[M4IF Discuss] What is the REAL advantage?

[Craig Birkmaier]

At 11:09 AM -0800 2/6/02, Yuval Fisher wrote:
>  > Of course, it didn't stop MPEG-2 from becoming the most successful codec of
>>  all time (without any help from the PC). I see few reasons why it should be
>>  any different with MPEG-4.
>
>I disagree. I think it will be very different for MPEG-4. The existing
>investment in MPEG-2 will not go away. Content owners do not want to
>live on the bleeding edge at all. As happened with mp3, the hardware and
>consumer markets will follow wide spread adoption only, and this can
>only happen over PCs.
>
>The success of MPEG-4 depends on adoption on PCs. (Wireless is another
>argument, which I'll avoid now).

I've got to agree with Yuval.
There is a good reason why MPEG-2 succeeded in the consumer 
electronics space, and now in cable. The major players saw both the 
need for and the opportunity to make MPEG-2 a success. DBS and DVD 
could not have happened without a video encoding technology that 
would deliver the legacy 525/625 line interlaced video formats at the 
target bit rates. And these vendors were only to happy to play the IP 
licensing game; they created a bunch of new IP for MPEG-2, and they 
share in the royalty revenues.
It is also noteworthy that ONLY MPEG-2 MP   ML is a commercial success, 
and even here there is no interoperability...set-top boxes do not 
inter-connect with DVD players...it's the same old box per function 
CE mentality.
MPEG-2 did not succeed on PCs because of the licensing issues AND 
because of content management issues. The latter is still more of a 
barrier than the revised MPEG-2 license fees. Decoded MPEG-2 video 
cannot be handled like any other data in a PC, it must be separated 
and protected.
MPEG-4 offers a variety of benefits that will eventually play an 
important role in CE products, PC products and IP distribution 
networks. All of these markets are converging in terms of the 
underlying technology - but the barriers to prevent marketplace 
convergence are as impenetrable as ever.
Yuval is correct about the desire to protect the investment in 
MPEG-2. This extends well beyond the video codec. It helps to 
preserve the legacy of interlace (a barrier to convergence with the 
PC) and the control over the appliances that deliver entertainment 
content. But some cracks are beginning to appear in walls that the CE 
industry has built.
Program distributors are looking for more efficient codecs to deliver 
more content in fewer bits (and in a few cases higher quality in 
fewer bits). The new Moxi platform, introduced at CES, supports DBS, 
Digital cable, DVD-V AND Real Player, and has all of the hooks needed 
for applications delivered via IP networks. It uses Flash for the 
development of the user interface.
If DirecTV and DISH are allowed to merge, they will have a strong 
incentive to replace the existing MPEG-2 set-tops with a platform 
that uses more efficient video coding and a variety of IP based 
interactive applications. MPEG-4 could be the solution, but not with 
the proposed licensing model.
If MPEG-LA were to pursue a PC based "proliferation" model for MPEG-4 
it would stand a very good chance of displacing the big three, or at 
the very least, full support by the big three. Even with this it 
won't be an easy path, as fully conformant implementations of MPEG-4 
across multiple platforms will be a significant challenge from both a 
technical and business viewpoint.
Unfortunately, this discussion is centered only on MPEG-4 visual, 
which is but one of many codecs that could be used in a full 
implementation of the MPEG-4 composition model. There is far too much 
emphasis on MPEG-4 as an audio/video streaming format. The real power 
lies in the composition model and the impact it will have on the 
localization and personalization of digital media content.
-- 
Regards
Craig Birkmaier
Pcube Labs