One of the longest standing challenges facing photographers has been working with photographic media that inherently has a lower dynamic range than what we, the photographer, see with our own eyes. This challenge has spurred a lot of innovative solutions enabling photographers to push film and digital media to their limit. These innovations have included various darkroom techniques for film, graduated neutral density filters for in the field use and a variety of digital dark room techniques including High Dynamic Range (HDR) processing.

Of all the approaches to address the limited dynamic range challenge, High Dynamic Range (HDR) processing is one of the few photo techniques that I’ve grown to hate and here is why:

Mathematical vs Aesthetic Solution
One of the two main reasons that I hate this technique is that it is a solution centered more around math than it is around aesthetics. Mathematically the output of an HDR rendered photo tells us that it represents what the human eye would see, but the software creating an the HDR photo lacks the intelligence to display the output as the human brain would process it when naturally viewed.

As I see it two fundamental problems face HDR ouput:
1. Dynamic range is an interpretive process.
The human brain determines dynamic range in a relative fashion by formulating lighting relationships and physical relationships learned from years of experience in our environment.

2. Current display technology lacks the ability to accurately display HDR output. Mathematically an HDR file might equate to what we would see with the naked eye, but current monitors are unable to display the entire tonal range of that file. In fact HDR files are output as tone mapped files that match the dynamic range capabilities of standard monitors or prints. As a result much if not all of the added benefit is unrealized.

Novelty vs. Applied Solution
The next reason why I hate HDR photography is that most people doing it approach its use as a novelty rather than an applied solution. In the process most images are completely butchered. This is most especially true of landscape and nature photographs. Much can be said for artistic style and there are a few photographers out there that have used HDR processed files rather creatively, but by and large most photographers are using HDR as a “gee whiz neat” function that does nothing other than create throw away images worthy neither of repeat viewing or print.

With all new output solutions, whether film or digital software functionality, there is an evolution of use. Photographers first experiment, use to an extreme / over use and then settle on a refined more balanced output. With digital there seems to be more of an extreme swing of experimentation and use to an extreme / over use and this is all to apparent with HDR processing. See this for yourself with HDR landscape photos on Flickr.

I find it quite ironic HDR processing, the functionality to create imagery with a dynamic range equivalent to that the human eye sees naturally, is seldom used for this purpose. In fact the pursuit of non-realistic tonal range is often what people seem to be aiming for. Consistently, as I have investigated current HDR trends, the images that look the most unnatural and unattractive to me are created using Photomatix. Adobe Photoshop CS2 was one of the first widely available applications to provide the functionality to create HDR imagery, but as of late Photomatix seems to be the application of choice for many would be photographers/digital artists. How Photomatix came to be the application of choice I’m not sure, but its output is truly the worst. Examples of Photomatix created imagery can be found in the Photomatix group on Flickr.

So what is it about HDR processed images that I hate?
The vast majority of HDR processed photos do not display the world in a natural fashion.

1. Uneven Shadow Tonality – where there should be shadow detail there is often not and shadow tonality is often inconsistent for areas that should be equivalent.

2. Nearly Reversed Tonality – sky and clouds often are dark while areas you would expect to see natural shadows are bright

3. Unnatural transitions between highlights, midtones and shadows – in combination with the previously mentioned reversed tonality, tonal transitions often look to have an inconsistent if not backward appearance.

4. Extreme Contrast – tonal extremes, in combination with previously mentioned reversed tonality and unnatural transitions, often create results that look cartoon-ish and overly graphical.

5. Reduced Contrast – depending on the scene and application used the net output of an HDR processed image can result in an image with unnaturally reduced contrast. The net result is an image that appears too bright with too much shadow detail.

HDR Alternatives
If you’re looking to create a more natural looking image with a broader dynamic range I would recommend a few other approaches.

1. Graduated Neutral Density (GND) Filters – using these types of filters will assist in balancing your exposure by selectively reducing the intensity of light reaching your film or digital sensor. The net effect is narrowing the dynamic range of the scene so your camera can achieve a proper exposure. The upside to using GND filters is that this reduces the need for some post-processing saving you time. The downsides are that additional equipment is required and GND filters are designed with a linear graduation impacting (albeit minimally) photographed scenes with uneven horizons (think tree lines).

2. Masking and Local Adjustments in Photoshop – one trick of the trade to deal with complex and uneven horizons is to make adjustments via masks in Photoshop. There are a variety of tutorials out there and the technique is quite easy to learn. The upside is you gain a lot of tonal control over your image, but the downside is that you can burn a lot of time while post-processing your image.

3. Double Processing RAW Files – if you’re shooting digitally in RAW format you can process one image twice (potentially more) to achieve an output for each tonal range of interest to you. This in essence is a scaled back version of HDR processing. The upside is you gain a great deal of control over your adjustments, but the downside is that it takes additional time and is a manual process requiring a little more hands on work.

Final Thoughts
Although HDR processed images theoretically can come the closest to matching the dynamic range we see with our eyes, they are never fully realized in output on monitors and print. So given the display and print limitations are you creating a better photo if you can never see the difference? I say, “No”. Dynamic range limitations of film and digital sensors have created opportunities to create less clinical reproductions of reality enabling the introduction of mildly artistic elements such as contrast, vignetting and so on.

Since dynamic range gains from HDR processed images are not seen with current display technology many have taken to exaggerating settings to produce wildly out of whack images that are now becoming the associated image type with HDR. This is regrettable two fold as HDR has become the latest photo adjustment fad and strangely when an image is properly exposed and processed viewers are now more apt to ask if the photo is an HDR processed image. Sadly HDR has long crossed over into the category of photo technology porn. The availability of the technology has resulted in its over use and abuse resulting in a flood of sub-par images. Amidst the flood of these sub-par images there are some photographers producing very naturally looking HDR images, but regrettably they are the exception.

Links:
High Dynamic Range Photography Taken with GND Filters Natural looking HDR photography by Sean McHugh

Using the High Dynamic Range (HDR) Feature in Photoshop CS2 by Sean McHugh

Adjustment Layers Are Your Friend by Russell Abraham

Double-Process RAW For Better Tonality – PCPhoto

[tags]HDR, High Dynamic Range, Photoshop, Photomatix, graduated neutral density, filters, adjustment layer, RAW, double-process, tonality, dynamic range, [/tags]