[PuxaVida]

Anyone have read this article in LL:



[url="http://www.luminous-landscape.com/tutorials/optimizing_exposure.shtml"]http://www.luminous-landscape.com/tutorials/optimizing_exposure.shtml[/url]



Quite interesting, but I guess the real life experiences with various raw converters are more valuable.



Serkan

[boren]

This is a well know technique, but for high contrast scenes (such as nightscapes) it's not a very good idea. Here's a very interesting article about the subject:



http://theonlinephotographer.typepad.com...-bull.html



And here's another article which rebuts the idea:



http://chromasoft.blogspot.com/2009/09/w...wrong.html

[bryan conner]

Boren is correct. In fact, ETTR is intended to be used when the dynamic range of the scene is less than the capability of the sensor. If your image produces a histogram that has a sizable gap both right and left, the idea is to increase the exposure to place as much info in the right side of the histogram (without clipping important highlights). By doing this, you are increasing the signal to noise ratio when possible.



The links above bring up the positive as well as the negative aspects of implementing ETTR. In the end, you have to experiment for yourself and make your own determination as to whether it is a useful method to implement into your photography life.



It has it's place in mine, but not in all situations.

[PuxaVida]

[quote name='Bryan Conner' timestamp='1320301292' post='12637']

...



The links above bring up the positive as well as the negative aspects of implementing ETTR. In the end, you have to experiment for yourself and make your own determination as to whether it is a useful method to implement into your photography life.



It has it's place in mine, but not in all situations.

[/quote]



Thanks for the input Bryan...



In the past (before knowing about the ETTR) I had made some experiments by myself and I can confirm that higher ISO settings produce better results than lower ISOs with same aperture & shutter speed, then exposure push-up in ACR. Actually if I remember correctly I had used ISO 200 and ISO 800 to compare results, and it was obvious... Infact, the article in ChromaSoft tells the same as one of the aspects. However there are some other arguements there regarding the tone curves applied before or after the raw converter loads the image (and manipulates the exposure). Actually such kind of inputs based on experiences are what I'm looking for. The author of the article tells that various raw converters behave differently, and NX2 for instance, does a real trick for nef files resulting from active d-lighting usage.

[bryan conner]

[quote name='PuxaVida' timestamp='1320306031' post='12638']

Thanks for the input Bryan...



In the past (before knowing about the ETTR) I had made some experiments by myself and I can confirm that higher ISO settings produce better results than lower ISOs with same aperture & shutter speed, then exposure push-up in ACR. Actually if I remember correctly I had used ISO 200 and ISO 800 to compare results, and it was obvious... Infact, the article in ChromaSoft tells the same as one of the aspects. However there are some other arguements there regarding the tone curves applied before or after the raw converter loads the image (and manipulates the exposure). Actually such kind of inputs based on experiences are what I'm looking for. The author of the article tells that various raw converters behave differently, and NX2 for instance, does a real trick for nef files resulting from active d-lighting usage.

[/quote]



I use Capture One for almost all of my raw processing. I do not see the color shift (or Hue Twist) when utilize ETTR. If it is there, I do not notice it.

[Brightcolours]

Exposure to the right is a reaction to the strange exposure to the LEFT, the very odd idea to avoid "clipped highlights"at all cost, resulting in under exposing images.

The idea behind it is that:

1. RAW has more room to recover "highlights" as shown in histograms than people realize, and

2. The information density at the left is MUCH lower than the information density at the right.



With 2, when you do adjustments ETTL can more easily result in banding/posterization (banding in the right sense of the term, nothing to do with noise patterns).



If you use a RAW converter which shifts the hue to a noticeable extent when adjusting "exposure", it will show with ETTL too. Of course, a correct exposure is preferable to ETTL or ETTR.



What I find a bit silly, is to bring ISO up like done in the articles. When one would ETTR, it is not about noise levels, but rather because one does not know the correct exposure and then the information density on the right has the preference. Not because one is struggling to get a suitable exposure time.



The article "against" ETTR with that night shot is silly. As if ETTR would mean, shoot the night scene as if it was a well lit scene, then bring it back. Ridiculous.



Another point: If one notices a hue shift, it will be shift with all hues... Not hard to shift it back in PP.



Anyway, ETTR was a reaction to the mantra of ETTL which was evangelized for years on at least internet fora, to protect highlights at all cost. And ETTR has PP advantages. But when possible, I think it is best to just nail the exposure. Which then means exposing on mid tones, not histogram.

[bryan conner]

[quote name='Brightcolours' timestamp='1320314458' post='12640']

But when possible, I think it is best to just nail the exposure. Which then means exposing on mid tones, not histogram.

[/quote]



I agree. But, if you have a low contrast scene, utilizing ETTR would place the mid tones, and the other tones in a range that would result in a higher signal to noise ratio...therefore lower overall noise in the image. Why would this not be better than exposing to place the midtones in the middle of the histogram?



I guess the underlying question that I have is: How do you define "nail the exposure"?

[PuxaVida]

I mostly agree with BC... I think the ETTR can be surely applied depending on the particular cases. During "zone"ing or exposing on midtones of the scene for instance, it is sometimes good to know that we have more headroom for the highlights and select the exposure point in the scene accordingly (without clipping). Experimenting the SNR performance with ETTR is fun, but does not sound practical in terms of field usage. Actually, if the scene allows, I would rather change shutter speed instead of playing with the ISO setting...



What you say Bryan about shifting the midtones to the right sounds reasonable at first sight (more headroom for highlights in raw data). But as stated in the ChromaSoft article, I also find it quite important when it comes to the impact of the tone curve applied by the raw converter... BC, any experience with different raw converters? Is it possible to avoid the interference of the applied tone curve during PP to some extent?



Serkan

[popo]

What we have here is probably yet another case of trying to find one rule that works in all situations. Just expose for what you want.



In my case, with wildlife there are often white or light detailed coloured parts amongst darker regions. The default exposure will tend to clip those areas more often than not, so I tend to shoot with various degrees of negative compensation. A better way to describe it might be to Expose to Retain Wanted Details. It could be left or right and doesn't really matter. If I don't have time to dial in near-perfect settings, underexposing always gives more scope for recovery than overexposing the same amount.

[bryan conner]

[quote name='popo' timestamp='1320324912' post='12645']

What we have here is probably yet another case of trying to find one rule that works in all situations. Just expose for what you want.

[/quote]



Exactly. I am not saying that ETTR is the end all to exposure/noise problems. My position is that it has it's place in my knowledge bank. I have experimented with it, and I do see some definite advantages in some situations...not all situations, but some.



In the end of my experimentation, I arrived at the following conclusion: If the scene contains less dynamic range than the capability of my camera, then I will take at least one shot with the image exposed to the right as far as possible without clipping important highlights.



Of course, there are an infinite number of variables present in every shooting situation. The photographer has to make his/her own decision concerning each of these variables in order to capture the image that he/she wishes to do. Sometimes ETTR can do this. Other times ETTR would make the sought after image impossible to capture.