Ugly Hedgehog - Photography Forum
Home Active Topics Newest Pictures Search Login Register
Main Photography Discussion
Why and When to use ETTR
Page 1 of 21 next> last>>
Apr 26, 2016 06:26:48   #
selmslie Loc: Fernandina Beach, FL, USA
 
The question of "why" and "when" to use ETTR usually gets lost in the shuffle of the discussions of "what" (science) and "how" (procedure). The last two have been covered quite thoroughly others and arguments over definitions and nuances in these areas can get silly. But the first two questions get little coverage.

Often the "why" is brushed off with a blanket claim about noise reduction, wanting to capture more detail or wanting to use the sensors full range. "When" is almost never mentioned. The problem is that why and when are the most important questions we need to be thinking about.

You cannot convince someone that ETTR will reduce noise if they use a modern camera like a D8xx, D7xx or D6xx at low ISO and they can't see any noise in their own images from a normal viewing distance, or even much closer than normal viewing distance. Noise reduction is not a sufficient reason to use ETTR. The solution to noise is a high S/N and we all know how to get that - low ISO. One or two stops of additional exposure with ETTR only begins to reduce noise.

Neither are you going to persuade anyone to use ETTR to get more "detail" if their concept of detail is resolution of sharp edges and something they can actually see in their prints or images. There is no expectation of seeing detail based on its conventional meaning in out of focus areas or in Zone II - "Textured black; the darkest part of the image in which slight detail is recorded" or in Zone I or 0.

The detail that is being referred to in relation to ETTR are the fine separations between tonal levels - the number of distinct levels that drop by half in the raw file with each reduction of 1 stop of exposure. This is "loss of detail" has been discussed ad nauseam but I have yet to see it demonstrated in a real image. Even from the ETTR viewpoint, "detail" has little relevance in very dark areas, the same areas where the raw file appears to be deficient.

It is also taken for granted that we should want details and fine tonal distinctions in the shadows. Not always, maybe rarely.

The sensor's physical range (the point at which the buckets fill to overflowing) and the raw file's numeric range (the point where all of the bits are 1) are not the same thing. Other than at base ISO both are seldom fully used. ETTR tries to use up the bits without going too far. This is often glossed over because it is harder to explain.

What is almost never mentioned is the dynamic range of the scene. There are plenty of cases where the range of luminance values reaching the camera is narrow and others where it is quite wide. With a narrow range there is no benefit to be derived from ETTR, a normal exposure is adequate. With a very wide range, even ETTR cannot help, only HDR. So there is only a particular subset of cases where ETTR can actually provide some help.

To justify the effort and inconvenience of using of ETTR we have to be convinced that there are real visible benefits. It's the "visible" part that needs to be addressed.

I am inviting proponents of ETTR to answer these questions and provide full images as evidence.

We do not need PowerPoint slides, charts, graphs, tables or screen shots of edit sessions. We don't need 100% crops from an image unless you also provide the complete image. And the images in question should be scenes that we face in our normal lives like landscapes, street scenes, night scenes, still life, etc. We do not need images specifically staged to illustrate a particular topic unless these are subjects we normally capture.

The most important consideration in your example is relevance. An image of the Crab Nebula may be interesting and beautiful but it is probably not relevant.

Reply
Apr 26, 2016 06:50:44   #
gwong1 Loc: Tampa, FL
 
Just want to follow this thread.

Reply
Apr 26, 2016 06:54:05   #
Gitchigumi Loc: Wake Forest, NC
 
I don't have an opinion on ETTR, but I am interested in the discussion. Thanks for raising the topic.

Reply
 
 
Apr 26, 2016 07:08:32   #
dochal
 
What is Ettr

Reply
Apr 26, 2016 07:25:20   #
rmalarz Loc: Tempe, Arizona
 
dochal wrote:
What is Ettr


dochal, it stands for Expose To The Right. You'll also see the term EBTR, Expose Beyond The Right.
--Bob

Reply
Apr 26, 2016 07:29:14   #
selmslie Loc: Fernandina Beach, FL, USA
 
selmslie wrote:
... I am inviting proponents of ETTR to answer these questions and provide full images as evidence. ...

