Chinese scientists claim neural network tech unlocks 10,000X speedup in optical fiber bandwidth

Optical fibers
(Image credit: Shutterstock)

Earlier this month, Chinese researchers from the University of Shanghai revealed that they had found a way to increase current-generation fiber optic speeds by orders of magnitude—that is, by 10,000 times (considering the current 100 Gigabit standard of fiber optics, this could be as high as 125 Terabytes per second) by overcoming the surprisingly strict bandwidth limitations that come with standard networking using optical fibers [h/t Interesting Engineering].

Those already somewhat savvy with networking, fiber optics, or both may be surprised to hear that these speeds are even possible, but also that fiber optics were leaving this much potential on the table when we already understood them to be the fastest data transfer method. You can't change the speed of light, and light's as fast as anything gets in the universe— so what trickery is at play here?

But it's important to recognize that this isn't technically a speed increase—it's more of an exponential bandwidth increase. So, while a future of faster loss-less downloads and streaming media does become more possible with technology like this, you're still probably going to have 200-300 pings trying to game overseas and get kicked for your troubles. Light may be fast, but it's not skip-into-the-future-past-all-existing-server-relays fast—low latency will remain somewhat sensitive to distance for a long while into the future, perhaps forever.

Christopher Harper
Contributing Writer

Christopher Harper has been a successful freelance tech writer specializing in PC hardware and gaming since 2015, and ghostwrote for various B2B clients in High School before that. Outside of work, Christopher is best known to friends and rivals as an active competitive player in various eSports (particularly fighting games and arena shooters) and a purveyor of music ranging from Jimi Hendrix to Killer Mike to the Sonic Adventure 2 soundtrack.