How much DR is enough?

[BillB]

Once again, we await the rollout of a new fullframe Canon camera, so once again posts are peppered with speculation, opinion and information about "Dynamic Range", along with the magic nambers associated with it. And once again, some questions come to mind. Of course, the answers to these questions vary from person to person, so there are no right answers.

Is more DR always better? In some theoretical sense the answer may be yes, but my questions have to do with the practical value of more DR.

First, is there a level beyond which additional DR has no practical value?

Second, how much difference in DR is of practical value?

Third, what other features are important enough to be traded off against DR?

 

[snoke]

Second, how much difference in DR is of practical value?

Like human eye.

 

[denstore]

This might sound stupid, but I have never found any need of more DR. But that might be because I rarely spend lot of time in post. I do what I feel is nesessary at the moment, and if I can’t bring up shadows or whatever more than to a certain point, I don’t waste time crying about it.
I don’t make a living from photography, even if I take a lot of photos at work. It might be that the absence of extreme DR doesn’t matter to much to my pay check that makes me care little, but I do find that people tend to be a bit fixated on DR and DxO scores....

 

[Durf]

Well, for what its worth, I shoot with the 6D2 and have very seldom ever ran in to an issue concerning DR. The very few times I have had to bracket in extreme conditions, I would almost bet those using a 5D4 would of had to likely do the same.

I do admit I have to plan my shots in extreme conditions, but I've always had to do that even when I was shooting film 40 years ago!

Yes, there's cameras better in the DR than the 6D2 but why are these folks shooting 5 stops under-exposed?????? dah.

regardless, I am predicting Canon's mid range and flag ship mirrorless FF cameras will be rather nice and very competetive with or even better than Sony's.......the M50 has some cool features for under 700 bucks, just imagine what Canon will put in a 3000.00 dollar FF mirrorless. It'll likely have very nice DR capabilities too...

 

[Orangutan]

Is more DR always better? In some theoretical sense the answer may be yes, but my questions have to do with the practical value of more DR.

Here's my benchmark for that: if I'm in a deep forest, I should be able to expose for the deepest shadows, but not blow-out the sun that's poking through a hole in the canopy. (Alternately, no blown specular highlights off a stream in the frame). If I remember correctly, this is about 18 stops at brightest day.

First, is there a level beyond which additional DR has no practical value?

Beyond the scenario above, I don't know that more would help because, in effect, I'm simultaneously taking two separate, well-exposed photos at extreme ends of the illumination scale.

Second, how much difference in DR is of practical value?

It depends on your subject matter. For a studio situation, you control the light so we're probably past that point now, even with APS-C cameras. For landscape, see my answer above.

Third, what other features are important enough to be traded off against DR?

Everything else. If you mean just regarding the sensor, then the features that could trade against DR might include low-light sensitivity, resolution, frame-rate, heat dissipation (for LV/EVF).

 

[wsmith96]

I don’t believe there is a hard and fast rule on DR requirements as it depends on what your subject matter is and your taste in how you process. Per DXO, 12 EV is considered excellent and many cameras meet or exceed that value. Personally, I don’t shoot landscapes primarily, but I’ve had no problems using a 60D to do so when I did and it has a DR score of 11.5. I also don’t lift the shadows that much, which is important to some.

 

[3kramd5]

At this point I think we generally have enough in most raw capable cameras. There are some scenes which I can not image and don’t expect to without something like MIT’s modulo camera that keeps clearing the saturated sensor and continues counting.

I’d rather see improvements in the contrast ratio of media. Even if I can capture 14 stops in a digital file, I have to map it to maybe 7 for display on screen and 5-6 for print.

 

[rrcphoto]

Once again, we await the rollout of a new fullframe Canon camera, so once again posts are peppered with speculation, opinion and information about "Dynamic Range", along with the magic nambers associated with it. And once again, some questions come to mind. Of course, the answers to these questions vary from person to person, so there are no right answers.

Is more DR always better? In some theoretical sense the answer may be yes, but my questions have to do with the practical value of more DR.

First, is there a level beyond which additional DR has no practical value?

Second, how much difference in DR is of practical value?

Third, what other features are important enough to be traded off against DR?

the higher the DR, the less noise in the midtones and shadows, and arguably the higher bit depth the raw files could be.

while practically today's modern sensors are probably "good enough" there's always room to get better, and why not take the lead again? why always look at Sony sensors as "the benchmark".

