Frank Capria
QuickTime and Windows Media 9 Series
The recent release of Apple QuickTime 6 and the upcoming release of Microsoft Windows Media 9 Series (WM9) portend significant changes to the digital media authoring and distribution landscape. Apple is betting on the open MPEG-4 standard to help advance QuickTime in the marketplace. Microsoft, not surprisingly, is sticking to a traditional proprietary model with its large installed base of clients and servers, and its scores of licensing agreements with third-party vendors. For content producers and distributors, what may be the most interesting shift in this ongoing battle is that, unlike previous skirmishes between Redmond and Cupertino for digital media dominance, neither foe expects the winner will be decided on the PC desktop. Although no one is using the phrase, it appears convergence has finally arrived.
QuickTime 6 was formally released in summer 2002, and its features and performance can be fairly evaluated. Windows Media 9 is still in beta, so for now, it can only be judged based on what can be gleaned from the upcoming public beta and what Microsoft and third-party developers are willing to state publicly. Thus a head-to-head comparison at this time is impractical and unfair, but looking at where these two powerhouses are heading can help production facilities plan for the future.
QuickTime 6
The most noteworthy feature of QuickTime 6 is its native support for the MPEG-4 standard. Apple press materials from the past two years have been quick to note the MPEG-4 standard is in fact built upon the QuickTime file format. Yet this doesn't mean QuickTime 6 and MPEG-4 are completely interchangeable-far from it. Myriad features in each format aren't accessible in the other. For example, QuickTime and MPEG-4 define interactivity layers and objects differently. A QuickTime 6 movie's interactive objects aren't accessible to an MPEG-4 player, as MPEG-4 BIFS (binary format for scenes) information can't be understood by a QuickTime player.
It helps to think of both the MPEG-4 and QuickTime 6 formats as huge content containers. Each format's specification defines the type of content that can be stored in its files, with the two formats sharing a large subset of features. Files comprised solely of items within that subset are compatible. This is what allows the QuickTime Player to play many MP4 files natively.
Currently most of the MPEG-4 content we see is being delivered over the Web, but Apple's Aimee Nugent, senior product manager for QuickTime, sees that situation changing quickly: "MPEG-4 is pretty much designed to go anywhere MPEG-2 has gone. You could theoretically have MPEG-4-based DVDs." She also anticipates development will happen rather quickly because the MPEG-4 licensing terms were set in July 2002 and full licenses will be available in fall 2002.
If MPEG-4 goes where MPEG-2 has gone, MPEG-4 may be used for broadcast and satellite delivery of video content as well. This can only help Apple solidify QuickTime's position as an authoring format among broadcasters and video producers. Of more interest to content developers and distributors may be where MPEG-4 is poised to break the bounds of MPEG-2. MPEG-4 includes specifications for many forms of digital content distribution, from cell phone audio codecs to HDTV and digital cinema projection (see "MPEG-4 Codecs Compared," by Ben Waggoner, June '02 DV).
If Apple's vision comes to pass, it's pos-sible that in the near future, editors will be outputting sequences to videotape, encoded for the Web, and ready for wireless devices from within a single Final Cut Pro project. Such an environment would surely help sell a lot of Macs. As Apple continues to expand its consumer electronics offerings, having content delivered in a QuickTime-compatible format will make the task of developing software powering consumer devices vastly easier and more efficient for Apple. It's not a huge leap to imagine Apple developing iPod-like PVRs and DVD players, reinforcing its "digital hub" strategy.
It's easy to understand Apple's logic in tying its cross-platform authoring and distri-bution format to a promising open standard. MPEG-4 has much going for it during the early stages of the battle against WM9. MPEG-4 has a family tree that includes the successful MPEG-2 DVD and transmission standard and the wildly popular MP3 audio format, so MPEG-4 will get a lot of attention just for showing up to the party. But significant questions linger. MPEG-4 spent many years getting from the drawing board and into commercial distribution. It now faces stiff competition in key areas from not only WM9, but also from DivX, Real, and other distribution platforms. Critics of MPEG-4 are quick to point out that MPEG-4 is no longer cutting- edge technology after languishing for years in the lab and through all of the well-documented disputes over licensing fees.
Most users will first become acquainted with MPEG-4 via Web delivery. Apple's first release of its MPEG-4 video codec for QuickTime is a serviceable Web video codec, but its quality doesn't compare with existing Web video codecs from DivX, Microsoft, Real, and Sorenson. Web content producers probably won't adopt a codec that delivers less picture quality and that may end up costing a lot in licensing fees, if their sites are successful.
Remember that Apple's MPEG-4 video codec is only a first-generation offering. It currently lacks two-pass VBR (variable bitrate encoding) capability and other features present on more mature video codecs. As third parties integrate MPEG-4 features into their applications and Apple continues to tweak the codec in the lab-"to add that special sauce," as Nugent puts it-we can expect improvements in the video codec.
The most pleasant surprise while I was working with Apple's MPEG-4 codec was its encoding speed. Casual tests in my studio consistently returned encoding speeds of 1.2 to 1.4 times realtime.
The two most commonly used MPEG-4 audio codecs are well developed. The AAC codec is touted as a replacement for MP3, and has the performance stats to back up the claim. It can support up to 48 channels and 96 kHz sampling rates. At low bitrates, its performance at compressing speech drops off significantly. For such applications, developers should use the CELP audio codec.
