[Mp4-tech] H.246 on DVB

[Gary Sullivan]

This would not be a "safe" assumption in general.  We discussed whether
to impose such a constraint on encoders, and decided not to.
The concept that we considered to be a viable alternative was to operate
the parsing process of a decoder in a serial fashion, and take advantage
of parallelism in the rest of the decoding process but not in the
parsing.  Designing a decoder in this fashion prevents us from needing
to impose constraints on encoders to force them to be "nice" to
parallel-oriented decoder designs.
Best Regards,
Gary Sullivan
________________________________
	From: mp4-tech-bounces lists.mpegif.org
[mailto:mp4-tech-bounces lists.mpegif.org] On Behalf Of Stephen Henry
	Sent: Friday, July 22, 2023 6:28 AM
	To: mp4-tech lists.mpegif.org
	Subject: [Mp4-tech] H.246 on DVB
	Hi all,
	I have a couple of questions, could someone please help me with
them?
	I'm interested in using H.264 for HDTV and higher resolutions.
To me, it seems like the natural way to enhance a hardware accelerator
for H.264 is through parallelizing the system using multiple slices. For
example, four independent encoders can encode 4 quarters of the one
frame in a quarter of the time of just one. What I wanted to know is
whether it would be safe to assume whether this can be done at the
decoder as well? I was thinking that it is unlikely that for broadcast
H.264, high resolution frame would need to be partitioned into slices to
be encoded in real time and therefore it would be safe to assume that
such parallelism is available at the decoder as well.
	Does anyone know whether the DVB standards impose a number upon
the minimum number of slices that a decoder can assume are present?
	Thanks,
	SCH
-------------- next part --------------
An HTML attachment was scrubbed...
URL: /pipermail/mp4-tech/attachments/20050723/76a6dcb5/attachment.html