ETTR, when and why

[pgk]

Advertisement (gone after registration)

4 hours ago, SrMi said:

I believe that ETTR, its application, and the reasons behind it are still misunderstood.

ETTR is a technique to be applied in low contrast images when the DR of the scene is smaller than the DR of the sensor. 

The basis of ETTR is to expose the sensor to more photons than with 'regular' exposure. More photons equal better SNR. Better SNR is preferable, regardless of sensor generation.

I believe that ETTR can be applied correctly without image deterioration.

It is not misunderstood. Been there tried it. It is hit and miss even with low contrast scenes. For a start you are missing out the in-camera processing applied to the data before the RAW file is written and its effect. This optimises the file but is not based on an ETTR exposure, and this in any case is tricky to determine because if you are aiming at maximum saturation of the sensor then any variation in contrast will have an effect on the ETTR exposure and subsequent processing of the file. I tried using ETTR but gave up due to inconsistency and the need for more post processing. The theory is ok but doesn't take all factors into account. If we had access to a 'real' RAW file it might be marginally better but we don't, so we start off with a file we don't full appreciate the nuances of, with variability of exposure and subsequent colour and tonal artefacts which have to be corrected. It can be applied but I see little actual advantage doing so and sometimes significant disadvantages. But its easy enough to try and determine for yourself. Just don't think its a 'solution' to a problem that many of us don't have.

[willeica]

I have been using digital cameras since 2004 and I have never consulted a histogram, even though I know it is there and what it shows. I am perfectly satisfied with the exposure of my photographs and so are others who have seen them. My main technique is to meter from the area where the light represents the middle of the range that I want and then compensate according to the conditions and what I want to achieve. Most times I do this quickly without thinking and I rarely have to bracket with digital. I am more inclined to bracket with film which I have returned to increasingly in recent years. I accept that others may find using a histogram useful, of course. My resistance may have something to do with my wanting to control the camera rather than the camera controlling me and my photographic decisions. Most required adjustments can be made in post processing and I can see the histogram changing, but that means nothing to me as I use the image that I can see on the screen and I also use selected area tools for adjustments. For pretty much the same reason I don't use presets and I adjust my usual flow slightly for each image. If I were a professional I would probably use presets in order to handle much increased volumes, of course. I am also very much used to using a camera with no meter for which I often use a hand held meter.  I have a friend (Paul pgk also knows him) who has been a professional photographer for 61 years. Recently I was using a camera without a meter and I asked him what exposure did he think I was going to use (having consulted a handheld meter myself) for a roll of 100 ISO black and white film I had in the camera. The number he came up with without consulting a meter was within half a stop of what I had set on the camera. He is widely admired around the world for his advertising photography and he is an absolute genius with studio lighting. In fact the well known photographer Terence Donovan declared him to be a genius many years ago. He switched some years ago to digital cameras for his renowned advertising work, but he never ever uses RAW. He shoots everything in JPEG and what he produces is as close to perfection as you will ever see. 

William

[Advertisement]

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

[BernardC]

22 hours ago, SrMi said:

Why would you have different exposures with ETTR and not with automatic exposure?

You also see a difference with regular AE, but cameras have multi-zone and/or intelligent metering, so the difference is less. An example of intelligent metering is one that uses face-detection data to favour skin tones.

I am thinking here of scenes with strong side- or back-lighting. What old-timers called "Rembrandt lighting." It's commonly used for portraits, but many photographers approach landscape and still life the same way. This technique is especially fruitful with Leica lenses, because they are exceptional at holding detail in flare-prone situations.

One thing about such lighting is that your highlights can change significantly with small changes in composition (or if your subject moves). If you use ETTR, you'll end-up with a proof sheet (physical or virtual) where every shot has a different exposure. That makes it harder to judge which shot has the best composition/expression/etc. Of course, if it's harder for a photographer with a trained eye, it's almost impossible for a non-photographer to separate exposure from other content. Their eyes will automatically be drawn to images that "pop," even though other images can be processed to look the same way.

Of course, what I describe is a limitation of the technique, not a show-stopper. It should go without saying that photographers who are drawn to ETTR, aren't afraid of extra post-processing work...

