Is 99% loss-less compression of audio and video files even possible? Apparently it is
Over the first half of 2015, TheServerSide has had its favorite freelance writer, Barry Burd of Java Java Programming for Android Developers For Dummies">Programming for Android Developers For Dummies fame, covering what's going on with Google's Solve for X program, largely from the standpoing of the North New Jersey Google Developer Group. Not familiar with the Solve for X program? Here's how Mr. Burd describes it in his article entitled Google's Solve for 'X' program goes into high gear:
The ambitious Solve for X initiative brings together entrepreneurs and experts to help solve the world's most pressing problems. On the initiative's website, Google compares Solve for X projects with the moonshots of the 1960s and 1970s: Each moonshot project addresses a big, global problem, with a radical new solution/perspective/approach and leverages a breakthrough in science or technology.
A wide range of new ideas
There are plenty of interesting ideas festering under the Solve for X umbrella, ranging from saving the environment to creating new medical diagnostic technologies, but from the standpoint of a developer and IT professional, it's the XLABS project that asserts that they can achieve 99% compression rates on audio and video files that seems most interesting, if not a little bit impossible. So when given the chance to investigate the technology a bit deeper, Barry Burd jumped at the opportunity.
Mr. Burd chronicles his interaction with XLBAS much better than I ever could in the article entitled A new approach to condensing data leads to a 99% compression rate. The basic takeaway from his experience was that he shot a video on his phone, he provided XLABS the file, the file was compressed at a ration of 460 to 1, and then the compressed file was expanded to produce the exact same, original raw video file that had been created moments before.
Is this a pig in a poke?
From my understanding of how compression works, that shouldn't be possible. In the article describing his experience, Barry provides a simple description of how most compression technologies work, and again, describes why there is always a hard limit to just how small you can squeeze a file. But of course, that's what Solve for X projects are all about - thinking outside of the box and coming up with solutions that nobody ever thought possible.
So, what do you think? Is a 460 to 1 loss-less compression of a video file all smoke and mirrors, or are we looking at a new era of data compression that will change how organizations like Netflix and Amazon stream data?
Give Barry's article a read:
I'd love to hear your opinions on exactly what you think is going on with XLABS and this revolutionary new technology.