Whether you're pumped for 4K or you think it's a passing fad (it isn't), the truth is it's going to be a long time before there's enough 4K content that we can forget about HD and Full HD.
Netflix may have been been the first to offer 4K streaming last year, but the company is very much aware that the lion's share of streamed content is still HD or Full HD. To that end, Netflix has announced a new algorithm to increase the quality of non-4K content, called "Per-Title Encode Optimization."
That sounds complicated, but what it boils down to is a smarter way of getting 720p and 1080p video from its raw, uncompressed source to your phone, TV, streaming box, or gaming console, using the least data possible.
Netflix stated on its tech blog that the algorithm is the result of several years of "significant research and engineering" into just how much visual quality users can appreciate and the maximum required bitrates to achieve that quality. Allegedly, the new algorithm is the answer to those questions.
The algorithm works by assigning different bitrates to TV shows and movies depending on how demanding the content is to encode/decode. The amount of detail in a show is a huge factor.
For example, Netflix knows that an animated show like Bojack Horseman, which doesn't feature much fast motion, requires less streaming bandwidth than something like Planet Earth. For each show or movie, the algorithm allocates only the minimum necessary bandwidth to allow it to look its best, using as little data as possible.
This is good news for almost anyone with a Netflix subscription. Assuming it works as intended, it means anyone watching movies and TV will likely be using less bandwidth than before, meaning more will be available for everyone else to browse the web, stream music, or even watch their own Netflix content.
Hopefully, other popular streaming providers like Hulu Plus and Amazon Instant will adopt a similar method. While the success of the new algorithm remains to be seen, it sounds like a promising way to improve streaming across the board.