[jaapv]

On 8/29/2020 at 12:43 AM, willeica said:

I have been using digital cameras since 2004 and I have never consulted a histogram, even though I know it is there and what it shows. I am perfectly satisfied with the exposure of my photographs and so are others who have seen them. My main technique is to meter from the area where the light represents the middle of the range that I want and then compensate according to the conditions and what I want to achieve. Most times I do this quickly without thinking and I rarely have to bracket with digital. I am more inclined to bracket with film which I have returned to increasingly in recent years. I accept that others may find using a histogram useful, of course. My resistance may have something to do with my wanting to control the camera rather than the camera controlling me and my photographic decisions. Most required adjustments can be made in post processing and I can see the histogram changing, but that means nothing to me as I use the image that I can see on the screen and I also use selected area tools for adjustments. For pretty much the same reason I don't use presets and I adjust my usual flow slightly for each image. If I were a professional I would probably use presets in order to handle much increased volumes, of course. I am also very much used to using a camera with no meter for which I often use a hand held meter.  I have a friend (Paul pgk also knows him) who has been a professional photographer for 61 years. Recently I was using a camera without a meter and I asked him what exposure did he think I was going to use (having consulted a handheld meter myself) for a roll of 100 ISO black and white film I had in the camera. The number he came up with without consulting a meter was within half a stop of what I had set on the camera. He is widely admired around the world for his advertising photography and he is an absolute genius with studio lighting. In fact the well known photographer Terence Donovan declared him to be a genius many years ago. He switched some years ago to digital cameras for his renowned advertising work, but he never ever uses RAW. He shoots everything in JPEG and what he produces is as close to perfection as you will ever see. 

William

I find  this a bit hard to follow, William.  How can a camera  - or postprocessing program- control me through a histogram? It is just an informative readout of the exposure, nothing different from any kind of exposure meter, but providing more data for the photographer to use than just a needle and scale. It  is in no way related to presets.

[willeica]

9 hours ago, jaapv said:

I find  this a bit hard to follow, William.  How can a camera  - or postprocessing program- control me through a histogram? It is just an informative readout of the exposure, nothing different from any kind of exposure meter, but providing more data for the photographer to use than just a needle and scale. It  is in no way related to presets.

It could hardly control you, Jaap! This is completely a personal thing for me, perhaps a psychological thing on my part in which I believe that I am the one determining what the camera does as regards exposure. I fully understand what you are saying, but I don't use a histogram to determine exposure either in the camera or in post processing, nor do I use presets to do the same thing, nor do I feel the need to do any of those things. I judge images and alter them according to what I can see. Every image gets my own bespoke treatment according to what I can see in the image. Whether the histogram is to the left, to the right or in the centre is of no interest to me and does not influence my photography at the point of taking or in post processing. I hope that I have clarified the point. I also assume that using a histogram is not compulsory, as I have got on quite well for many years without using one. 

William

[jaapv]

But why do you use an exposure meter then? 
I never use presets either- rather dislike them except in very specific cases. 

[willeica]

Advertisement (gone after registration)

49 minutes ago, jaapv said:

But why do you use an exposure meter then? 
I never use presets either- rather dislike them except in very specific cases. 

I can actually guess exposure pretty well as I am used to using very old cameras with no meters, but I don't always trust my own judgement. Believe me when I say, as an experienced photographer, that I have never felt the need to use a histogram.

We are ad idem on the question of presets. 

William

[jaapv]

Different strokes for different folks. I find it a significant improvement in exposure precision. 

[pop]

There used to be light meters for use with film with explicit support for the zone system, which is quite similar to a histogram showing the distribution of light values within the frame.

Since digital sensors have so much less latitude than film, using a more accurate light meter is very useful for many people. The digital sensor clips all signals exceeding the maximum value it can render. Film does not do that. Hence, minding your highlights is more important when shooting on sensors instead of negative films. Hence, you adjust your exposure to the brightest parts of the image you want to record on the sensor, instead of the darkest part that you want to record on film. No arcane knowledge needed.