Better yet, how about a pair of images - with and without ETTR, complete with exposure and ISO information.

Reply
Apr 26, 2016 08:00:27   #
Bill_de Loc: US
 
I always wonder why digital seems to complicate simple, long standing situations.

When the range of brightness in a scene exceeds the dynamic range of film or sensor (HDR processing aside) the photographer needs to choose between exposing for the highlights or shadows.

With film you then compensate with developing times. With digital you compensate with software.

ETTR is in effect saying expose for the shadows because it is easier to correct over exposure than it is to correct under exposure.


It should not be some big mystery.

--

Reply
 
 
Apr 26, 2016 08:00:51   #
DaveO Loc: Northeast CT
 
gwong1 wrote:
Just want to follow this thread.


If you wish,you can click on "WATCH",which is right above the name of the OP.

Reply
Apr 26, 2016 08:04:10   #
dochal
 
Thanks

Reply
Apr 26, 2016 08:09:52   #
DaveO Loc: Northeast CT
 
dochal wrote:
Thanks


Conversely,you may have noticed that you can "UNWATCH" a thread to prevent notices of further activity. :)

Reply
Apr 26, 2016 08:25:02   #
Gene51 Loc: Yonkers, NY, now in LSD (LowerSlowerDelaware)
 
selmslie wrote:
The question of "why" and "when" to use ETTR usually gets lost in the shuffle of the discussions of "what" (science) and "how" (procedure). The last two have been covered quite thoroughly others and arguments over definitions and nuances in these areas can get silly. But the first two questions get little coverage.

Often the "why" is brushed off with a blanket claim about noise reduction, wanting to capture more detail or wanting to use the sensors full range. "When" is almost never mentioned. The problem is that why and when are the most important questions we need to be thinking about.

You cannot convince someone that ETTR will reduce noise if they use a modern camera like a D8xx, D7xx or D6xx at low ISO and they can't see any noise in their own images from a normal viewing distance, or even much closer than normal viewing distance. Noise reduction is not a sufficient reason to use ETTR. The solution to noise is a high S/N and we all know how to get that - low ISO. One or two stops of additional exposure with ETTR only begins to reduce noise.

Neither are you going to persuade anyone to use ETTR to get more "detail" if their concept of detail is resolution of sharp edges and something they can actually see in their prints or images. There is no expectation of seeing detail based on its conventional meaning in out of focus areas or in Zone II - "Textured black; the darkest part of the image in which slight detail is recorded" or in Zone I or 0.

The detail that is being referred to in relation to ETTR are the fine separations between tonal levels - the number of distinct levels that drop by half in the raw file with each reduction of 1 stop of exposure. This is "loss of detail" has been discussed ad nauseam but I have yet to see it demonstrated in a real image. Even from the ETTR viewpoint, "detail" has little relevance in very dark areas, the same areas where the raw file appears to be deficient.

It is also taken for granted that we should want details and fine tonal distinctions in the shadows. Not always, maybe rarely.

The sensor's physical range (the point at which the buckets fill to overflowing) and the raw file's numeric range (the point where all of the bits are 1) are not the same thing. Other than at base ISO both are seldom fully used. ETTR tries to use up the bits without going too far. This is often glossed over because it is harder to explain.

What is almost never mentioned is the dynamic range of the scene. There are plenty of cases where the range of luminance values reaching the camera is narrow and others where it is quite wide. With a narrow range there is no benefit to be derived from ETTR, a normal exposure is adequate. With a very wide range, even ETTR cannot help, only HDR. So there is only a particular subset of cases where ETTR can actually provide some help.

To justify the effort and inconvenience of using of ETTR we have to be convinced that there are real visible benefits. It's the "visible" part that needs to be addressed.

