
And this is also problematical: "Linear capture has important implications for exposure. If a camera captures six stops of dynamic range, half of the 4,096 levels are devoted to the brightest stop, half of the remainder (1,024 levels) are devoted to the next stop, half of the remainder (512 levels) are devoted to the next stop, and so on. The darkest stop, the extreme shadows, is represented by only 64 levels—as shown above."
This ignores the fact that cameras almost always have enough noise to properly dither the ADC inputs. Maybe in Bruce's day, they didn't, but I doubt it, since in his day were were mostly talking about CCDs with off-sensor ADCs.
I remember Mike Collette used a similar approach in one of his introduction to raw capture talks. I approached him about it, and he said that it wasn't really germane, but provided motivation to give generous exposure, which is what the takeaway should be. "Hmph," I replied.
-
This is just an excellent source of misconceptions.
-
I highly recommend reading anything from Bruce Fraser. He writes in a manner that's clear, concise, and approachable by a wide audience.
-
True. On a funny note, 6 is not the same as 5 (p.2, fig. of linear distribution vs. gamma-corrected distribution). Those and other raw things in the article were pointed out to Bruce and Adobe many years ago, but...
-
and wrong.
-
They also write in a manner that is clear and concise for beginners.
-
Again,
"For every complex problem there is an answer that is clear, simple, and wrong."-- H. L. Mencken