If anything else, it would be nice to see what other things DPR looks at to say that Sony sensors are still ranked the best.. color accuracy or tonality or something I'm sure.

high DR is also harder to achieve as you increase the speed of the sensors. the increased readout lines take away from the sensor's efficiency,etc.

so any core improvement in DR makes up for losses in other areas.

 

[dak723]

I understand that there are instances when you need more DR, but I have almost never encountered one using my 6D and now my M5. I have tried the Sony FF A7 II - and found that in my landscapes (compared to my 6D) there was no benefit and perhaps a detriment. Compared to my old 300D, I now find that I have to post process more and add contrast to get my photos the way I want, again realizing each person has their own opinion on how a photo should look. I consider contrast to be far more important than a wide DR range in the majority of my shots. As the DR increases, I see so many dull, washed out photos that have a lack of contrast.

Coming from an art and painting background, I may be looking at things a bit differently than the mostly "tech" crowd on this forum. One general rule of painting is that the human eye does not simultaneously see details in both the light areas and the shadows, thus one should not be painting detail in both. Either the light areas have detail and the shadows are essentially dark and featureless or vice versa. Yes, I understand that you don't want those shadows to be totally black in most cases, but a bit of shadow lifting is easily done with all Canon cameras that I have used. But I don't want anything close to the same level of visible detail throughout the value range of the photo. In my opinion, that is why too much DR - and especially the DR of HDR photos - is often a negative and the end results look fake, washed out, and rather boring. Again, that is just my opinion.

I have found it interesting that the desire for more DR is so overpowering that even the reviewers fail to see the negative side. A number of years ago I remember reading a camera review of a new model. The reviewer complimented the camera for adding a stop or more of DR and then in another part of the review lamented the fact that photos from the newer version seemed to have lacked the "punch" of the previous generation. They failed to make the connection that the lack of punch was directly related to the lesser contrast and the greater DR that was now available on the newer model.

Yes, I understand that with enough post processing one can make use of the greater DR and then add more contrast. But having tried the "better" (more DR) Sony sensor and compared the results with identical shots taken with my older sensor generation 6D, I preferred the look of the 6D shots. So more DR is not anything I care about or am looking forward to. Not in the least.

 

[stevelee]

In what I'd consider even vaguely normal conditions, I've not experienced any problem with dynamic range. I shoot RAW and when needed can touch up things in ACR with the Shadows slider and a bit of highlight recovery with the Highlights slider. Like Durf, I don't shoot five stops underexposed. There is also the matter of trade-offs. Would I give up the quality of what my 6D2 can get at higher ISOs just to add a fraction of a stop dynamic range at ISO 100? My answer would end with "José." YMMV.

In extreme situations, I bracket. On the wall across the room from me is a picture from the Order of the Thistle chapel in St. Giles Cathedral in Edinburgh. I shot one exposure for the interior and another for the stained glass. That maybe took me an extra fifteen seconds when shooting and a lesser time for Photoshop to put them together back home. The result is glorious (he said modestly). I chose to keep the converging verticals to retain the feel of looking up. It represents well how I remember seeing it. In real life, my brain likely put together a composite of what I was seeing when I looked different places, so maybe even my eyes could not handle simultaneously that range of light while preserving the deep colors of the windows and the rich woods and stone walls. (Rods and cones, and all that.) Even if the camera could handle that dynamic range, it might not achieve the balance between things and give me the control I had in Photoshop, so I might still make a couple of bracketed exposures.

When I visited the Garden of the Gods in Colorado Springs, it was almost sunset. I took backlighted pictures of the rock formations. I wanted both the beautiful sky and the rock formations to look good, so I bracketed and merged as above. I was much less experienced with my then new G7X II, so I would do a better job of that now, but still got some decent pictures. Later in the trip at Arches National Park I did better. By the next spring in Edinburgh I did even better. I'm not fond of the HDR look, but am not above using the technique to try to make natural looking pictures.

 

[stevelee]

Here's my benchmark for that: if I'm in a deep forest, I should be able to expose for the deepest shadows, but not blow-out the sun that's poking through a hole in the canopy. (Alternately, no blown specular highlights off a stream in the frame). If I remember correctly, this is about 18 stops at brightest day.

I still have my filter for shooting the eclipse last August. I hadn't thought about this until your post. I realize now that I could do extreme bracketing in a landscape by exposing a shot or two for earthbound things maybe while the sun is partly behind a cloud, and then put on the filter and expose a shot for the sun after it emerges. If I post any landscapes here that also let you see the sunspots, you'll know how I did it.

 

[3kramd5]

One general rule of painting is that the human eye does not simultaneously see details in both the light areas and the shadows,.

