Panoramas are one obvious use. Creating files with more pixels is another. But can we find enough to outnumber dalmatian puppies ? Onwards, woof.
This is probably my most frequent use of stitching. The square format (or close to) is a classic. It is possible to crop a rectangular frame and get a square, of course. Dynamically, it is a very different experience to getting closer and making two shots. There’s an action to getting the two shots that feels a lot more involving than getting away from the subject and framing a square within the rectangular viewfinder. It feels more active and involved.
Of course, it’s possible that the alignment between the two frames ends up being slightly imperfect. But, is that such a bad thing? Is our eye scanning fixed on a rectilinear referential or do we move about based on emotion? I’m not advocating sloppiness. But it is quite unlikely that the few degrees of misalignment detracts from how we actually perceive the photograph. To me, the dynamic experience more than makes up for a tiny misalignment and experience beats technical perfection 10 times out of 10.
Here, things become even more interesting.
The first frame in this image is focused on the bottom of the V trunk formation in the foreground. The second is focused on the single tree in the middle ground. Getting one frame in sharp focus from 3 feet away to the cave entrance at least 30 feet further, without a tilting lens, good luck with that. Here, using a small enough aperture (f/6.3, but it could have used f/8) insured that the two could be stitched OK, giving the impression of one continuous frame made at f/32 and ISO 10 million (it wasn’t very bright in there).
“Oh the horror” I hear the purist cringe. Stitching frames with different focus points? That’s against every rule in the book. Guess what, it gets worse. Really worse 😉 😉
If the previous photograph was against the rules, then this one comes with a health warning!
Let’s go over the rules of panoramas and try to understand their purpose to determine how and when they can and should be bent!
OK, start with a tripod. Not necessary but it makes all the others a lot more convenient. Level the tripod head. That’s important if you want your horizon to be level and not have to correct it in post (any PP always introduces a little bit of image degradation). Rotate around the nodal point of the lens. That’s not always easy, depending on where the manufacturer placed the screw under the camera. This matters because it ensures that no parallax is recorded between individual frames and the resulting stitch is seamless. This is particularly important if your pano incorporates multiple layers of objects (some in the foreground and some in the background). In practise, I shoot all my panos hand held and very very rarely encounter any issues with parallax. Same focus point. We’ve covered that. Same exposure. We’ll cover that below.
Now let me describe the process for the photo above. The camera was tilted very slightly to create that slope towards bottom right and increase the feeling of the house being dragged down. The frame on the left was shot as stable as possible. The shot on the right was shot in silent mode, which introduces huge rolling shutter on the X1D, and distorted the trunks and branches. Admittedly, I was very lucky that it worked out (Lightroom is excellent for this). Rushing after my family made another attempt totally impossible. But … what if it didn’t ? Who cares? It’s more important to try something original, get a few wins and take a few losses than follow rules in a book that all lead to identical looking images.
You see where this is going, right? I know you know. I know you know I know you know.
Dark foreground, bright sky. Wet weather. Good luck keeping those Lee filters dry. In my case, good luck finding the time to set up a tripod. What this photo doesn’t show is my leg horizontally blocking the family trolls that wanted to move on. 2 shots, 4 seconds, is what I got. Aperture priority, auto exposure. Two frames, two different exposures.
And, guess what. 99.9% of the time, Lightroom’s algorithms produce a much lower visual signature than a ND filter does. I can see no downsides.
Here is the same photograph in colour. The b&w post processing isn’t hiding any hideous band 😉
And here’s another.
The photograph above also demonstrates the benefits of stitching for 3D pop. The Brenizer Effect is an extension of this, and while that can be a bit gimmicky, adding a bit of 3D this easily is not to be sniffed at.
Another advantage of stitching, aptly pointed out in a comment, is that it approximates a curved sensor. Get yourself a nifty fifty. Even better, an old nifty fifty with gorgeous character and flaky corner correction and start stitching with a lot of overlap. You’ll get only the center goodness and produce images that are far less sterile than what modern glass can ever give you. All for 150 bucks 🙂 🙂
So, that’s not really 101. But, if you apply all those ideas in as many combinations as mathematics allow, you probably get close to a dalmatian puppy number of ways to break the rules creatively. And that’s a lot of fun, in my book.
Got any other ideas? Objections? Lauding? Looking forward to reading you!
Never miss a post
Like what you are reading? Subscribe below and receive all posts in your inbox as they are published. Join the conversation with thousands of other creative photographers.
Please log in again. The login page will open in a new tab. After logging in you can close it and return to this page.