High Dynamic Range imaging is one of the most polarizing topics in photography today. Critics have an easy day: due to it’s emergent nature and technical complexity, many HDR photos are awful. This fact is then used to support various tedious and hyperbolic arguments that HDR is ruining photography, promoting sloth, and making people’s eyes bleed.
HDR will never turn a technically flawed, boring or poorly composed image into a great photo. It’s a powerful tool, and used without discretion, skill or artistic intent, HDR techniques can turn a great photo into an overblown, gaudy, noisy atrocity … or worse.
I have produced hundreds of truly regretful HDR images over the years…and a handful of good ones. With the benefit of 20/20 HDR hindsight, I’ve figured out some of the things that I did — out of laziness, stupidity, oversight or cluelessness — to produce my worst HDR images and remembered some of the things that I’ve done to make them better.
Note: This isn’t a tutorial, it’s a confessional. A tale of woe and words of warning. I’m not going to give you step by step instructions on how to do it right…I’m going to tell you what I’ve done wrong. I’m also going to assume the following:
- You have at least an intermediate-level understanding of photography
- You have a DSLR and you’ve read the manual (or know where to find it)
- You have software capable of making HDR images and tone-mapping the results
So, without further ado, and in no particular order, here’s what I’ve done to make my HDRs suck, and what I’m trying to do about it.
I thought I could take shortcuts
Shooting for HDR requires a significantly greater degree of technical precision and frame-to-frame consistency than regular photography. Many DSLR features that we have come to rely on are poison to HDRs, including aperture priority metering, auto-focus, auto-ISO, auto-white-balance and vibration resistant lenses. All of these features can — and will — change important values in your images, even in the time it takes to shoot two or three frames at burst rate.
What I’ve learned:
- I really needed to use my camera’s fully manual mode when shooting for HDR. In general, you want to use the lowest ISO possible and the lens’ sharpest aperture but it depends on the scene and subject.
- After my subject is in focus and I’m ready to shoot, I now turn auto-focus off so the camera doesn’t decide to focus on something new. It’ll also help to decrease shutter lag and latency because the lens won’t even try to re-focus.
- I always turn off auto-white-balance. I set it to an appropriate value and don’t mess with it. I was horrified to find out how often the camera changes this over the course of three bracketed shots.
- Auto-ISO is evil.
- I changed my color mode to the factory defaults. Most digital SLRs have a color mode selection menu. Leave this set to “Normal” and avoid the custom settings until you know what you’re doing.
- Assuming that the rest of your technique is good, you will probably get sharper results if you turn off your lens’ vibration resistance or image stabilization feature.
I used HDR techniques on inappropriate scenes or subjects.
It’s very difficult to produce a good HDR image if the scene involves moving objects. I had no idea just how much stuff — even seemingly immobile stuff — moves around until I started doing HDR. This means that when I shot scenes that included people, cars, clouds, trees or grass I was inducing artifacts or distracting elements into the final image.
Alas, HDR will make anyone with less than supermodel-quality skin tone look like they are coming down with smallpox. HDR techniques usually expand mid-tone contrast and exaggerate tonal irregularities, which is great for shooting skylines at night, but not quite so amazing when it turns last week’s pimple into a pea-sized divot. And we’re not even going to get into what it does to eyes.
Finally, people move. Even when they’re not moving…they’re moving.
- I choose my subjects more carefully than I used to. HDR isn’t for every scene.
- I don’t do HDRs of people. People that I like, anyway.
- I try not to shoot at “peak” times. Shooting at sunrise is lovely. Not only are there fewer people and cars but the light is awesome. If you are a night owl like me and can’t get up at 5 AM, stay up all night working on photos and go out at first light.
I had sloppy workflow practices
Even though I was shooting RAW, I used JPGs as source files for my first HDR images. It was a nightmare. I switched to 16 bit TIFF files, and the results were a bit better, but still not quite right. My colors were crunky and there was too much noise. The problem was that the initial import into Aperture or Lightroom added adjustments that didn’t seem to matter until the HDR creation process blew them out of proportion.
Simply put: garbage in, even worse garbage out. The HDR conversion process exaggerates everything, including any tonal colorations that your RAW processing software induces during the import/export process.
Painful experience has taught me:
- Don’t be a slacker. Workflow exists for a reason.
- Shoot RAW and use a 16-bit workflow or pay the price.
- When exporting source files from a RAW photo management application, zero out every setting before doing so. The image in Lightroom or Aperture will look pale and washed out and awful. This is OK. No one but you will see those files anyway. After you’ve merged the images you will have more control over color and tonality .
- After you’ve merged your images to HDR, save the HDR before you tone-map it. This might seem obvious, but it saves time and will let you try new ideas without having to re-create the master file every time.
- If you’re using Photomatix you can generate .hdr files straight from your RAWs. This is often a very good idea. Not always. Try it.
