[MPlayer-dev-eng] [PATCH] [TEST AFILTER] Encode ac3 at runtime using lavc

[Ulion]

2007/11/21, Reimar Döffinger <Reimar.Doeffinger at stud.uni-karlsruhe.de>:
> [...]
> > The real AC3 format not include iec 61937 header and padding zeroes,
> > with a 61937 header, the format should be IEC61937 or SPDIF I think.
> > So the current AC-3 stuff totally need refixed.
>
> I think it is currently not possible to support non-IEC AC3 without
> adding AC-3 parsing code everywhere, whereas removing the header is
> trivial.

Except dump it out by ao_pcm, or in ao_mpegpes.c it indeed mean to use
raw AC-3 stream instead of IEC's one.

> And having the same format ID for e.g. AC3 and DTS would create problems
> as well.

You mean currently? If introduce IEC or SPDIF stuff, it should bit-or
combine with the inner raw audio format flag (AC3/DTS/WMAPRO/etc) to
mark the format.

> So the only thing that is missing IMO is documenting that the AC3/DTS
> etc. formats actually mean those wrapped in IEC-headers.

Currently yes, but it's better confirm what ao_mpegpes.c mean to use
the AC-3 format.

> And a function that builds the IEC header from the af_data_t struct and
> whatever else is needed probably would be good as well.

If currently need it, or in the future make a filter to add the
header. Filter is a little hard to see frames than decoders, but still
can  be done.

btw: any suggestion for generate the last ac3 frame of at the audio
track end, it has not enough bytes to triger the encode action in
current code.

-- 
Ulion