It is easier to recover shadows than it is to recover highlights.
There is truth in what you say but there is also the converse argument of shooting to the right: see https://en.m.wikipedia.org/wiki/Exposing_to_the_right
Personally, for general photography I’d take the middle ground and opt for a balanced exposure, using exposure compensation to apply either of the techniques mentioned according to conditions and preferences for the subject being taken.
Shooting raw format will anyway give a 2 to 3 stop adjustment range in post production.
It would be interesting though to see how others approach the subject of exposure.
Nice subject you introduced here!
For me it depends on the camera as well… With Canon, I prefer to expose into the light, as if not it introduces too much noise. For my Fuji xt2, I prefer a bit to the shadows, as noise managmente for my taste is quite good… In general, I prefer kind of middle, but with Canon Cameras, 1.3 to the light always… Let’s see what the rest prefer!
Nice to see other discussions around!
Daniela from Uruguay.
The general idea is that it is easier to make darks brighter than it is to make brights brighter!!!
Using a raw format , I usually expose 0.3 to 0.7 EV to the left or under exposure. I usually find that highlights, mainly in reflections or the sky are harder to bring back so underexposure helps. There is a reason for this, which is, camera metering is mostly an average of the total scene. If the subject area is mostly shadows and mid tones the exposure will increase and not account for the small bright areas of clouds (for example). Of course, if the scene has no bright spots like a flower close up or indoor portraits then set the exposure compensation to zero.
Since scenes tend to vary, I like to keep exposure comp on 0 and then over or under expose as I view the scene. When in doubt…bracket. and re-evaluate later.
Xiao Lin, I tend to do the same, I prefer any bracketting, before under or over exposing.
As the lawyers love to say: it depends
The trick with ETTR (Expose to the right) is that you want to go as far to the right as you can without blowing the highlights. Once you have blown highlights, you cannot recover them.
So, if you are shooting from a tripod in relatively set conditions, it is not that hard to do this, if you have tested your camera system.
But if you are not in fixed conditions and things are changing dynamically (you are moving, your subject is moving, lighting conditions are changing, etc), it is pretty hard to reliably ETTR without risking blowing highlights.
So the advice that it is easier to recover shadows than highlights is spot on. If you can be certain that you are not losing highlights, you can ETTR. But if you are not sure, it is safer to make sure you are not overexposing.
I think the concept of compression is a better way to understand the ideas surrounding ETTR/ETTL and shadow/highlight recovery. Consider extreme examples of compression are to overexpose or underexpose so much that your resulting image is completely white or black respectively. Another extreme example would be to reduce contrast until your image is a single grey value, or increase it until you get an image with a mix of white and black pixels. What all 4 of these extreme cases have in common is the lack of flexibility they offer in post processing and printing/display. The key point of this thought experiment is: increasing compression reduces choices/flexibility; decreasing compression expands choices/flexibility.
So if you photograph a scene multiple times with increasing exposure, you’ll be compressing highlights and decompressing shadows. That is ETTR. Doing the same with decreasing exposure will cause compression of shadows and decompression of highlights. That is ETTL.
Also remember that the scene you’re photographing may already have highly compressed highlights or shadows that you have minimal control over and thus minimal flexibility.
What’s best then? Best is to understand the scene you’re photographing in the context of which portions of value you’re interested in - such as highlights, mid tones and/or shadows - and how to control compression/decompression to maximize flexibility in those values. That means, exposing to the right may be the best choice in some case and not others. Or maybe you need brackets and HDR.
In this way of thinking,
I’d say recovery is really about flexibility, and flexibility arises out of your use of compression. What ever you compress will be more difficult to recover.
As practical examples, consider two scenes: street level in the city shrouded in cast shadows with a bit of bright sky and clouds at the top, and landscape with the horizon very low and a sky filled clouds and broad, delicate highlights.
You might consider exposing to the right for the first scene so the shadows span a larger portion of the histogram with highlights compressed on the right. Such an exposure would contain many more shadow values to control and process. The highlights would be more difficult, but they’re not what’s important in the image. Likewise, in the second scene, you could expose to the left to decompress the highlights (like the post title says).
This line of thinking also agrees with the basic premise to expose for what’s important and let the rest fall where they may. Ultimately, you probably can’t have it all; something has to be sacrificed, even with HDR stuff.