At 1:26 PM -0500 11/22/04, Manfredi, Albert E wrote: >I'm not completely positive about this, but >isn't this article full of BS? Apparently not. > >To begin with, H.264/AVC is either a little >older than or about as old as WM9. The article >here states that "'Windows Media 9 is a few years >old and it's evolved a few times,' said Casanova. >'AVC is brand new -- it's just at the very >beginning of its quality and optimization curves.'" > >Bull. H.264, ex-H.26L, was already defined by >August of 2002, when it was described in >Communications Design Magazine and doubtless >many other sources. Not exactly. Yes, the ITU had been working on H.264 for several years. It was one of SEVERAL codecs, including WM9 that were submitted to ISO/MPEG for evaluation. Based on those evaluations ISO and ITU agreed to create the Joint Video Team, which then created MPEG-4 Part 10, aka AVC, aka H.264. The resulting codec incorporates a number of additional tools that were not part of the original ITU work or that of Microsoft. As for Cassanova's comments, I think that he has a valid point. Microsoft has had some of the top compression experts in the world working on Windows Media technologies since the late '90s. There are signifricant advantage for an engineering team that develops the core technologies and then works to optimize them in a real world product, versus the "documentation" of a set of tools that can be used for video compression. At the end of the JVT process (in truth it has not ended, as reported by Tom McMahon regarding the Fidelity Range Extensions), potential implementers had to take the spec and turn it into products. If Bert monitored the relevant MPEG developers e-mail lists, he would understand that many companies are just now beginning to understand the tools that they are working with. There have been many demonstrations of AVC implementations, but there are few shipping products today, as they are still being optimized. >Also, in 2001 MPEG joined with the ITU-T VCEG to >form the Joint Video Team (JVT), to bring forward >work on H.26L. I'm not claiming that Apple was >absent in any of these groups, but they were not >explicitly mentioned, either. Sounds like a large >ITU effort to me. As with all ISO/MPEG developments, a wide range of companies are involved. To this you can add those companies that focus their efforts on ITU activities, rather than participating in ISO/MPEG. I am not ware of any patents that Apple holds with respect to AVC; to the best of my knowledge they have not submitted any IP to MPEG-LA for the AVC patent pool ( but I could be wrong). But Apple has been very active in development of the AVC codec for QuickTime, and this work overlapped with the JVT work. Apple has developed a full software codec implementation of AVC; they demonstrated the HD version of this codec at NAB. It is likely that it will be rolled into the next release of QuickTime (it might have been released sooner, were it not for the ongoing delays related to the development of acceptable licensing terms by MPEG-LA and Via. > >I'm not disputing that QuickTime uses MPEG-4 file >formats, but it's news to me that MPEG-4 file >formats and AVC were Apple inventions, and that >AVC is any newer than WM9. I think the cart got >placed in front of the horse? Here's what I'm >referring to: Apples and oranges (literally). ISO/MPEG started working on the MPEG-4 container format back in the late '90s, when I was actively participating in MPEG meetings.They published a request for technologies for the container format Apple submitted the QuickTime File Format; Microsoft submitted their streaming video format (ASF). Here is a February 1998 press release from Sun Microsystems announcing the decision by ISO/MPEG to use the QuickTime File Format as the basis for the development of the MPEG-4 file format: http://www.sun.com/smi/Press/sunflash/1998-02/sunflash.980211.1.html As you will see, the QuickTime File Format was supported by a number of companies that were working on the MPEG-4 standard. What is particularly relevant is that there was a strong correlation between the QuickTime File Format and the requirements established by ISO/MPEG. It is important to remember that MPEG-4 is MUCH MORE COMPLEX than just audio and video codecs and a transport protocol (e.g. MPEG-2). It provides a complete architecture for the composition of digital media content in the receiver, just as QuickTime is an architecture for digital media content, not just a collection of codecs. The following page on the Apple web site provides a quick overview of the MPEG-4 architecture and its relationship to QuickTime. http://www.apple.com/mpeg4/ Bert seems to have fallen into the same trap as many others by inappropriately thinking of AVC as MPEG-4. AVC is nothing more than a new video codec that is PART of the MPEG-4 architecture. With respect to the MPEG-4 file format, AVC is just another addition to a table of supported codecs that can be included inside the file container as part of an MPEG-4 composition. > >"'The investment we made in the MPEG-4 standard is >paying off incredible dividends for QuickTime and for >Apple,' Frank Casanova, Apple's director of QuickTime >product marketing, told MacCentral. 'Allowing the ISO >the use our file format has turned out to be the best >decisions we could have made.'" > >Is this for real? Anyone know? Yes Bert, it's for real. There are some differences between QuickTime and implementations of MPEG-4 that use the MPEG-4 file format. Apple has control of the technologies that it chooses to incorporate into QuickTime. There are MANY codecs and object types that are included in QuickTime that are not part of the MPEG-4 toolbox. It would be more accurate to say that MPEG-4 is a subset of the content that can be stored in the QuickTime File format. The marketplace reality is that a significant group of companies have been working together to create open standards for digital media content under the banner of the Internet Streaming Media Alliance. The notable absentee is Microsoft. Apple decided to put its eggs into the ISO/MPEG and ISMA baskets in the late '90s, and has been turning these efforts into real products for several years. Much is made of the "proprietary" nature of the Apple iPOD, however, the only thing that is proprietary is the Digital Rights Management component - the underlying audio compression technology comes straight out of MPEG-4 in the form of the Advanced Audio Codec (AAC). I store all of my iTunes music using the AAC codec, which means that I can easily move this music to an iPOD, if i decide I need a portable music player. What we can hope is that with AAC, AVC and the rest of the tools that are incorporated into MPEG-4, we may soon start to see more standards-based efforts to create and distribute RICH digital media content that is optimized for various delivery platforms. This is where QuickTime may finally come into its own. It would be interesting, if in the end, it is not the OS, but the digital media architecture that turns out to be the real crown jewels of the personal computing revolution. QuickTime can EASILY be used at the core of the digital home entertainment systems of the future. It already runs across the MacOS, Windows, Unix, Linux, Solaris and Irix. Consumer Electronics products like Set-top boxes are a logical extension. Regards Craig P.S. I wonder if Bert noticed what happened to Apple stock yesterday? Two analysts raised their strike point to $65/share and $100/share respectively, based on strong sales of iPODs and more important, improving sales of Macs, as more and more people begin to understand the many advantages that Apple offers for home entertainment computing. I was amazed to open the business page of the Gainseville Sun this morning to see two stories about Apple. One reported the 11% bump in share price yesterday, after the announcement of the analyst reports. The other was part of a holiday technology buyers guide series by Gannett News service. The story recommended two personal computers for this holiday season - the "affordable" Apple iMacG5 and a low cost E-Machines PC. ---------------------------------------------------------------------- You can UNSUBSCRIBE from the OpenDTV list in two ways: - Using the UNSUBSCRIBE command in your user configuration settings at FreeLists.org - By sending a message to: opendtv-request@xxxxxxxxxxxxx with the word unsubscribe in the subject line.