NIK in all its glory....incidentally easy controllable HDR

So what's the difference? You still have not told us what makes what you did HDR versus my brightness edit example. CAN you explain what HDR is?

I processed a single image in AuroraHD, which the program calls "HDR" but is in fact, nothing more than tone mapping. Adjusting brightness is a component of tone-mapping, but tone-mapping consists of adjusting many levels in all channels, plus mapping key areas in the image for separate processing. If you can't see the difference in my example plus the single-adjustment followup by @paddler4, then more words won't suffice. Is it time for a "graphic novel" approach?

"HDR", to my understanding, is the process of rendering bracketed images into a 32 bit/channel image file, so that the lightest and darkest tones fit inside the limits of 0 to 100%, with 2^32 steps in between. Tone-mapping is the usual finishing process making the HDR image suitable for display or printing. The tone-mapping process is more complicated than simply mapping the 32 bit channels to 16 bits (which changes the number of steps without changing the end points).

In summary, traditional HDR processing has two steps, the last being tone-mapping. "HDR" processing of a single image "HDR" has only the tone-mapping step, hence is not actually HDR, regardless of what the software calls it.

Glad to see that you agree with Rodney and me about this.

I thought it relevant to show that not all HDR/tone-mapping produces grunge results.

I spend an unfortunately large amount of time grading log2 and log3 video. This too a means to compress the huge dynamic range of a cinematic camera into a limits ultimately conformed to REC.709 (~ sRGB) or REC.2020 (~ Adobe RGB). Like tone-mapping, log gamma doesn't increase the dynamic range of capture, but makes that dynamic range fit into a smaller standard space as the work product.

My little P&S Sony RX100iv has HDR. I think it's just appling shadow slider control in camera.

So how does that make it HDR?

So how does that make it HDR?

It doesn't. This whole thread (except of course where we tenuously discuss sensor heating and DR reduction :):eek::rolleyes:) is a prime example of utter confusion. It seems to me that the proper use of the term HDR is when you combine two or more shots with different exposures to go beyond the DR limitation of a single shot. Using a single image, whether or not manipulated after capture to create multiple copies with different post-processing, is just tone mapping.

In @AlanKlein's example, "HDR in the camera" would be better described as a picture profile or style. My iPhone, on the other hand, takes three exposures at different settings, and correctly renders the results as HDR.

"Log gamma" in a video camera is correctly listed as a "picture profile" rather than HDR, although it has a similar effect. This setting lowers the exposure by 2 or more stops (depending on the implementation) and flattens the characteristic curve at the upper end, making it nearly impossible to blow highlights. Judicious use of ND filters keeps the exposure in range of the shutter angle (speed), aperture and gain (ISO). Log gamma addresses only the upper part of the curve, but most of the time that's sufficient. RAW video carries that another step in using the entire range of the sensor with little or no processing (profiling). Both log gamma and RAW video would qualify as "grunge" before grading and finishing.

You beat me to the punch talking of iPhone which does actually provide HDR using multiple exposures: Adjust HDR camera settings on iPhone

HDR (high dynamic range) in Camera helps you get great shots in high-contrast situations. iPhone takes several photos in rapid succession at different exposures and blends them together to bring more highlight and shadow detail to your photos.

So how does that make it HDR?

It isn't. I was comparing so-called single shot HDR claims to cameras that do the same thing. However, I just checked my manual. It says the Sony RX100 HDR is actually using three shots, not one, that it combines for HDR. You can select the EV spread from 1 to 6 stops. For example -2, 0, +2 stops. So in that case, it really is HDR. I don;t see how you can vary the allocation. It seems fixed unless SOny is using some oddball algorithm. . I would think HDR software works better but I;ve never used it. Someone else has to explain that operation.

Here is an example of a single frame of a landscape before and after HDR (q.v., tone mapping) treatment in Aurora2019HD

Then

It isn't. I was comparing so-called single shot HDR claims to cameras that do the same thing.

So there is no such thing as a 'single shot HDR' yes? So called doesn't make it so, it only seeds confusion towards those who understand less about a process (any process) than those using so called incorrect terminology.