I was color (space) blind
Color space is one of those things that you never think about until you are banging your head against the wall and crying for your mommy at 2 AM. When I started shooting HDR I was importing images shot in Adobe RGB color space into a program that uses the ProPhoto color space….and then viewing my photos on a monitor that was calibrated for sRGB. And after I exported the images in Pro Photo I converted them to HDR using software that didn’t support that color space. This was bad. Very bad. My bad. Those photos are awful.
I reluctantly realized that:
- I needed to learn about color space. And it’s a pithy, complex topic. You can’t just read the FAQ.
- I had to calibrate my monitor. I did it wrong the first couple of times and that didn’t help.
- I needed to check my photos on lots of systems. Mac, PC, cheap, expensive. The differences are amazing.
- I needed a nice monitor. A nice monitor costs as much as a nice camera. Note: I still don’t have a nice monitor.
While you can shoot for HDR without a tripod, it’s hellishly difficult to do so unless you are an android or a brain surgeon. Sure, it’s a pain to carry a tripod with you, but even a cheap tripod will improve your HDRs significantly.
I came to the conclusion that I had to:
- Get a tripod.
- Get a better tripod. The first tripod was a cheap piece of crap.
Even with the tripod, the camera wasn’t stable
Camera instability is a great way to ruin a potentially decent HDR. While most HDR software products include mechanisms that help to correct for this, all induce noise and make your images less clear.
Touching the camera for any reason at all after you’ve framed and focused your shot (even just pressing the shutter release) makes the camera camera wiggle. Other seemingly miniscule vibration sources that can degrade your image include: the reflex mirror popping up, traffic, trains, footsteps, wind and loud music. The HDR process will exaggerate even the most insignificant vibrations into fuzz, crunk and noise. And if you’re shooting telephoto it’ll be worse.
So I decided to:
- Get a shutter release cable. They don’t make these for my camera, so I got a a wireless remote. I’m currently lusting after Cadillac of remotes, the $299 Promote Control, which is aimed directly at people who shoot HDR.
- Use the self-timer when I forget or can’t find my remote.
- Use the mirror lock-up feature for exposures between 1 and 1/80. The hard part is to remember to turn it off when you’re done.
- Shield the camera from wind. If it’s windy enough to make my dangling camera strap move, I take it off or stabilize it however I can. Rubber bands work great!
- Find the most stable ground to shoot on. If you’re shooting on a wood floor try to position your tripod so that two of it’s legs are over a joist
- I found that many seemingly-stable things aren’t stable at all. Concrete floors move a lot more than you think. So do tall buildings.
- Chop up an old memory foam pillow and make little platforms for my tripod. This helps to help dampen vibrations, especially when doing longer exposures.
I used bad lighting
No, not during the shoot, but during the post-production process. The light in the room that you work in will greatly affect how you process your photos, especially when you’re working on heavily saturated or high contrast images. Direct sunlight? Bad. Fluorescent lights? Worse. Working on your HDRs in bed with the monitor brightness turned all the way down so you don’t wake your significant other? Don’t even think about it.
I’ve come to understand that:
- Indirect, consistent incandescent lighting is good.
- I need to take breaks. Go outside, walk the dog, stand my head. My eyes need a rest as much as I need the exercise.
And then there’s the obvious part…
Beginners suck. This is one of those near-universal facts that we like to think doesn’t apply to us.
It takes practice to learn how to use complex, powerful tools toward an intentional artistic goal. Think of the first photos you shot. Unless you are completely delusional or you shot those photos yesterday, you know that they….well, they sucked, didn’t they? But you learned about the interrelationships between light, glass, film or sensor, paper or screen. You shot more. Your photos sucked a little less. You you got interested in optics and imaging technologies, you picked up a whole new vocabulary, you probably spent a bunch of money on a better camera, a faster lens…and you started to see things differently. So you shot some more. And slowly, maybe over months, probably over years, your photos started to suck less and less until finally one day you pressed the shutter and you knew deep down when you heard that click that you got a good one. Sure, it had imperfections, but by and large, it didn’t suck.
And so you shot some more. And you got better. And now it’s now, and you’re shooting for HDR. And your HDRs suck, and you’re frustrated, and there are people leaving comments on your Flickr page saying that you are being lazy, and ruining the art of photography, and making their eyes bleed. And worse.
So, congratulations. Welcome to the club
Shooting for HDR and manipulating deep-bit images is a lot like learning to shoot in the first place: It’s hard, it takes time, it takes practice. Depressingly enough, it takes a lot of failure and screwing up as well. It’s part of the process. Keep at it.
I could give you some tips on how to tone map your images, and how to deal with noise. I could hold forth on how to make a “realistic” HDR, or how to make one that makes your viewers feel like they’re on mushrooms. But if you’ve read this far, it’s time for you to go out and make some mistakes.
Me too, for that matter.