Pushing the Dynamic Range Envelope

You can treat the shadows separately from the highlights by getting as much light onto the sensor for those shadow elements without blowing anything in and near them in the scene.

You can in this case let the highlights clip as much as you like because they will come from a different shot in the final blended HDR image.

A typical scenario could be when you are inside a dark room with windows to the outside you need to protect. Here I would denitely ETTR the inside of the room and let the windows clip.

You then adjust the shadows' lightness in post back to what you want before blending them into the final HDR image.

Therefore, I disagree that ETTR is redundant in making HDR images.

In the same way they can make HDR processing totally redundant as well if someone is happy to expose for the highlights, raise shadows in post to suit and then use any reputable noise reduction app to remove any visible noise.

There is no real need to fiddle with bracketed exposures, ETTR and then blending them into a single final image.

The technique someone chooses to handle HDR scenes simply comes down to style and preferences as described earlier.

I don't begrudge you a strict technical approach, such a dogmatic adherence to a process can be a useful way of abstracting an image. But +/-1 stop will not make any real visible difference in the amount of noise. And if you're setting a +3 stop on your bracketed set anyway, then how much are your exposures really going to differ? Not by any appreciable amount. Besides, in my brief foray I found flare to be more of the phyisical limit to over exposure than excessive noise in finished photos.

As I said, what technique someone chooses to handle HDR scenes comes down to styles and preferences.

Just do what suits you best.

With this scene I'm not sure bracketing images makes much difference.

Using the 1/125s shot and some quick and basic processing in ACR and Photoshop Elements you can come up with something very usable.

This is from an 8 bit jpeg. If the 16 bit raw file of the 1/125s shot was available it is highly likely you could come up with an even better final image by using the extra information from 16 bits.

With this scene you can set the largest exposure* within dof and blur constraints without clipping important elements in the sky and then raise the shadows and adjust the colours in post from the raw file to make a very good "documentary" image of the scene.

*exposure - amount of light striking the sensor per unit area during a shutter actuation.

And the same applies with the 1/50s shot in your other HDR sequence. You don't necessarily need HDR processing for these types of scenes.

The dynamic range is not so high that most cameras cannot handle it.

Even with the 1/50s jpeg you can get a very usable final image as shown below.

If the 16 bit raw file was available I'm sure an even better final image could be made.

In your HDR image the shadow circled in red does not match the "lighting" on the building at all. The sun is clearly in the distance in the direction of near the top left corner of the frame as indicated by the area circled in green.

The large building wall facing towards the right needs to be in shadow to a large extent to match the shadow circled in red.

That is not what ETTR is all about at all.

ETTR is all about maximising the quality of the raw data by maximising the Signal to Noise Ratio (SNR) in the raw data within dof and blur requirements without clipping important highlights. The higher the SNR the less noise will be visible.

After ETTR'ing a low light scene the image will most likely look way too light when the raw file is opened in a raw converter. You then adjust the image lightness in post down to what you saw at the time of the shot.

Because more light hit the sensor in the ETTR'ed shot the SNR will be higher than if you simply centered/zeroed the camera meter. The higher SNR results in less visible noise when you dial back the image lightness in post compared to the same scene shot with a centered meter.

This thread was just about a technique I found to be useful. The progress in picture processing software is enormous. The fact I can easily create hand held HDR sets, opens a lot of doors. Dealing with burnt out skies becomes easier to deal with. I can just set my Z7 to a short delay an it will fire off a 3 or 5 frame sequence with just one shutter actuation.

Hand holding in interiors can produce better images, as I can get detail into the window areas, that are usually burnt out, with a single frame, like the shot below.

Yes but it was side tracked to some extent when you posted your opinion that ETTR is redundant in HDR processing.

I posted why it isn't.

I have also shown that for these types of scenes you don't even really need HDR processing to get a good usable image.

Perhaps more suitable scenes would have been better to discuss what is a commonly known technique anyway.

Think you misread the photo, if we glance we tend to jump to assumptions that support our viewpoint more easily than we see where the photo doesn't.

The sun is over the photogs left shoulder and the shadow circled in red is the shadow of the building where the light hits it circled in green, and to me the front of the building looks to be in shadow, not sunlit. Processing choice, each to their own?

I disagree for the reasons posted earlier.

Have a look at the op's 3 photos in the op.

No-one is saying it is sunlit.

I said the lighting on the wall in the HDR image does not match the shadow circled red. In the HDR image the lighting on the wall is way too light in comparison to the shadow in red.

Even in the HDR bracket shot in the op (see below) shows the level of lighting on the wall and the shadow in red match much more closely than in the op's final HDR image.

And this is a quick and basic edit of the 1/50s shot in the op's HDR bracket.

For me, the lighting on the large wall and shadow in red are closely consistent with each other.

And in doing so you convert every image you see so it conforms with your view of how you think it should logically be, which being honest has a fair amount of confirmation bias (this is why even when looking at the same photo we do not see the same thing). You not only treat every image as though it's an absolute, which assumes our sight is absolute and so ignores the effects of being human, but you also refuse to leave others the room in the photo to see their view, their memory or confirmation bias. Perhaps missing the real import of your much used phrase?

But if it works for you all well and good.

Personally I find your edit cold and unreal, the shadow previously circled in red is far to light, too blue and too much contrast. The colours as a wjhole are far too blue and just don't work as an image. The building looked far better in NCV's edit as being more yellow which contrasted with the shadow on the rock far better. By enforcing your logic rather than your visual sense you've made it rather like an authoritarian taking over a friendly and very loose game of softball, I don't feel involved any more and it's no fun. - In a pictorial sense, not a personal one, to be clear what I'm saying here. 😋

That's nonsense because you like me and everyone else is free to view the photo as they like.

You are not agreeing with me any more than I am agreeing with you. That's fine. I have no issue with that.

That's fine. As I said, I was editing a jpeg and with a raw file a much better final image should be able to be made.

The point I was making is that you don't really even need HDR processing for scenes of this type because you can expose for the highlights, raise shadows and adjust colours in post from a 16 bit raw file and make a nice "documentary" version of the scene.

The building was in fact in the shadows, very deep shadow in fact.

This was an experiment, to explore the possibilities of HDR. I wanted the building to be readable with a mid tone level for the stonework, as if I had been using a huge fill in flash. Yes, I should have lifted the rock shadows too.

The picture very quickly falls apart and looked totally artificial when I tried editing some single exposure shots, I took before trying hand held HDR

I usually try to get the right time of day using an app that tels me when the sun is illuminating a building. Or at other times I want a grey day. But this is not always possible.

It shouldn't be too much of an issue for the middle exposure shots if you used the raw files.

For a scene like that you can make the building just about any colour you like with any lightness you want in post fairly easily as well for those that don't want to do HDR bracketing for whatever reason.