Expose to the right (ETTR) or underexpose?

[kdriceman]

Advertisement (gone after registration)

It is highly dependent on the raw converter that you use. From what I've seen, LR doesn't quite measure up to boosting ISO in camera, but RPP does.

 

The two images are quite comparable using LR, but I have no doubt that other raw convertors could do the job better. I've tried to use RPP, but haven't worked my way up the learning curve. It is hard enough to keep me in front of the computer screen long enough to process my photos with what I know (LR/PS), so I haven't been motivated to dedicate the time to RPP. Maybe sometime.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[pgk]

FWIW this info might be of interest. If I use ETTR in very low contrast conditions (in which that the centered data reaches to neither end of the histogram) I find it more difficult to produce as good a result than if I expose so that the data is actually centered. I've found this out by empirically trying different exposures and if I use ETTR I struggle to deal with slightly adrift tonality in the mid-tones. I achieve a more effective result (with far less post processing) with data centered on the histogram. Has anyone else found this to be so (and also FWIW this is less so on Leicas than my Canons!)?

[Muizen]

(in which that the centered data reaches to neither end of the histogram) I find it more difficult to produce as good a result than if I expose so that the data is actually centered.

 

Paul, I like to understand the difference between:

"the centered data reaches to neither end of the histogram" and "the data is actually centered"

When centered can't it be that the data also do not reach either end of the histogram?

Harry

[adrianw]

I am with denoir all the way. I use mine for landscapes. Dark shadows are easy to retrieve in LR4 while highlights, even if not really blown out, are much harder to do anything with. While I am at it, that is also my experience with the 5DII. The M9 is so much better than the 5DII, I now rarely use the latter. The weight difference is also profoundly in favour of the M9 and all I need is a vintage Summicron 50. Underexpose.

[Tonki-M]

sunny 16!

 

really though ETTR w/o clipping highlights. may have preferred underexposed before, but with LR4, and how it handles highlights when we increase exposure, i now prefer ETTR.

[pgk]

Paul, I like to understand the difference between:

"the centered data reaches to neither end of the histogram" and "the data is actually centered"

When centered can't it be that the data also do not reach either end of the histogram?

Harry

Yes, but with ETTR the upper limit of the data would be at the right of the histogram, leaving a large gap at the left of the histogram.

[1JB]

Advertisement (gone after registration)

Personally I found spending too much thought about exposing to the right, left, and center to be a trap if I let it dominate my photography. It's a technical aspect of digital files and exposure that is easily understood but of little practical benefit to me. Rather than trying to make each photo what I wanted it to be when I shot it, exposing for a face and just let the highlights be blown out if they had to be, I found myself trying to capture a shot technically. Get it as close as possible to being able to bring back the highlights and pull up the dark areas to get a result where everything was visible after post processing. I ended up with a lot of snapshots and postcards that left me uninspired and often with large prints to create some sort of impact. I just wasn't capturing photographs that inpsired me or anyone else. I can see them at every festival with vendors and I don't want to buy them nor sell them.

 

I learned a great deal from the technique and still rely on it for pp particularly for landscapes. The M9 files are very resilient with regard to what they can take and deliver in pp. But I had to go through a black and white phase to appreciate dark shadows, white skies and contrast that stirred some emotion. For me the two had to be a balance. If I wanted every shot to have no dark eye sockets, blue skies and pretty green trees I would bring a flash everywhere. My wife will take perfect shots with a little pocket camera and pop flash that detects faces and auto exposes. It even shoots hd movies.

 

For me ettr was a good learning experience but not an end in and of itself. My best shots are exposed how I want them when I take them with a bit of margin for error from what I learned from the technique. They tell a story, raise questions or define a subject and moment. Almost all of them have some aspect of the exposure a miss. A face a bit too dark, an area blown out or something almost completely lost in shadow but just there to be explored.

Edited October 16, 2012 by 1JB

[douglasf13]

Well, the beauty of the M9 is that it is essentially an ISO-less camera, so there is a lot of margin for error, as long as you don't blow highlights. You could feasibly shoot everything at base ISO and boost exposure in the raw converter. Anyways, ETTR really only applies to base ISO in the first place.

