Standing at the base of a stanchion on the St. John’s Bridge in Portland, Oregon, I realized that I’d really like to shoot with an 8×10 camera. In order to frame the picture the way I wanted, I had to use a 14mm lens, which gave too much distortion and foreshortening. I wanted the look of a 50mm lens from where I was standing. That’s when I realized that a larger sensor gives a larger window on the world without using a wider angle lens. Not only does the larger sensor give a bigger view, but using a longer lens also allows a shallower depth of field.
I’ve used almost all the APS-C sized (cropped) sensor cameras on the market, as well as most of the full frame cameras. The reason I like full frame cameras is that I get more in the frame without moving back and without using a wider lens. The reason I like cropped sensor cameras is that they are smaller, lighter, and much cheaper (these are the same reasons I’d like to get into shooting micro four thirds bodies).
But I’d really like an even bigger sensor. To me, this expanded view of the world is the best reason to shoot medium and large format film bodies. Oh, for the means to own them all.
Here’s a technique that mimics a larger film/sensor plane, but can be done with any small camera, even a point and shoot, and I employed it at Arches National Park the other day. Ryan Brenizer has popularized a similar technique in his portraiture, so this may not be totally new to you.
Tools Required:
- A tripod with a panning head (even better if it has degree marks for measuring rotation)
- Camera with at least a 50mm lens–the longer the lens, the shallower the depth of field that can be achieved, and a prime will also be better
- A remote trigger is nice, but not necessary
- Photoshop, or other software that stitches panoramas
First of all, this method makes a really large image file, so it’s good to have a subject worthy of the effort. On the other hand, it’s definitely worth practicing before you find yourself sitting on a cliff in a national park at sunset with the light quickly fading. Simply put, you’ll make a grid of images and stitch them together as if it was a panorama. My image of Delicate Arch is a grid of nine images, three by three. The key is to have plenty of overlap between images. Normally, including 25% of the previous image is enough for Photoshop to create a pano. In order to maintain a more flat perspective in the image, I recommend 50% overlap. If your tripod head has degree marks it’s simple to reproduce the right amount of overlap without even looking through the view finder.
Level the tripod and choose your lens. It’s got to be level so that each frame remains in the same plane as you pan, or the final composite will not be square. My nine frame composite shot at 105mm includes a little more area than my single shots made with a 50mm lens. I used 105mm at f/2.8 to make the background a little soft. A longer lens at the same aperture would be softer still in the back. I imagine that if I used a 200mm lens I would have shot twice as many frames to make the same picture (I wimped out, however, and didn’t carry my 70-200mm the 3 miles to make this image; some of the micro four thirds lenses would solve this problem).
Some tips for shooting the grid:
- Shoot in rows, not in columns, and start at the top. Shooting across (left to right) and starting at the top helps ensure that the sky portion of your image will be the same brightness, and if it’s a little darker by the time you reach the bottom, it’ll probably look natural to have that kind of gradient.
- Use the built in horizon in your camera (if it has it). The tick marks on the indicator acted like the degree marks on the tripod head, allowing me to tilt the camera up and down the same amount between frames.
- Use autofocus to focus on the subject, then switch to manual focus so it doesn’t change between frames. Use manual exposure mode to make sure the brightness of the image doesn’t change, either.
- Shoot a frame with your hand in front of the lens at the beginning and end of your grid so that you can easily identify which frames to include in the composite.
Once I’d successfully hiked back to the car in the dark, I selected the frames between hand shots and opened them in CS6’s Photomerge. I used the automatic option and was quite pleased with the result. Shooting with a Nikon D800, my final image ended up over 113 megapixels. I’d recommend practicing with fewer frames–maybe a 2×2 grid–so that you don’t overwhelm your computer.
_______
Good information, but just as a reference, you mentioned that the gigapan only shoots in columns. Actually you can enter the setup menu and change it to: Columns, Rows – up, and Rows – Down. So you can tell it the order you want to pictures taken.
Yes. I updated post to clarify. I appreciate your catch.
Thanks, Colin! I can’t wait to try the Gigapan again–it was so simple.
Good article! Very similar to the “Brenizer Method”
http://blog.buiphotos.com/2009/07/the-brenizer-method-explained-with-directions/
I am not sure it’s similar other than in intent but the article is interesting.
Where as the brenizer method is used to create an abnormal DOF you would normally see, this method aims to get more pixels to work with. To be honest… and in my opinion… this is a very dull shot of the Delicate Arch. Maybe post the color version?
Trent, this method does provide the shallow DOF of the Brenizer method–105mm at f/2.8 will do that, and that’s what I was after. I don’t want my entire world to be in focus in every landscape image–it’s not the way my eyes see the world. As far as the image, I agree that it’s not the best Delicate Arch image ever made, but it serves to illustrate the point. I’d love to practice enough to make this sort of image quickly enough that I can beat the changing light before I’m done shooting the grid…
Great read. I have been experimenting with this on and off. When I first started doing this I was zoomed all the way out. This compounds the distortion from each photo. I found it best to be at least at 24mm on a cropped sensor. What is the minimum zoom on a full frame sensor to prevent distortion?