However, I just checked my manual. It says the Sony RX100 HDR is actually using three shots, not one, that it combines for HDR.

Yes, like the iPhone or any other blending of bracketed captures, the results can be/are HDR. Got nothing to do with bit depth either although much HDR work is done high bit, floating math. But that isn't a requirement.

Although many HDR photos involve multiple exposures, sometimes you might want that HDR look but only have one image to work with. With HDR Efex Pro an advanced tone-mapping algorithm can expand the dynamic range of a single image to produce that classic HDR look.

Marketing rubbish; no, they cannot expand the DR of the capture; it is what it is. Can they alter the image appearance or make it have an HDR look (undefined term, perhaps that IMHO ugly, "grunge look"), sure. I can do that in LR or Photoshop or any number of non actual HDR products.

"HDR", to my understanding, is the process of rendering bracketed images into a 32 bit/channel image file, so that the lightest and darkest tones fit inside the limits of 0 to 100%, with 2^32 steps in between. Tone-mapping is the usual finishing process making the HDR image suitable for display or printing. The tone-mapping process is more complicated than simply mapping the 32 bit channels to 16 bits (which changes the number of steps without changing the end points).

That's how I understand it too.

In summary, traditional HDR processing has two steps, the last being tone-mapping. "HDR" processing of a single image "HDR" has only the tone-mapping step, hence is not actually HDR, regardless of what the software calls it.

At least some of the programs that do HDR from a single image require that at least one more bracketed shot is created from that image beforehand - then the program runs the normal HDR engine. I don't have a complete overview over all the HDR processing software out there but it is entirely conceivable that those that don't require those bracketing shots be created from a single image do so internally, hidden from the user.

So it appears that in my two scenarios given above there might be - as digitaldog pointed out - a difference in actually needing to do an HDR to get all the scene DR into a single image - but from a procedural point of view, nothing prevents anyone from running any bracketed shots (done with real bracketing or created on the computer) through an HDR program and create an HDR image - nevermind if there actually is a dynamic range need for doing so. I am not going to dissect various definitions of what is and isn't HDR nor am I willing to dive into the underlying math to see what exactly is done to the data. I am satisfied if I can produce results that satisfy my wants and needs - how the underlying algorithms work doesn't really concern me; learning how to get out of them what I want much more so.

BTW, here's what NIK says about the single image HDR:

Although many HDR photos involve multiple exposures, sometimes you might want that HDR look but only have one image to work with. With HDR Efex Pro an advanced tone-mapping algorithm can expand the dynamic range of a single image to produce that classic HDR look.

They reveal that it is actually just tone-mapping. Though one can question if "expand the dynamic range of a single image" is actually true.

Marketing rubbish; no, they cannot expand the DR of the capture; it is what it is. Can they alter the image appearance or make it have an HDR look (undefined term, perhaps that IMHO ugly, "grunge look"), sure. I can do that in LR or Photoshop or any number of non actual HDR products.

On that we can agree. As is so often the case, in everyday life strict definitions of terms are ignored or not even fully known and thus strictly speaking used either incorrectly or totally out of context. That just the way it is - and in this particular case, it appears that no harm comes from it. I can certainly dig up the original literature on HDR and possibly even understand what it says - it's not going to cure that terms are used incorrectly all the time. And it hardly appears to be an issue that's worth going on a crusade for.

Just look at a parallel thread where the term "exposure" was dissected. I admit I've gotten it wrong too in posts here on PN several years back when I learned about the "exposure triangle". I have now learned where I went wrong - but frankly in the end it amounts to "distinction without a difference".

Shakespeare has been credited with "inventing the English language."

"HDR" is a popular visual notion or impression.

"Full stop" defines constipation.

I doubt anybody thinks multiple iphone shots expand dynamic range of any individual shot, even if identically timed.

I doubt anybody thinks multiple iphone shots expand dynamic range of any individual shot, even if identically timed.

And that would be incorrect. That's a problem when you think for others and ignore how stuff actually works (in this case, with respect to an iPhone). Do you own one, care to do some testing?

”HDR" is a popular visual notion or impression”

An incorrect notion and impression for some, hence such discussions.