That may be true to a certain extent, but my eyes can see simultaneous bright and shadow details beyond what any single exposure camera I’ve ever encountered is capable of recording, and significantly beyond any media (be it paint on canvas, ink on paper, or pixels in screen with back lighting).

 

[stevelee]

One general rule of painting is that the human eye does not simultaneously see details in both the light areas and the shadows,.

That may be true to a certain extent, but my eyes can see simultaneous bright and shadow details beyond what any single exposure camera I’ve ever encountered is capable of recording, and significantly beyond any media (be it paint on canvas, ink on paper, or pixels in screen with back lighting).

And some of that comes from the way that the brain puts together information together as you look at different things. Light and color sensitivity varies all over the place on the retina. https://en.wikipedia.org/wiki/Fovea_centralis

 

[Kit.]

What I think is important is not high "linear" DR as such, but the ability to nonlinearly compress overexposed highlights in order to keep some detail in them.

 

[stevelee]

What I think is important is not high "linear" DR as such, but the ability to nonlinearly compress overexposed highlights in order to keep some detail in them.

And I think that is a lot of the problem. The eye is quite non-linear (the ear even more so). Digital sensors are pretty close to linear. Something like one stop of exposure on the highlight end takes up half of the information, or so it looks to me on a graph. That leaves stops and stops of the shadows all scrunched together. No wonder there is so much noise in the deep shadows. Converters (in camera, ACR, Lightroom, et. al.) all compensate for that. I wonder if anyone has tried to make an image look like the RAW picture would look if we could see it. It might approach a line drawing on a white background.

 

[3kramd5]

One general rule of painting is that the human eye does not simultaneously see details in both the light areas and the shadows,.

That may be true to a certain extent, but my eyes can see simultaneous bright and shadow details beyond what any single exposure camera I’ve ever encountered is capable of recording, and significantly beyond any media (be it paint on canvas, ink on paper, or pixels in screen with back lighting).

And some of that comes from the way that the brain puts together information together as you look at different things. Light and color sensitivity varies all over the place on the retina. https://en.wikipedia.org/wiki/Fovea_centralis

As does resolution across our FOV.

 

[SUNDOG04]

I love this question and very interested to hear the thoughts of others. I use a 6D, mostly landscape, and I am able to lighten shadows without a problem to get photos to try to match what I saw. And, even in shadows, I see more detail in the photos than my eye sees. So I am happy. If I took the same photo taken with a camera with a higher dynamic range, would I then not be as pleased? I don't know.

At what point do you go the HDR? Those pictures look stunning and seem to get very positive feedback. But they also do not look quite realistic, even if they do look stunning.

 

[IglooEater]

IMO DR would only be objectively enough when I can photograph any scene without adjusting my exposure. Enough to handle white snow in full sun, and the Milky Way in a single shot. So, what? 30-40 stops? This is why I generally disregard DR as a major differentiation between cameras. They’re all so dang close to the same.

 

[3kramd5]

IMO DR would only be objectively enough when I can photograph any scene without adjusting my exposure. Enough to handle white snow in full sun, and the Milky Way in a single shot. So, what? 30-40 stops? This is why I generally disregard DR as a major differentiation between cameras. They’re all so dang close to the same.

What do you mean by “without adjusting my exposure?” Do you mean in camera never changing focal plane exposure (result of aperture and exposure time)?

Or do you mean after the fact?

If the latter, having 40 stops of range in a digital file will require more post processing work than ~13, as it vastly exceeds the ability of anything to display it.

 

[stevelee]

What I think is important is not high "linear" DR as such, but the ability to nonlinearly compress overexposed highlights in order to keep some detail in them.

And I think that is a lot of the problem. The eye is quite non-linear (the ear even more so). Digital sensors are pretty close to linear. Something like one stop of exposure on the highlight end takes up half of the information, or so it looks to me on a graph. That leaves stops and stops of the shadows all scrunched together. No wonder there is so much noise in the deep shadows. Converters (in camera, ACR, Lightroom, et. al.) all compensate for that. I wonder if anyone has tried to make an image look like the RAW picture would look if we could see it. It might approach a line drawing on a white background.

After I posted that, I decided to give it a shot. I applied an antilog curve to a picture with a fair dynamic range, and this is what came out:

Maybe that is what the RAW picture would look like. Maybe not. The curve and then the JPEG conversion lost highlight detail, the opposite of what is in the RAW file, even if to our eyes it would look like this. Something like that might make a nice print.