My current camera of choice is a Canon Rebel XS — mainly because it’s all I have available at the moment. My external intervalometer is a Vello ShutterBoss (model RC-C1) — a little on the cheap/flimsy side, but it has served me well so far.
I typically shoot high-quality JPEG with all the camera settings locked down to match the subject. The images come in at 3888×2592, or a hair over 10 megapixels. The image sequence is imported into QuickTime 7 (yes, some of us still use it) where it is then re-exported — in its full, native resolution — as a single video file.
I was in Lower Manhattan in 2012 when Hurricane Sandy struck. As a result, my entire neighborhood lost power for about four days. On the second dark night, after the initial thrill had worn off, I set up my camera to shoot the interesting patterns that the passing cars’ headlights created.
I let the camera run until its battery drained completely. I “flew by the seat of my pants,” as they say, having no idea what I had until I was able to view the final video. An interesting experiment. This was also the shoot that made me painfully aware of how many “hot” pixels my camera’s sensor has.
I set up a time-lapse rig in my kitchen to capture images of various things I had lying around as they dissolved in hot water.
That last one was a joke.
For a time, I had an old iPhone 3G pointing out the window of my apartment. It was running InterCam and uploading images from the built-in camera to an FTP server I was running on my desktop computer. The setup ultimately proved infeasible for any sort of long-term usage, due to the crashiness of the app, the not-quite-reliable nature of the WiFi connection, and the fact that FTP servers reeeeally don’t like having tens of thousands of files pile up in a single directory.
I’m half-seriously considering repeating these experiments with my old yet somewhat-more-reliable iPhone 4.
Technically, I did not take any of these pictures. But using a log of GPS coordinates from a road trip a friend and I took to California, I was able to programmatically scrape the Google Maps Street View tiles and construct a time-lapse video of our journey.
The technical details of this video are explained in excruciating detail here.