[Jaybob]

Well, the beauty of the M9 is that it is essentially an ISO-less camera, so there is a lot of margin for error, as long as you don't blow highlights. You could feasibly shoot everything at base ISO and boost exposure in the raw converter. Anyways, ETTR really only applies to base ISO in the first place.

 

This statement, from beginning to end, is inaccurate, except for "don't blow highlights".

 

Regardless of camera model, sensor type, film stock, OR ISO, "proper" exposure will always yield a better technical photograph. The BEST way to accomplish proper exposure in digital photography is to utilize ETTR. The only real "correctable" mistake you can make shooting RAW is in white balance, exposure errors will almost always compromise the image quality, unless of course, that's what you're looking for.

 

Jay

Edited October 16, 2012 by Jaybob

[haroldp]

In high contrast situations, with modern sensors, I find shadows to be easier to deal with than blown highlights, and the result of boosting shadows is much better than trying to tame highlights that are either blown or up to the last bit.

 

For myself, In high contrast situations, I prefer to leave headroom fort the highlight, which ETTR does not.

 

Others should do what works for them.

 

.... H

[Jaybob]

ETTR is expose toward the right, not THROUGH the right. The reason I use ETTR for digital is because it works, regardless of model.

 

That's an OPTIC white jacket with detail to the thread and no blocked shadows. It's probably about as high contrast as you get.

 

LR4 develop module with the reset button pressed (no correction applied).

 

Summicron Rigid on a NEX 5 ARW.

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Edited October 16, 2012 by Jaybob

[pgk]

The BEST way to accomplish proper exposure in digital photography is to utilize ETTR.

If you read my post above you might find that whilst I do not disagree entirely, exposing to the right can in certain circumstances, in my experience, produce files which require extensive post processing to produce images with acceptable tonality. The increased discernible 'quality' improvement by using ETTR in such circumstances (as extremely low contrast) is IMHO so marginal as to negate the use of ETTR. This is from my own empirically derived data based on actual real usage testing. All that said, I find the Leica Ms to be far more tolerant than my dSLRs in this respect and I continue to ETTR with them.

 

If I may say so, the problem with many sweeping statements or accepted facts is that it all too often they need caveats, and so should not be accepted in their entirety without some degree of checking when they are applied and results are not quite as anticipated. Just my 2 pence worth.

[Jaybob]

If I may say so, the problem with many sweeping statements or accepted facts is that it all too often they need caveats, and so should not be accepted in their entirety without some degree of checking when they are applied and results are not quite as anticipated. Just my 2 pence worth.

 

As I showed in my example, If presented with a full range of values from really dark to really bright..and given the choice "Expose to the right (ETTR) or underexpose?" I would choose ETTR 100 times out of 100, attempting above all else to keep highlights safe. Regardless of ISO, an ETTR, non blownout highlight image, Whites, Highlights, and Exposure will always be easier to adjust than the exposure, shadows, and blacks of an ETTL or underexposed image.

 

If it's a low contrast scene, (or a lower contrast lens, like my 90 Elmarit) then the values of the histogram should immediately reflect that, and it's really unimportant where your curve is on the graph, as long as you're not jammed up to or off the edge on either side. In these low contrast situations, I prefer my histogram to be in the middle or higher, but that's just me. It's up to you how you adjust that exposure to a higher level of contrast in post processing or not.

 

Blowing highlights or blocking shadows and fixing it later is not for me.

 

Does that work out for me everytime? No. Am I the all knowing, all seeing, king of all photography? No. Sorry for the sweeping generalizations...typing often makes me sound like a pompous douchebag.

[mjh]

If presented with a full range of values from really dark to really bright..and given the choice "Expose to the right (ETTR) or underexpose?"

“ETTR or underexpose?” is the wrong (or misleading) question to ask. ETTR means to expose so that the brightest highlights worth preserving sit flush with the right edge of the histogram. If the contrast of the scene is lower than the sensor’s dynamic range, this amounts to overexposure, but it amounts to underexposure when contrast is greater than dynamic range. So the answer to “ETTR or underexpose?” may very well be “Both.”

