High Efficiency Video Coding (HEVC) aka H.265 has had a tumultuous journey over the past year or two.
Back in January 2015 HEVC looked like a sure thing as industries seemed willing to adopt the codec as a standard replacement for H.264 codecs under a coalition of licensors. Later in July, it was in trouble with differing ideas regarding royalties and the development of the codec was put on hold. In December it was happily resuscitated, but during the in-between months, open source codecs were developed creating industry confusion and it seemed everyone was chaotically moving in different directions. Then, in the rarest of occurrences, most of the original groups involved, reversed course and converged toward a single standard, which almost certainly helped shove HEVC back on its intended path…or did it?
H.264 is an incredibly popular codec that forms the basis of most online video shown today. It was released way back in 2003 and, despite taking a few years to catch on, became the workhorse codec. This leads to an inevitable question: If it’s been working for so long, and is compatible with virtually every device consumers’ own today, why replace it? Simply put HEVC is just more efficient. The less data a codec can use to display an image without degrading its quality, the more efficient it is. If codec X and codec Y both display the same image quality, but the size of X’s file is 20 percent smaller than that of Y, then X is more efficient.
With the rapidly growing interest and demand for Ultra High Definition Television (UHDTV) services both in broadcasting and live streaming, the developments in video compression technologies have key challenges to face. The H.265/HEVC standard is considered as the coding technology of choice due to its enhanced compression efficiency, to help cope with the significantly increased data rates in UHDTV signals.
However, this superior compression performance of HEVC has come at the expense of increased computational requirements, which is estimated at up to four times that of its predecessor H.264/AVC. Therefore the development of real-time H.265/HEVC encoders for UHDTV, particularly the software implementations that can run on general purpose hardware is an ongoing challenge. It is interesting to note that Microsoft’s Windows 10 has H.265 playback, without the need for third party plugins, embedded in its system. Hardware-decode support will become more prevalent, and it’s likely that Intel will add some hardware support as a feature in sixth-generation Core processor. This will entirely solve the potential performance issues facing high-quality, UltraHD and HEVC video.
From a broadcast perspective, the International Telecommunications Union (ITU) has approved the H.265 video standard and the European Broadcasting Union (EBU) has decided to test HEVC encoders and decoders for live content transmission throughout the 2016 motor sports season. Open source developer and UK based start-up video compression firm V-Nova, have finalised the development of their Perseus encoding technology, considered by many as the current gold standard of HEVC.
They recently formed an Open Innovation consortium of over 20 global major industry players including Broadcom, the European Broadcasting Union, Hitachi Data Systems, Intel and Sky Italia. The firm claims to have at least halved the required bitrate for high video quality compared with HEVC and have achieved the same gain in encoding efficiency over its predecessor MPEG4/H.264. But the difference is that while HEVC belongs to the same technology family as H.264/MPEG4 and indeed MPEG2 before that, Perseus is based on radically new methods designed to avoid some of the previous limitations as well as improve performance.
But these are early days and the proof of the pudding will be in the eating. Many broadcasters and operators will want to see further evidence of performance and wait for proof that the technology is robust in the field and really does integrate with existing workflows.