I am inviting proponents of ETTR to answer these questions and provide full images as evidence.

We do not need PowerPoint slides, charts, graphs, tables or screen shots of edit sessions. We don't need 100% crops from an image unless you also provide the complete image. And the images in question should be scenes that we face in our normal lives like landscapes, street scenes, night scenes, still life, etc. We do not need images specifically staged to illustrate a particular topic unless these are subjects we normally capture.

The most important consideration in your example is relevance. An image of the Crab Nebula may be interesting and beautiful but it is probably not relevant.
The question of "why" and "when&quo... (show quote)


Here is your relevance:

http://photographylife.com/exposing-to-the-right-explained

And in the majority of my images found here:

http://www.flickr.com/photos/gene_lugo/

I don't have comparisons because I don't waste time on proving the obvious to myself or others. I use ETTR 100% of the time to get correct exposures. Since I shoot manual exposure, there are times when I shoot a fast moving subject in changing light and make a mistake on exposure because I didn't get a chance to read and adjust in time to get the exposure I needed.

Having shot film for many years, and in particular, cut sheet large format, the opposite - expose to the left, was the rule - to minimize the chance of missing shadow detail.

In digital and with reversal film, there are benefits to exposing to the right. In cases of average or below average contrast, you record more information over the noise level, so shadows are less noisy. In this case you generally lower your exposure in post processing.

When you have high contrast situations, it is the only way you can avoid clipping important highlights (not specular highlights, streetlights in night scenes, etc) - the shadows will be murky and will take more effort in post processing to reduce noise, improve contrast and detail, but the result is usually worth the effort. In high contrast lighting I find that I often need to reduce both the highlights and exposure, and lighten the shadows.

ETTR along with the zone system, are two techniques that are worth mastering. It leaves less to chance.

Reply
 
 
Apr 26, 2016 08:53:25   #
selmslie Loc: Fernandina Beach, FL, USA
 
Gene51 wrote:
Here is your relevance: ....

Sorry, but that's not what I am asking for. We do not need more explanations of what and how.

But open the first link and scroll down to the section labeled 9) When to Avoid ETTR and you will find, "If you aren’t shooting at your camera’s base ISO, ETTR is all but useless." And later, "Similarly, ETTR is useless if it requires that you leave base ISO."

That's the author's answer to "when". Thanks for pointing that out.

Gene51 wrote:
... I don't have comparisons because I don't waste time on proving the obvious to myself or others. I use ETTR 100% of the time to get correct exposures. ...

That's the entire point of this thread. The benefits of ETTR are not obvious to anyone who is curious. The cost are easily apparent.

Your response does not illustrate why or when (other than always) you use ETTR so it does not contribute to this discussion.

Reply
Apr 26, 2016 09:00:41   #
brucewells Loc: Central Kentucky
 
Despite having seen it mentioned in print, I never paid much attention to ETTR until a couple weeks ago. It seems to be the "hot topic" on UHH right now, so I've been experimenting some.

I understand the concept. I'm just not sure the best way to get there. I typically shoot in aperture priority, so can I not use EC to ETTR? If so, on a 'sunny 16' scene, how far to the right would one expose to reap the benefits of the concept?

As in most things photographic, I'm not on one side or the other of this fence. I believe any of us can use these tools to create the image we visualize. I'm just not sure what/how to visualize what the ETTR proponents are after. I'm hopeful this thread will elaborate on that to some degree.

Thanks.

Reply
Apr 26, 2016 09:10:26   #
selmslie Loc: Fernandina Beach, FL, USA
 
brucewells wrote:
... I understand the concept. I'm just not sure the best way to get there. I typically shoot in aperture priority, so can I not use EC to ETTR? ...

Yes, you can use it with aperture priority. All you need to do is use exposure compensation which will slow the shutter down and add exposure.

But how about this for a novel approach? Set the exposure manually and use Auto ISO. Exposure compensation will increase the ISO and get you to the same place (no doubt arguments to the contrary are coming).

