New Chinese Involvement Could Trigger HD DVD Price Plunge
If ever there were a time for either Blu-ray or HD DVD manufacturers to play a trump card, now may be the time, and this could be the one: A consortium of Chinese university engineers and government officials, in cooperation with a Chinese video standards group that includes globally recognized manufacturers, plus the DVD Forum, have come to an agreement on a standard specification for a blue-laser disc mechanism and format specifically for the Chinese market.
The new group will be called the China High Definition DVD Industry Association, or just CHDA, and this is not the last you will hear of it.
The agreement announced early this morning US time is critical, because up until now, China has been reluctant to participate in the high-definition video disc industry unless it had an opportunity to bring its own video encoding standard to bear: AVS, a codec which incorporates elements of MPEG-2, but is otherwise different from -- and some argue, better than -- other MPEG encoding standards and VC-1 in important respects.
Now, the DVD Forum's involvement has evidently made it possible for Chinese manufacturers to produce components that play blue-laser, high-def discs using the Chinese national standard, though which are fundamentally compatible with HD DVD with only minor adjustments.
The upshot here is this: The same country that has literally upset the LCD TV industry on its ear in just the last year alone, now has the specifications it needs to do the same with high-def video discs. While it makes so-called CH-DVD players for the home market (the name is subject to change, the new consortium says), China can also produce HD DVD players for the rest of the world, at prices that can best be described as Chinese.
All of a sudden, the incentive for studios such as Warner Bros. to call a halt to exploiting new disc technologies its own engineers had patented, and for Paramount to jump ship and abort its Blu-ray support, may be becoming clear.
What a difference a macroblock makes
Since 2004, China has taken a public stand against becoming too involved in what it described as "Western-influenced technologies" in the development of high-def technology. The country opted instead to create its own national standard - government-owned intellectual property for video encoding - and go its own way. But it needed a clear technological distinction between its own way of encoding video and the rest of the world's.
It found one, most likely, in the person of one Professor Ishfaq Ahmad. Educated at Syracuse, for nine years Ahmad taught at Hong Kong University of Science and Technology's Department of Computer Science, before moving back to the US to teach at the University of Texas at Arlington. But while in Hong Kong in 2001, he and two colleagues produced a thesis on the subject of reducing the computing overhead necessary for encoding H.263 video. One of the systems he used in the test was a Sun UltraSPARC-1 workstation; the other was a 233 MHz Intel Pentium II-based PC.
The Hong Kong team's goal was to demonstrate algorithmic optimization techniques that could be applied to relatively low-power systems such as these, that could demonstrably improve encoding times by a factor of between 3 and 6. Part of encoding optimization involves being able to divide frames into few enough components so that algorithms can still correctly detect motion - and thus, the need to change - without overtaxing the processor.
Rather than divide pictures down to the most granular level - pixels - video encoding bunches pixels together into what are called macroblocks - cubes, in a way, made up of squares of luminance and difference data bunched together. Typically with MPEG, macroblocks using 4 x 4 pixel squares are used. But for Prof. Ahmad's team's experiment, they tried a technique that, based on others' theories, should not have worked.
From the Hong Kong team's 2001 paper: "A picture is divided into macroblocks, since such division results in more efficient coding. Each macroblock consists of four luminance blocks and two spatially aligned color difference blocks. Each of these blocks [is] of size 8 x 8 pixels. One or more macroblock rows are combined into a group of blocks (GOB) to enable quick resynchronization after transmission errors."
It's the 8 x 8 macroblock that was key to simplifying the algorithm and accelerating the encoding sequence.
Three years later, the Chinese government announced it would be developing its own video encoding/decoding standard, to be called Advanced Audio Video Encoding Standard in Information Technology, or just AVS. It was described thus:
"China initiated the new standard to reduce its reliance on technologies that use international standards heavily influenced by companies in Western countries. The standard will be used in the development of video players, stereos, interactive educational programs, surveillance systems and other audio and video products developed for the Chinese market."
Leading the development of this standard would be Prof. Ahmad, who by that time had joined the University of Texas at Arlington. There, he and the university would receive a $200,000 grant from Sun Microsystems, which was evidently impressed by the use of the UltraSPARC-1 in the Hong Kong team's tests.
Next: China's strategy as a high-definition IP licensor...