MPEG-4's licensing fees garnered some unfavorable attention earlier in 2002 when Apple held up the release of the QuickTime 6 public beta due to Apple's dispute over licensing terms with MPEG-LA, the licensing authority that represents approximately 100 patent holders of MPEG-4 technologies. After some behind-the-scenes negotiations and a public rapprochement, Apple agreed on the fees and released QuickTime 6 during summer 2002.
On the surface, the licensing terms to distribute MPEG-4 video streams don't appear onerous. Fees are only levied on paid subscription and advertising-supported sites with over 50,000 subscribers. There's a charge of $0.25 per subscriber and an additional $0.02 per hour of streamed MPEG-4 content. Fees are capped at $1 million. Not a lot of money is being made on streaming content these days, so even these relatively small fees may be a disincentive to a midsize content provider.
Translate these fees into the broadcast, cable, and satellite environments, where a typical American household may have the television on for eight to 12 hours a day, and MPEG-LA may be levying $5 per month, per household. With margins under pressure in a tight economy, this fee may be enough to get cable, satellite, and DVD distributors to look at MPEG-4's competition.
Windows Media 9
Long before the licensing fee flap, Microsoft made it clear the company was moving decisively into the consumer electronics and digital rights management spaces with the upcoming release of its digital media platform. Early in 2002, graphics chip makers Nvidia and ATI announced plans to embed support for DirectX and WM9 technology in their chips. This is a significant performance enhancement. Relieving CPUs of much of the burden of processing of large video and multimedia files, Microsoft technology is in essence enabling content creators to access twice the processing power of previous architectures. At nearly the same time, Pioneer announced it would support the WM9 platform in its upcoming Digital Network Entertainment series of home theater products (see Industry Buzz, Oct. '02 DV). This is Micro-soft's most direct move into Apple's digital hub space.
Microsoft is proud of the performance of its new codecs in the lab and in field tests. (As of this writing, the public beta of WM9 hadn't been released, so I wasn't able to perform any testing of my own.) The company claims a 20 percent improvement over the WM8 codec-meaning picture quality improves 20 percent at the same bitrate, or the same picture quality can be achieved at a 20 percent bitrate reduction. Microsoft claims a 2x improvement over MPEG-2 at D-1 resolutions.
The WM9 codec has been well received among key third-party developers. Industry leaders such as Adobe and Avid have announced plans that Adobe Premiere and Avid|DS will allow users to export WM9-format files directly. Avid|DS HD will enable the export of HD-format WM9 files.
On the audio side, the new WMA Pro codec will be capable of delivering a 5.1 experience streamed at rates as low as 128 kbps. The codec is capable of sampling all the way up to 24 bits at 96 kHz, making it a strong authoring format that can compete with AAC.
Although both Apple and Microsoft aggressively target the small and smaller screens of televisions, PCs, wireless devices, and PDAs, Microsoft is aiming for the big screen in a big way. Back in March 2002, six months before a public beta of WM9, Microsoft announced the new technology was breaking into the digital cinema arena and the company was specifically targeting independent filmmakers working with digital video formats. Citing the high cost of creating video-to-film recordings and prints for distribution, Microsoft presented WM9 as an alternative. Electronic distribution teamed with its Digital Rights Management (DRM) platform could be a very attractive alternative to traditional film distribution for independents. The same technology could also be used for delivering digital dailies.
Microsoft firmly believes DRM gives it the edge over all competing technologies in digital media distribution. Mike Aldridge, lead product manager for the Windows Digital Media Division, cites DRM's track record and client list as proof. "We're already in our second generation of software and we have tens of millions of transactions completed with it." He goes on to note studios such as DreamWorks and MGM are already using Microsoft's DRM to handle licensing of content delivered over the Internet.
Although Microsoft has been public about the need for, and its desire to be, a digital gatekeeper, it isn't clear the entertainment industry is willing to hand Mr. Gates the keys without the answers to some serious questions first. Microsoft was accused and found guilty of anticompetitive practices in the PC world, so many content producers want to ensure open and fair competition in the digital licensing space. They are wary about becoming beholden to a monopoly charged with guarding their intellectual property.
Conclusion
Perhaps what is most impressive about Apple's and Microsoft's current digital media technology push is that both companies are aggressively seeking dominance in places beyond the PC desktop and the Web. Neither technology is likely to change content production workflows significantly in the short term. Producers who acquire and author in QuickTime will probably stay there for the time being, even if the final delivery medium is going to be WM9. This may change if Microsoft takes the logical next step and merges the AVI and WM9 technologies into a single platform in subsequent Windows Media updates.
QuickTime's cross-platform capabilities continue to be a strength in QuickTime 6. By keeping the door to the Windows world wide open, Apple ensures producers who choose to acquire and author in AVI and other Windows-based technologies will have a clear path for distributing QuickTime 6/ MPEG-4 content.
For years, we heard about the inevitability of the convergence of the PC and the tele-vision. Now it finally seems to have happened. It will be interesting to see what becomes the next battleground in the on-going competition for digital dominance. Will Microsoft move aggressively to take some of QuickTime's authoring turf? Will Apple counter Microsoft's agreements with Pioneer and the chipmakers with some interesting consumer electronics plays of its own? For us producers, this aggressive competition will help open more channels to deliver our content to our audiences.
Frank Capria is the president of Kingpin Interactive, Inc., a consulting and design firm in Boston. He is a member of the summer master faculty at the International Film and Television Workshops in Rockport, ME, and a lecturer at Boston University.
Copyright 2002, CMP Media LLC