 

[Jeff S]

34 minutes ago, pop said:

There used to be light meters for use with film with explicit support for the zone system, which is quite similar to a histogram showing the distribution of light values within the frame.

Since digital sensors have so much less latitude than film, using a more accurate light meter is very useful for many people. The digital sensor clips all signals exceeding the maximum value it can render. Film does not do that. Hence, minding your highlights is more important when shooting on sensors instead of negative films. Hence, you adjust your exposure to the brightest parts of the image you want to record on the sensor, instead of the darkest part that you want to record on film. No arcane knowledge needed.

 

And of course those who have experience with slide film know the drill with sensors.

BTW, I still have my Pentax digital spot-meter with zone labels.  That was a big help for eliminating guesswork when shooting expensive 4x5 b&w film negs.
 

Jeff

[jaapv]

13 hours ago, pop said:

Since digital sensors have so much less latitude than film,

Although I agree with your conclusion, Philipp, this one is not tenable (any more). For instance Ilford FP4+ has a DR of 13 ( https://adrianbacon.com/simple-photography-services/simple-film-lab/films/125fp4/ ) which is rather the same as a modern sensor.

[pgk]

14 hours ago, pop said:

Since digital sensors have so much less latitude than film.....

I actually find that RAW files are quite flexible and have good latitude providing they are not overexposed much - another reason that ETTR is risky, get it wrong and highlights are blown and cannot be recovered. Slight underexposure mimics higher ISO except that it has to be adjusted for in post processing rather than by the camera (and yes I do know underexposure  reduces the DR that is recorded but often absolute shadow detail is less important than many seem to think).

[SrMi]

On 9/2/2020 at 3:02 AM, pgk said:

I actually find that RAW files are quite flexible and have good latitude providing they are not overexposed much - another reason that ETTR is risky, get it wrong and highlights are blown and cannot be recovered. Slight underexposure mimics higher ISO except that it has to be adjusted for in post processing rather than by the camera (and yes I do know underexposure  reduces the DR that is recorded but often absolute shadow detail is less important than many seem to think).

The only time that I have blown highlights is when I let the camera determine the exposure. Once I use histograms and blinkies, I am typically on the safe side. For high contrast situations (no ETTR), I usually underexpose or bracket. I typically overexpose (ETTR) for low contrast situation, but am far away from clipping highlights. I am not doing the optimal ETTR but the safe ETTR, which is better than what automatic exposure would do in low contrast situations.
The RAWs are quite flexible, but that does not mean that we should not try to get the best possible exposure (with or without ETTR).

[pgk]

4 hours ago, SrMi said:

The RAWs are quite flexible, but that does not mean that we should not try to get the best possible exposure (with or without ETTR).

In film days we referred to this as 'latitude' and yes RAW files can have quite a lot although a 'good' or 'accurate' exposure is always the best option. This isn't always so easy to define as it depends on subject and the photographer's preferences which in turn depend on the desired end result.

[Stuart Richardson]

I am with pgk here. ETTR can work in certain circumstances, but more often than not, I have seen color shifts and otherwise undesirable effects to the tone curve. At base ISO with modern cameras, I do not think the benefit of slightly lower grain (of which there is often none to begin with these days) is worth the risk. Add to this the fact that every photo comes out of camera looking too hot, and you wind up with a lot of bother for minimal gain and a higher potential failure rate. If you nail the exposure, you nail the exposure...there are few photos that cannot be rendered nicely with a good exposure, especially with 10+ stops of DR. Look at how good slide film looks with 4 or 5 stops...

I will add the caveat, that if you are shooting high ISO, then giving it very generous exposure really helps. Then again, you often use high ISO when you don't have the light to begin with, so it is a tricky balance. With respect to all these "systems" (zone system, ETTR etc), a wise photographer once told me, "photography is about choice. Focus on the part of the image most needed to be in focus, expose for the most important tonality. If you cannot make the photo work doing that, it may not be a photo worth taking." I know that some might object to that, and I am not saying that it should be applied in all cases, but I think sometimes these systems wind up doing more harm than good....pursuing some technically perfect ideal at the expense of clarity of vision. If I aim to capture the beauty of the light in an landscape image, wouldn't it make more sense to photograph it as closely as possible to how it appeared, rather than to overexpose it and hope to remember it correctly whenever the times comes to edit it, possible days, months or years later?