I doubt anybody thinks multiple iphone shots expand dynamic range of any individual shot, even if identically timed.

The iPhone takes three shots at bracketed exposure levels, which should do the trick.

I thought it was Chaucer, not Shakespeare with that attribution.

How To Use The HDR iPhone Feature To Shoot Perfectly Exposed Photos

So, how exactly does the HDR iPhone camera feature work? When you use HDR, the camera captures several different exposures at different brightnesses. The camera then blends the best parts of these exposures to create a single well-lit image.

As a result, the dynamic range of the photo increases. This means you’ll see more color and detail in both the shadows and the highlights.

While this process might sound complex, the good news is that the iPhone does it all for you. All you need to do is switch on HDR and take a photo. The camera will shoot and combine the three exposures to create your HDR image.

That's a problem when you think for others and ignore how stuff actually works

Very smooth, Andrew.

Very smooth, Andrew.

“There are two kinds of fools: those who can't change their opinions and those who won't.“ -Josh Billings

I've not followed or understood much of this thread, life's too short. But it seems to me that one view is that HDR can be achieved by developing the same RAW file to different exposures, then combining them in whatever programme claims to do it. The other view is that no, it's only HDR if the range of tones in the scene is outside the dynamic range of the camera, so different exposures are required, at different camera settings.

My problem with this second view is that, remembering that the dynamic range available has increased significantly as digital cameras progressed, that would mean that whether or not it was HDR would depend on the dynamic range of the camera, so it would change as designs improved. This doesn't seem to make sense to me.

Or have I taken leave of my senses?

But it seems to me that one view is that HDR can be achieved by developing the same RAW file to different exposures...

Hold on before we continue. Exposure ONLY takes place at capture and is ONLY an attribute of the amount of light striking film or a sensor. From Aperture and Shutter. It isn't ISO, it isn't something that takes place after capture. Despite what a control in some software may be named. One never develops a raw or anything else with differing exposures after the image is taken.

As to DR, no available 'consumer' digital camera can capture the entire DR of many scenes. See: http://www.lumita.com/site_media/work/whitepapers/files/pscs3_rendering_image.pdf

Including:

Many people erroneously think of a photograph as an “exact” reproduction of a scene. The reality is that light levels in a natural scene can’t be reproduced using any current technology and certainly not in a print. We must squash the vast dynamic range of sunlight and shadow onto the paper of a print that simply can’t hold it.

One way to record our scene would be a perfect scientific capture called a spectral pixmap. The image plane would be divided up into a grid of pixels (picture elements); for each pixel we would record 300 values representing the energy for each wavelength of visible light. Spectral pixmaps are huge. A 10-million pixel image would be 5.8 gigabytes.

Oh right. I guess the confusion stems from using the word exposure loosely to mean the appearance of an image.

This has prompted me to buy and hopefully re-read a new copy of the book "Eye and Brain" by R. L. Gregory, which I read some years ago. It explains the astonishing range of sensitivity of the eye / brain system, only a very small part of which is derived from the opening and closing of the iris. The book is also very revealing about our perception of colour and movement, if I recall correctly. It made me realise that as photographers we are trying to achieve the impossible, that is, to represent on a flat coloured rectangle, the things that our eyes and brain convert into what we think of as the actual appearance of the world.

Oh right. I guess the confusion stems from using the word exposure loosely to mean the appearance of an image.

Alter brightness (or alter tone, tone 'mapping' etc). Again, exposure in photography is pretty simple; the amount of light sticking film or sensor as a result of Aperture and shutter.

The super easy method is to choose the brightest channel, copy it to the new layer in soft light mode, invert it to negative. Copy the background layer on top of it and use lighten or darken or normal mode to decrease shadows, highlights or both. Additionaly you can blur that negative with gaussian blur to get more detail. Creating an action makes it very fast.

Also using Apply Channel from Image menu allows for the same procedure.

I guess the confusion stems from using the word exposure loosely to mean the appearance of an image.

I think that's exactly right. It's interesting that some people use terminology that avoids that. For example, rather than calling an image that appears too bright "overexposed"--which it may or may not be, depending on ISO and postprocessing--they call it "too hot".