[MPEGIF Discuss] [systems] [discussion] Who framed MPEG-4 within H264 ?

[igino manfre']

Hi people,
It seems me that we are forgetting that MPEG-4 is NOT ONLY a visual
compression scheme.
Really it is all but BUT NOT ONLY a visual compression scheme.
I presume you all agree.
If we consider the subjects of this newsgroup, in the last two-three years
the focus migrates from the visual (who was in or out MPEG-4) to the MPEG-LA
licensing schemes (that could better defined "mpeg-4 suicide") and again
back to the visual (h264). It can be also felt a growing interest from India
and far east.
In this landscape, I find almost no mail related to the "system" of MPEG-4,
and I do not think because they are better explained.
Two years ago there had been a lot of development of SDKs and authoring
systems in the area, many of which used to work (even fine). Probably all
these have been commercial flops since now at the exhibitions no-one is more
pushing fully MPEG-4 based solutions. All and only H264. Does it means that
MPEG-4 (in its complex) is dead ?
Really, for what I know, excluding the many experiments and demo targeted to
demostrate the power of object encoding and multiplexing within MPEG-4
systems (meanly two-three years old), I do not know any "true" (i.e.
marketable) application of "true" MPEG-4 after the part "-2".
Today the entire world is fashioned by H264. Its astonishing 2-3 Mbps 720P
video will probably be the first "true" MPEG-4 part really diffused on the
Earth. All the broadcast encoder manufacturer are today able to show their
h264 solution...
But the broadcast usage will probably be a terrible mix of H264 ES video &
MP3 or AC3 audio over MPEG-2 transport. The goal is to "squeeze" enough the
required bandwidth to allow the deploiment of high definition TV over cable
modem. To make this mix dirtier, probably these broadcast will carry OCAP or
MHP/OpenTV interactivity.
Does not it sound a little blasfeme ? What do you think about ?
The diffusion of MPEG-2 STB is not a key factor, since the current STBs must
be replaced in any case to decompress the H264. But a STB that will be able
to decode H264 could have enough computational power to implement the entire
standard.
Really it is an architectural problem since if the entire decoder will be
put in a dice of silicon any further expansion could not be possible...
An observation I expect is that as "the market trains the research", as well
it trains the newsgroups. But maybe the locomotives (or who designs the
railways) do not sufficiently know what COULD be available...
Are we really sure to have done the right work ?
Best regards, Igino Manfre'
-----------------------
Igino Manfre'  - igino.manfre tiscali.it
Broadcast Video Consultant
Mobile (+39) 3358235346
Home   (+39) 0650916416
13, Largo Nearco
I 00124 ROMA - ITALY