I really like the added dynamic range that HDR offers, but dislike the over-use of HDR to the point where images look unreal. I want my images to look “real” and sometimes HDR oversteps that. I’ve tried a variety of software for HDR and I have found one underlying theme that always results in a realistic-looking photo that also has high dynamic range:
- Use a single-image RAW file for HDR.
It’s true. Using a properly-exposed single image RAW file to create an HDR image, nearly always provides a more realistic-looking photo in the end. This is because with a bracketed set of photos everyone (including me) is tempted to overextend the dynamic range beyond the “real” level. Beyond what you would actually see if you were there.
Take shadows for example. My eyes can’t see into shadows, so why should my HDR process open up those shadows so much. A little, yes, but not to where everything can be seen in them. I can’t do that in real life, so why do I want to show it in a photo I took? I want to get only up to (and maybe slightly beyond) what I could see while I was there.
Highlights are the same way. Squinting into the sun is a natural thing. You’re going to do it. So why can’t your photo reflect that as well. It can and should. So blown out highlights are going to occur in a photo because they were definitely there when you were actually there viewing the scene you photographed. That doesn’t mean that you shouldn’t help out those blown out highlights a little bit. You should, because the camera artificially blows out the highlights sometimes because its sensor can’t handle so much light. Your eyes are different. They can resolve more detail in light areas, so you should bring in those highlights in your photo using the HDR software process, but only to where it looked like what you saw when you were there. After all, I can’t see texture on the sun or in glare, and your photo shouldn’t reflect this either.
But why does a single-image RAW file work for HDR? Simple. A DSLR covers about 11 EV (stops), but a JPEG file on your computer screen is about 8 EV. A printed image is only about 6 EV. Your eyes on the other hand are about 27 EV. So you can see more than you can ever represent in a digital photo. But, using HDR tonemapping techniques, you can extend that RAW file’s information into the upper and lower luminance ranges and “represent” the colors and light that you saw with your eyes, making the image seem like it is closer to what you saw with your eyes.
For example, an 11 EV RAW file can be manipulated through tonemapping to appear closer to the 27 EV that your eyes see.
Are you curious what your camera is actually capable of? Can it really take in an 11 EV image? You can find out exactly what your camera can do if you look it up on DxO Labs “DxO Mark” website.
Stick with a properly-exposed single image RAW file for your HDR tonemapping and you will consistently come away with photos that are closer to reality!