[Jeff S]

Here is a case generally against ETTR. (Sorry if this has already been cited.) Ctein is a regular contributor to the TOP site, and known to have strong opinions.

 

This is not to say that I agree; rather, to generate further discussion.

 

Jeff

[pgk]

If the contrast of the scene is lower than the sensor’s dynamic range, this amounts to overexposure.....

But why? Doing so should allow more data to be captured shouldn't it (ie the darkest areas should have more data than if they were captured at the left of the histogram)? My suspicion is that capturing by ETTR in low contrast conditions leads of abnormal tonality without considerable readjustment in post. So I find it is actually more (time) effective to capture with a centered histogram - but I haven't convinced myself yet that ETTR won't capture more data.

[mjh]

But why? Doing so should allow more data to be captured shouldn't it (ie the darkest areas should have more data than if they were captured at the left of the histogram)?

It’s the other way round: There is more data captured the farther to right it is captured – much more actually. So if you have a choice you want to move the tonal values as far to the right as possible without blowing the highlights.

 

If a sensor resolves, say, 16384 values (14 bits) and you divide the histogram into 14 zones of one EV each, there are 8192 different values resolved within the rightmost zone, 4096 in the next, then 2048, 1024, 512, and so on.

 

My suspicion is that capturing by ETTR in low contrast conditions leads of abnormal tonality without considerable readjustment in post.

Nothing abnormal about it. It’s just a high-key picture until you adjust the black point to your liking.

Edited October 17, 2012 by mjh

[giordano]

Here is a case generally against ETTR. (Sorry if this has already been cited.) Ctein is a regular contributor to the TOP site, and known to have strong opinions.

 

I wonder if Ctein was just trying to be controversial, or whether he's damning the term "ETTR" in the hope of persuading unskilled photographers not to E too far TTR. In practice, his advice boils down to exactly what ETTR has always meant to the discerning, namely "give as much exposure as you can without blowing the brightest highlights that matter". The important bit is where he points out that the histogram and LCD don't always give a reliable indication of whether and exactly where the highlights are being clipped.

[Advertisement]

Hallo,

Look here: Leica M

[01af]

If the contrast of the scene is lower than the sensor’s dynamic range, this amounts to overexposure ...

But why?

Because the mid-tones get moved away from the histogram's center. The overall impression of a capture being underexposed, overexposed, or properly exposed depends mostly on the mid-tones. If they are where they belong then we consider the exposure as being correct.

 

Still, proper exposure in a technical sense means that a feasible post-processing procedure exists that leads to a visually convincing picture without significant loss of image quality. So technically good exposure and visually good exposure may differ. The digital file coming out of the camera should be exposed technically right. The final picture for the beholder to look at should be exposed visually right. Between the two, there may be some considerable amount of post-processing.

 

 

... capturing by ETTR in low contrast conditions leads of abnormal tonality without considerable re-adjustment in post.

That's right.

 

 

... but I haven't convinced myself yet that ETTR won't capture more data.

ETTR in low-contrast situations will capture more data indeed—but a centered histogram will capture better data because a digital RGB sensor's colour gamut is wider for mid-tones than it is for highlights or shadows. So better don't ETTR for low-contrast scenes ... umm, except when you're using the M Monochrom

 

In fact, ETTR makes sense only for high-contrast scenes when a regular exposure, i. e. a centered histogram, means blowing some highlights and blocking some shadows. Since blocked shadows usually are less objectionable to the eye than blown highlights, it makes sense to reduce exposure in order to save the highlights from blowing, at the expense of more blocked shadows. Usually. Exceptions to the rule may happen; it depends on the subject and the photographer's intentions.

[Jeff S]

I wonder if Ctein was just trying to be controversial, or whether he's damning the term "ETTR" in the hope of persuading unskilled photographers not to E too far TTR.

 

He's usually controversial, and his ego precedes his strong views, from which he rarely wavers.

 

But, you're generally correct if you read his comments back to multiple people, about halfway down. There he says he doesn't dispute the theory of ETTR; rather he thinks (my paraphrasing) that the practice is dangerous as most people get it wrong, and the dangers of getting it wrong are far worse than erring in the shadows, which is now less an issue than years ago when the theory arose.

 

Jeff