During November and December of 2012, I had the pleasure to film several weeks of construction time-lapse in Gibraltar for SoEnergy International (formerly Energy International). The time-lapse footage was included in the following piece, produced by Roar Media in Miami, FL.
Here’s a version with an added voiceover and project overview:
When a casual viewer sees something like this, they may assume that there was a large crew involved with the production, and that they were able to operate in sterile, controlled environments. In reality, the story behind the time-lapse is far more interesting.
In mid-November I got an email from Roar Media, an advertising agency in Miami, Florida, asking if I’d be available to fly to Gibraltar the following week to film a construction time-lapse. I emailed back and said that I might be able to make myself available, and mentioned some specific numbers in terms of what the budget would need to be.
I learned that SoEnergy International installs temporary power plants in locations experiencing immediate power shortages. Gibraltar was looking at the looming reality of not having enough domestic energy to power the whole city/nation/territory during the holiday season, and SoEnergy was hired to install a 15MW power plant within about three weeks (start to finish).
They called and we talked through details, I wrote a bid, and it was accepted the Tuesday before Thanksgiving. I left at the end of that week, just after Thanksgiving, and had precious little time to prepare before the ships arrived and the crew began construction on Tuesday of the following week.
In an ideal world, I would have leased, bought or built a custom weatherproof time-lapse enclosure to hold my DSLR camera and intervalometer. In an ideal world, a sturdy steel post would be driven into the ground at the ideal location, complete with AC running into my enclosure, and I would have images transmitted via a T3 internet connection to a remote server (as well as saved to an internal memory card), creating an instant backup of images that could be easily retrieved without touching the camera enclosure (and risk a bump in the time-lapse image).
Several consumer-grade construction cameras, bird-watching cameras and hunting cameras are available for a few hundred dollars, but the image quality is abysmal. I wanted to shoot the entire three-week timelapse on a DSLR (in 14-bit raw images), so I could have the maximum resolution, dynamic range and color data. As far as I know, nobody had done anything quite as ambitious with tone-mapped 14-bit raw DSLR images, so there were very few sources of inspiration other than ingenuity and panic.
The client wasn’t sure what kind of site access I would have, nor where exactly the camera would be oriented (I didn’t want the camera to face East-West, for example, where the rising or setting sun would create lighting troubles for the construction scene). They didn’t know how readily available power would be, and I only had about a day to set up. Since Gibraltar is relatively remote and has a small population, if I needed any specialty camera gear I would have to bring it with me. Given the extremely short advance notice and business closures during the USA Thanksgiving holiday, mail-order also wasn’t much of an option. It was a daunting set of circumstances to face, but the plane tickets were in hand and the company was excited to see the results.
I loaded my luggage with dozens of pounds of random gear, ranging from Gaffer’s tape to C-clamps to solar panels to large-capacity battery cells, as well as three DSLR bodies with their corresponding battery grips and AC adapters. I brought all of my lenses, not being sure if I’d end up shooting wide or medium (or telephoto, from a nearby apartment complex), as well as hundreds of GB in memory cards, portable hard drives, and a few styles of intervalometers to perform the actual time-lapse photography.
Jet-lagged after the series of flights (and two trains, a bus and a taxi) between Seattle and Gibraltar, I showed up at the job site on Monday morning. After a few hours of assessment, I determined that there were two good locations to film from: one was inside of the restricted-access port authority office and the other was on the top of a privately owned industrial property. The helpful contacts in Gibraltar scrambled to secure permission for one of the sites, and in the end I was able to use the port authority office, shooting through a window.
Of all possible circumstances, this was far from ideal. There was no great place to mount the camera, and there wasn’t room to mount a tripod in the narrow walkway next to the window. I hadn’t brought any sort of DSLR-capable suction cup mounts with me (I remedied that lack of gear shortly after I returned home at the end of the job), but I did have a small clamp that was designer to mount much smaller cameras, which was able to hold the DSLR upside-down from the handle used to open the window. Coupled with a GoPro suction cup mount supporting the lens from below, I secured the camera to the window.
