Netflix is the undisputed king of internet video streaming in the world, boasting over 93 million members in over 190 countries. And these subscribers stream about 11 billion hours worth of content each month, imagine that!
That’s an absurd amount of internet traffic. In fact, 37 percent of all downstream internet bandwidth is due to Netflix streaming alone.
To give you an idea how much data bandwidth Netflix uses, its “High” usage setting runs through 3 GB per hour when streaming High Definition (HD) videos and 7 GB per hour for Ultra HD while the “Medium” setting only uses 0.7 GB per hour for a Standard Definition (SD) video.
In this age of mobile data caps, if you’re hoping to Netflix-cheat and continue your Walking Dead marathons while on the go, this can be an expensive way to entertain yourself on the grocery checkout lane.
Netflix has been diligent in solving this problem for mobile video streamers. The company recently introduced offline playback downloads and even SD card storage for Netflix rovers who just can’t stay put.
These are all welcome perks but wait, it gets better!
The streaming giant just announced its new technology that can significantly reduce mobile data usage without compromising the picture quality.
Dubbed the Dynamic Optimizer system of video compression, Netflix promises to reduce the data cost for its subscribers who love watching on-the-go.
How does Dynamic Optimizer work?
With this new video compression technique, artificial intelligence will analyze each movie’s scenes and adjust the compression accordingly for more efficient data usage and uniform video quality.
For example, this new system will compress an action scene filled with fast, complex images less than “quiet” scenes that feature static backgrounds.
Currently, Netflix’s current mobile video compression depends on bandwidth speeds and not for specific content, meaning every scene is squeezed uniformly depending on the connection. In effect, action scenes will appear grainier in low-bandwidth connections.
With Dynamic Optimizer, the compression will depend on what’s happening on the screen instead, optimizing the stream scene by scene, which will lead to less data usage overall.
This new technique is reported to be a result of a collaboration between Netflix, the University of Southern California and the University of Nantes. Armed with thousands of shots judged by hundreds of users, data was acquired to teach an algorithm for quality detection.
Speaking before journalists at the Mobile World Congress (MWC) in Barcelona this week, Netflix’s VP of product innovation Todd Yellin stated that this will reduce buffering and is all about “getting more bang for our bits.”
“Whatever the best picture we can give you with whatever your bit rate is, that’s what we’re going for. An HD picture for a lot less,” Yellin said.
Netflix’s Dynamic Optimizer system is set to be rolled out in a couple of months.