Back when I was using 4×5 and 8×10 sheet film, if I really cared about an image, I would make two identical exposures. I’d make the first normally, then flip the film holder over and do it again. When I developed the film, I’d sometimes do all of one side of the holders in one batch, and the other side in another. The purpose of all this rigamarole was to make sure I had a good neg if there were a processing oopsie, say uneven development that resulted in a mottled sky. The processing was usually fine, so I was left with a choice: which neg to print? Should be easy, right? Just pick one at random. I soon found out that didn’t work. Even with supposedly identical negs – and this was with static subjects and the camera on a tripod – there was almost always one that I liked better than the other. The difference wasn’t usually processing, either. It was subtle things like cloud patterns, leaf motion, wave action, and the like.
Hold that thought.
A couple of years after I started work as an engineer, the research outfit that employed me hired someone to work alongside me. One day I needed a certain resistor but couldn’t find one in the stock bins. I went looking for a switchable resistance box to use until new stock arrived. I couldn’t find one. After a while I dropped by the new guy’s bench to find it covered with resistance boxes. Rather than design the circuit properly and calculate all the resistances necessary to make it work right, this guy was twiddling knobs until he happened upon a design that would work. Needless to say, he didn’t last long.
Hold that thought, too.
I delight in surprises when processing photographic images. I enjoy setting up situations engineered to produce those surprises. My fondness for revelations in postproduction does not extend to exposure errors. I want files whose exposures are designed for the kind of processing I have in mind. I take some pride in being able to plan those exposures, using strategies appropriate to the intended development.
Got all three of those thoughts?
I recently encountered a person who said that his ISO-selection and exposure strategy was bracketing. My first reaction was, “That’s not a strategy; that’s rolling dice.” I thought immediately of the incompetent engineer and the lab bench full of R-boxes. All of my instincts told me that this was a terrible idea.
But a cooler side of my head prevailed, and I started thinking. Cameras automate bracketing, and that’s really useful for HDR, so maybe they can bang off shots so fast that my old experiences of supposedly identical shots being only quasi-identical aren’t applicable. I don’t think cameras automate ISO-setting bracketing, though. But still, just exposure bracketing could go a long way to finding one capture that positively nails the exposure, with the raw file just short of highlight clipping. And that’s hard to do in the absence of an in-camera raw histogram.
So, much as I hate the idea, I’m coming around to bracketing as an exposure protocol (I still bristle at calling it a strategy, for some crazy reason) for static subjects, and maybe some slow-moving ones. Practical difficulties that will make it unworkable for me in some situations include the reduction of dynamic range that some cameras experience in continuous shutter mode, and the logistics of combining this with focus bracketing, panos, and the like.
I sure won’t like what that will do to the culling process, which is tedious enough already. Fast Raw Viewer to the rescue? And I’m afraid that, should I start using it, all the while I’ll be thinking about a million monkeys, a million typewriters, a million years, and the complete works of Shakespeare.
Ah, as a pseudo-engineer (I taught maths to engineers at an Australian university for many years, and I build stuff), I sympathise: do it right, do it once. It is a trait appreciated by a pianist friend whose concerts I photograph. With me she hears an occasional click, with some photographers it’s like rain faliing on a tin roof 🙂
To bracketing then: I believe Pentax provide ISO bracketing on their DSLR’s. But I’ve never used it. The line between principle and pig-headed-ness is fine 😀
Auto-exposure bracketing is not necessarily similar to rolling dice. If one has a reasonable amount of experience with the in-camera meter’s behavior, then the bracketing window is based on the meter’s estimate. I find 0, +1/3, -1/3 works best. Occasionally +/- 1/2 stops is useful.
Perhaps the issue is whether or not an additional 1/3 to 1/2 stop of exposure will make a significant difference to shadow-region perceived image quality?
In dynamic situations (e.g. candids of people) auto-bracketing lowers the work load and helps me be more situationally aware.
When I was doing interiors photography gigs, I manually bracketed shutter speed because selectively pushing shadow regions in post production was unavoidable. The dynamic range of the scene almost always exceeded the camera’s dynamic range . This was not for blending multiple exposures. My goal was to intentionally overexpose unimportant highlight regions to their practical limit.
My favourite camera remains my Yashica TLR because of what it expects from me. WLF, zoom with your feet, 6×6 negative and 12 exposures per roll.
That thinking carries over to my digital work to some extent, but I do shoot more liberally with digital. Still, it’s not a spray and pray approach. Ialso try to do things like making serious images with my phone, just because; the limited dynamic range makes exposure bracketing a virtual necessity.