[SrMi]

I am still surprised that some are talking about color shifts with ETTR. Is the color shift in highlights, midtones or shadows? I can imagine color shifts happening in highlights, but only because ETTR was mis-applied to a high-contrast scene and/or was overdone and some color channels got clipped (this can be verified with RawDigger histogram). 

Would love to understand the issue with color shift and ETTR, and see some examples. 

Edited September 21, 2020 by SrMi

[SrMi]

Found the following information about color shifts in this DPR article:

Some care may be required when making large adjustments to shadows and mid-tones. Nonlinear transformations in moving from the camera space to the working space, and twisted converter profiles (such as Adobe's Standard Profiles), can cause color shifts during large luminosity lifts. In many cases this is not a practical problem, but, if it is, finding the right raw converter and using a proper camera profile become important. These are not issues that are peculiar to ETTR

[adan]

On 8/28/2020 at 10:59 AM, SrMi said:

ETTR is a technique to be applied in low contrast images when the DR of the scene is smaller than the DR of the sensor. 

This is the key point.

For me, low-contrast scenes are things that mostly only happen to other people.

Between the electric contrast range of daytime pictures under Colorado's raw direct sunlight and thin air, and nightime or indoor pictures with bright artificial light sources (or windows) and pools of shadow within the same picture, and the wide tonal range of a dawn/dusk sky behind a dim foreground, the percentage of low-contrast situations I encounter can be counted on the fingers of one hand (<5%).

Add to that my subjects are generally fast-moving humans, where the entire taking of the picture (including metering) has to happen in a couple of seconds, and I "EFTR" (expose for the right - the highlights) most of the time. Default -0.7-stop autoexposure.

(Plus, I do get a little extra help from using the older, lower-contrast 1980s Mandler lenses whenever possible - at least an extra stop of DR between darkest shadows and blown highlights)

But with the M10 I do use a rough-and-ready version of ETTR with the rare low-contrast (in the shade) subjects I find interesting - a thumb-swipe of the exp. comp wheel to +0.3 to +0.7.

And I have even started using OETTR (overexpose to the right) when the setting and subject demand it. In the extreme-contrast-range picture below, the rich dark tones far outweighed the importance of the windows, and I was perfectly happy to blow the windows through the roof for the maximum in shadow gradation. Again, a quick thumb-swipe to +1.5 EV.

Great photography is not about following a rule. It is about choosing and adjusting the many rules - to suit the situation.

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!

 

[pop]

Am 28.8.2020 um 18:59 schrieb SrMi:

The basis of ETTR is to expose the sensor to more photons than with 'regular' exposure. More photons equal better SNR. Better SNR is preferable, regardless of sensor generation.

I don't think so. Mapping a low contrast scene to a medium capable of rendering a higher contrast is straightforward. Of course, you can improve the result by increasing the exposure if the recording medium performs better for higher levels. 

The issue becomes more relevant when the contrast of your scene exceeds the capability of your sensor. In this case it seems useful to remind photographers who are mostly accustomed to negatives on film that a digital sensor clips the highlights while film clips the shadows. Featureless white areas in an image are often a more serious defect than featureless black blobs. YMMV, of course.

[Advertisement]

Hallo,

Look here: Interesting Leica gear

[pgk]

7 hours ago, adan said:

(Plus, I do get a little extra help from using the older, lower-contrast 1980s Mandler lenses whenever possible - at least an extra stop of DR between darkest shadows and blown highlights)

Sorry Andy but you don't and we've had this argument before. Older Mandler lenses record less tonality not more. The reduced contrast is a result of veiling flare and the deepest shadows are not tonally separable. This is regardless of the scene's dynamic range. If you don't believe me just try a very old lens such as an early TTH Cooke. It cannot record the tonality that a modern lens can because it clearly lacks shadow detail which is masked by veiling flare - even on a triplet. Using it in high contrast conditions tends, if anything, to increase veiling flare and record less tonality.