SoylentNews is people
SoylentNews
Google's VP9 codec can (sometimes) outperform H.265/HEVC at higher resolutions:
Netflix, being one of the biggest video streaming services in the world, tested how efficient various video codecs are for a given level of quality. The company discovered that the royalty-free VP9 codec developed primarily by Google is almost as efficient as HEVC, and can sometimes be even better at resolutions of 1080p and higher.
[...] Both HEVC and VP9 promise about 50% bitrate savings for the same quality compared to h.264, but Netflix wanted to test for itself to see if this is true. Netflix sampled 5,000 12-second clips from its catalog, which includes a wide range of genres and signal characteristics. With three codecs, two configurations, three resolutions (480p, 720p and 1080p), and eight quality levels per configuration-resolution pair, the company had more than 200 million encoded frames. Netflix applied six quality metrics: PSNR, PSNRMSE, SSIM, MS-SSIM, VIF and VMAF. This resulted in more than half a million bitrate-quality curves. Netflix's unused cloud-based encoding infrastructure allowed the company to complete this large test in only a few weeks.
The company learned that previous research showing up to 50% bitrate savings for both HEVC and VP9 compared to h.264 turned out to be true. HEVC's x265 implementation outperformed VP9's libvpx for most resolutions and quality metrics. However, at the 1080p resolution, the difference was either much smaller (in HEVC's favor), or, in some cases, VP9 even beat HEVC in bitrate savings. The fact that VP9 performs better at 1080p or higher is not a major surprise, considering VP9 was optimized for resolutions beyond HD. Google is currently using it for YouTube, where all videos are encoded in VP9.
As the article notes, new codecs are coming. Here's a little more about VP10.
(Score: 5, Insightful) by TheRaven on Tuesday August 30 2016, @08:27AM
First, a lot of 'hardware' implementations of H.264 are nothing of the sort. They are very specialised DSPs, which implement various operations that form the core of a number of CODECs as single operations and then provide a firmware implementation of the specific ones. Often the difference between the SoC with the H.264 decoder and the one without is that the one without doesn't include the royalty payment to the MPEG-LA and so doesn't come with the firmware blob. If there's demand, then writing the VP9 decoder to run on the same DSP is a relatively small amount of effort (doesn't require modifying the hardware), gives the SoC vendor another feature checklist item, and doesn't cost anything per unit (no patent licenses, just a fixed cost for writing and testing the software).
Second, the two biggest video sites, Netflix and YouTube, both stream in a large variety of formats. Netflix currently supports a huge number of old embedded streaming devices and so adding another format to their system is not particularly difficult for them if it's likely to be a net win. If they provide it for desktop users and start saying to mobile users 'it's going to cost you more to stream from your mobile device than your desktop at high quality because it doesn't support VP9' then that's a big incentive for table makers to start including VP9 decoders and 'cheaper Netflix streaming' as an advertising buzzword.
The problem with Theora and VP8 (not so much Dirac, as it wasn't really aimed for streaming) was that they didn't have any compelling advantages. They weren't noticeably better quality (or the same quality at noticeably lower bitrates). The license fees from MPEG-LA are structured in such a way that, for the big sites, it doesn't save them any money if they switch half of their users to a different CODEC, it just increases their complexity: they'll still be paying the same royalty fees. If VP9 lets Netflix reduce their bandwidth by 5% by switching all desktop users to it, then that's a big cost saving all by itself. Theora and V8 were never able to do this.
sudo mod me up