It felt very secure (which was surprising to me) but I needed it to be absolutely rock-solid for several weeks without even the slightest camera movement. As a third for of camera support, I attached some nylon cord to other places in the room and created a cradle that the camera also hung from. I put rubber and other types of padding all around the camera, then Gaffer taped the heck out of the whole thing.
My backup camera was a GoPro Hero3 (barely released at the time), mounted directly above the DSLR. I used a plastic mount for the GoPro, then taped it thoroughly to the glass, also supported by a cord cradle and some other methods.
I attached USB and power cables to each, and ran them to a nearby AC power outlet, taping down each cable to prevent snags. My plan was to return to the job site daily to retrieve the data to my laptop via the usb cable, then back it up to two external hard drives before I left the site each day.
When the installation was complete I put a few black plastic bags over the whole thing to prevent reflections on the glass from inside the office during the nights (since the construction timeframe was so short the workers planned to often work through the night, so the time-lapse ran 24 hours each day).
I started the cameras right away and returned every few hours to analyze the footage until I was satisfied that the setup was strong enough to get the job done. Aside from one lower corner of the black plastic that became detached partway through the construction period, giving me a few hours of office reflection in the far corner of my frame, the setup proved to work flawlessly. (When I cropped the 4:3 images to 16:9 for the final product, the reflection in question was out of the 16:9 crop area so I was saved from having to reconstruct something using After Effects compositing.)
Filming the Vessel Arrivals
Once I knew that the primary time-lapse camera was in place, I saw the opportunity to also film time-lapse footage of the arrival of the two vessels that carried the components to the power plant. I had two DSLRs available, along with a slightly-older GoPro Hero2, so I went to work as the ships arrived.
The docks where the ships arrived had very restricted access. My gear was thoroughly checked and my credentials as a contractor were confirmed each time before I was allowed to enter. I wasn’t able to bring two tripods in, so I brought the lighter of my tripods as well as a few crazy clamp rigs I had put together using a combination of specialty camera equipment gear and hardware store parts. Improvisation was often critical to get the right shot. Below you can see some photos of bizarre GoPro camera placements. In one, I used a yellow weight that I found on site, and in the other I used a folded tripod on its side, with the pan handle extended to keep the tripod from rolling.
During a slow moment before the first ship unloaded its cargo, I asked the crane operator if it would be possible to mount a small camera to the top of his crane house. He said that I’d be welcome to do so, but I needed to do it quickly and not get in the way.
Once I got up there with my GoPro and iPhone, I installed the camera, snapped a quick photo of the Gaffer’s tape-heavy installation (I used a clamp and the GoPro tripod mount to secure it to a metal bar), and hustled down to the ground. The beautiful thing is that while I watched the crane move over the next hour or two, I wasn’t worried about the camera. If it were to fall down, it had a good chance of surviving (that same GoPro camera had spent time at the bottom of a lake and lived to tell the tale… or at least it recorded the sights until it went dead). If it broke, I could replace it for a few hundred dollars. My biggest concern was that it would somehow work itself loose and I’d ruin my one chance at getting the shot that I wanted. Thankfully, it worked out great.
The following couple of weeks were fairly mundane in terms of photography: I went to the job site in the evening, retrieved the data, backed it up, went home, and did the next thing the following day. During key phases of assembly, I’d film and photograph down in the job site to provide B-roll for the final video. It was also fun to rush edit some photos to the Miami Herald for publication while the project was underway.
As usual, a large portion of the media I generated remained unused, so I put together a fun compilation of some additional time-lapse and video shots, most of which were shot using a GoPro:
What would you have done differently? Hindsight is 20/20, and it’s generally easier to solve problems when job and environmental stresses aren’t adding pressure. I’d love to know what you would have done, and what your approach would have been.
How would you have gotten better footage? How might you have installed the cameras differently? Would you have taken a different approach at any stage of the project? Leave a comment and let’s discuss it!