Nevertheless, you might want to first wait for the evidence that I hope ETTR fanatics will provide in response to this thread, if they are willing and able.

Reply
Apr 26, 2016 09:18:46   #
rmalarz Loc: Tempe, Arizona
 
selmslie wrote:
The question of "why" and "when" to use ETTR usually gets lost in the shuffle of the discussions of "what" (science) and "how" (procedure). The last two have been covered quite thoroughly others and arguments over definitions and nuances in these areas can get silly. But the first two questions get little coverage.

Often the "why" is brushed off with a blanket claim about noise reduction, wanting to capture more detail or wanting to use the sensors full range. "When" is almost never mentioned. The problem is that why and when are the most important questions we need to be thinking about.

You cannot convince someone that ETTR will reduce noise if they use a modern camera like a D8xx, D7xx or D6xx at low ISO and they can't see any noise in their own images from a normal viewing distance, or even much closer than normal viewing distance. Noise reduction is not a sufficient reason to use ETTR. The solution to noise is a high S/N and we all know how to get that - low ISO. One or two stops of additional exposure with ETTR only begins to reduce noise.

Neither are you going to persuade anyone to use ETTR to get more "detail" if their concept of detail is resolution of sharp edges and something they can actually see in their prints or images. There is no expectation of seeing detail based on its conventional meaning in out of focus areas or in Zone II - "Textured black; the darkest part of the image in which slight detail is recorded" or in Zone I or 0.

The detail that is being referred to in relation to ETTR are the fine separations between tonal levels - the number of distinct levels that drop by half in the raw file with each reduction of 1 stop of exposure. This is "loss of detail" has been discussed ad nauseam but I have yet to see it demonstrated in a real image. Even from the ETTR viewpoint, "detail" has little relevance in very dark areas, the same areas where the raw file appears to be deficient.

It is also taken for granted that we should want details and fine tonal distinctions in the shadows. Not always, maybe rarely.

The sensor's physical range (the point at which the buckets fill to overflowing) and the raw file's numeric range (the point where all of the bits are 1) are not the same thing. Other than at base ISO both are seldom fully used. ETTR tries to use up the bits without going too far. This is often glossed over because it is harder to explain.

What is almost never mentioned is the dynamic range of the scene. There are plenty of cases where the range of luminance values reaching the camera is narrow and others where it is quite wide. With a narrow range there is no benefit to be derived from ETTR, a normal exposure is adequate. With a very wide range, even ETTR cannot help, only HDR. So there is only a particular subset of cases where ETTR can actually provide some help.

To justify the effort and inconvenience of using of ETTR we have to be convinced that there are real visible benefits. It's the "visible" part that needs to be addressed.

I am inviting proponents of ETTR to answer these questions and provide full images as evidence.

We do not need PowerPoint slides, charts, graphs, tables or screen shots of edit sessions. We don't need 100% crops from an image unless you also provide the complete image. And the images in question should be scenes that we face in our normal lives like landscapes, street scenes, night scenes, still life, etc. We do not need images specifically staged to illustrate a particular topic unless these are subjects we normally capture.

The most important consideration in your example is relevance. An image of the Crab Nebula may be interesting and beautiful but it is probably not relevant.
The question of "why" and "when&quo... (show quote)


Scotty, If it is a new thread to discuss ETTR/EBTR, fine. We have another thread. But, if this is an attempt to reply to my post/invitation in an earlier thread, "Scotty, it would be interesting to have you write an article on the appropriate times to use ETTR and how to obtain the same results by lowering the ISO." it falls short.

If it is a response to that earlier thread, then the onus is on you to provide examples to back up your expressed opinion. It's not for the rest of us to provide photographs, which, I may add, have been done numerous times in other presentations.
--Bob

Reply
Page 1 of 21 next> last>>
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.
Main Photography Discussion
UglyHedgehog.com - Forum
Copyright 2011-2024 Ugly Hedgehog, Inc.