Seoul, Korea – June 24, 2013 – Chips&Media,Inc., a leading company in silicon video Intellectual Property (IP), has unveiled its innovative bandwidth saving technology to deliver premium 4K video content with merely a half memory bandwidth.
Now we are just in time for the industry’s next wave. Despite of persistent worries about lack of content, manufacturer’s response to growing consumer demand for higher resolution on larger screen drives widespread adoption of Ultra High-definition TV (UDTV). As pointed out at this year’s CES Show, more than 13 major TV manufacturers unveiled their UDTV sets at the show and began to push for a mass increase in the sale of them. According to the latest report from NPD DisplaySearch, UDTVs will be shipped with 2.6 million units in 2013.
One of the major bottlenecks to designing 4K solutions will be bandwidth. Basically Ultra HD(4Kx2K) is four times as many pixels as 1080p, which means that would take up four times more bandwidth. Fortunately HEVC(High Efficiency Video Codec)/H.265 would be a big help for bandwidth saving, which promises to drive bandwidth efficiency gains between 30 to 50 percent as compared to the AVC/H.264. Even with HEVC, however decoding up to 4Kx2K(4096x2304) pixels 30fps video stream would require a huge amount of bandwidth roughly 13.5Mbyte per frame only for write-back.
To address these bandwidth challenges, Chips&Media has actively implemented compact frame buffer technology - CFrame™- into its latest video IP, WAVE410, in order to maximize the benefits of HEVC’s bandwidth-saving capabilities without any quality loss. Chips&Media’s CFrame™ includes a frame buffer compression technique, which can minimize the significant amount of data transferred by using losslessly compressed reference frames and 2D smart cache limiting external memory access.
By employing CFrame™ technology, Chips&Media’s WAVE410 offers 50~60% bandwidth reduction in average for decoding 4K HEVC video stream compared to the assumed pure bandwidth. Considering the real bandwidth affected by various factors, it may achieve much higher bandwidth efficiency.
It is now available for HEVC decoder IP core, WAVE410 and will be widely adopted for future WAVE4 IP product.