[FFmpeg-devel] [PATCH] FFV1 specification: Merge of FrameHeader01() and GlobalHeader()

[Michael Niedermayer]

On Tue, May 19, 2015 at 10:24:02PM +0200, Jerome Martinez wrote:
> Le 19/05/2015 21:15, Michael Niedermayer a écrit :
> >On Mon, May 18, 2015 at 09:04:01PM +0200, Jerome Martinez wrote:
> >
> >>FrameHeader01() and GlobalHeader() have a lot of common fields
> >>and having a common function + default value for fields unused
> >>in previous versions is less complex and more coherent than repeating
> >>the common part.
> >maybe but calling the GlobalHeader FrameHeader is very confusing
> >and wrong
> 
> From my point of view, the GlobalHeader() is still a frame header
> because it still contains the same pieces of information about
> frames (frame width, frame height, frame bit depth...), and it is

In h264 its called a sequence parameter set, in MPEG-1/2 a
sequence header. Iam not aware of anything calling a global header
a "frame header"

version information is global its not a frame property
also w/h/colorspace/quantization tables and inititial states and all
that are global they are not frame properties
you could make w/h/colorspace frame properties in principle but

If you want to allow multiple w/h/colorspace, its probably better
to allow multiple global headers as in h264/h265 and put a index
in the frame header to choose one of the parameter sets
some of the tables are also large IIRC initial_state_delta is.
it makes not much sense for this to be anywhere but global

maybe none of these headers should have been called frame header
i dont know ...

[...]


-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

I have often repented speaking, but never of holding my tongue.
-- Xenocrates
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 181 bytes
Desc: Digital signature
URL: <https://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20150520/91d72a12/attachment.asc>