Uri Friedman is a staff writer at The Atlantic, covering national security and global affairs.
The Brazilian city like you've never seen it. Really.
Rio de Janeiro is called cidade maravilhosa, or the "marvelous city," for a reason. "How is one melancholic in Rio de Janeiro?" asks the Brazilian writer Tatiana Salem Levy. "You lower your head, but on your right side a hill rises up, majestic; on the left side, scandalous nature makes its presence felt; in front of you, the infinite line of the sea."
But Rio gets even more marvelous when you spend a couple days filming it with a $50,000, 80-megapixel camera—and a couple months whittling the footage down to a rhapsodic, five-and-a-half minute time-lapse video. The clip, as its creator Joe Capra told me, demonstrates "the extreme resolution of the camera and what the future may hold." (I suggest you watch the film in full screen.)
Capra, a Los Angeles-based photographer who was on assignment in Rio for Panasonic, says he took the exceedingly rare step of shooting the video in 10K resolution, which sounds pretty impressive but requires explanation. Resolution is usually expressed as image width (in pixels) multiplied by image height (also in pixels). High-definition TVs are typically 720 or 1,080 pixels high (the latter is referred to as Full HD, or FHD). Digital images that are roughly 4,000 pixels wide are considered ultra-high-definition, otherwise known as UHD or 4K. 4K is the Next Big Thing in television. 8K, or displays whose resolution is around 8,000 pixels wide, is the Next Big Thing after that. Hence Capra's excitement about capturing Rio in its 10K beauty.
Each raw frame in Capra's time-lapse is a staggering 10,328 x 7,760 pixels. But today's technology simply isn't equipped to transmit such vivid detail (plus, as Capra pointed out, there's debate over whether the human eye can even appreciate any resolution above 4K). Like most HD TVs, Vimeo (where Capra posted his video) can't support more than 1,920 x 1,080 pixels.
To overcome these problems, Capra divided his video into five shots, each of which consists of hundreds of still images strung together sequentially. Each shot sequence begins with his full-resolution footage scaled down to 14 percent to accommodate Vimeo's restrictions.
The next shot condenses the footage to 50 percent of its original size. "You are still seeing the 10K footage at its full resolution, but we have just zoomed into it," Capra explained. "The scaling does not change the resolution, only how much you see of the entire frame."
Then we zoom in again, to full scale (100 percent).
If that shirtless man on the balcony looks familiar; it's because he was barely visible in the 14-percent-scale image above:
"It is pretty amazing how much you can zoom into an image in 10K and retain all the detail," Capra told me. "Take any image off the web or from your own camera and zoom in really far, most likely it will look terrible. With this Phase One camera the shot looks great even at 100 percent."
Time-lapse videos have an ineluctable goshwow appeal to them ("5 years of the sun in 3 minutes!"), but they can also be recruited for more sober purposes like visualizing the normally imperceptible accretions of environmental damage and climate change. And Capra believes they have additional value. They allow you to see places "in a way that you can't see with your own eyes if you were just standing there," he said. "Time-lapse brings out all the hidden motion in the world."
This post originally appeared on The Atlantic.