[FFmpeg-devel] Fwd: Interested in NEW Seeking API of GSoC09

[Daniel Verkamp]

On Sat, Mar 21, 2009 at 11:59 AM, zhentan feng <spyfeng at gmail.com> wrote:
> Hi
>
> 2009/3/21 Daniel Verkamp <daniel at drv.nu>
>
>> On Fri, Mar 20, 2009 at 9:03 PM, zhentan feng <spyfeng at gmail.com> wrote:
>> > Hi
>> >
>> > This is a forward mail from ffmpeg-soc list without reply and forgive me
>> > repost it here.
>> > If the qualification task is necessary, I'll choose "Implement a Vivo
>> > demuxer for ffmpeg" as a qualification task.
>> >
>> > ---------- Forwarded message ----------
>> > From: zhentan feng <spyfeng at gmail.com>
>> > Date: 2009/3/20
>> > Subject: Interested in NEW Seeking API of GSoC09
>> > To: FFmpeg Google SoC list <ffmpeg-soc at mplayerhq.hu>
>> > Cc: Baptiste Coudurier <baptiste.coudurier at smartjog.com>
>> >
>> >
>> > hi,
>> >
>> > My name is zhentan feng, and I am a student from china.
>> > I have finished last year's soc project MXF muxer mentored by Baptiste.
>> >
>> > This summer, I plan to apply a soc project of ffmpeg again, and i am
>> > interested in the NEW Seeking API project.
>> > should I start from qualification task again or beginning to collect
>> > information for the new project?
>> >
>> > sincerely yours
>> > zhentan feng
>> > --
>> > Best wishes~
>>
>> Hi,
>>
>> I was playing around with writing a Vivo demuxer myself, but it
>> currently doesn't work; attached is the (horribly messy) work so far.
>>
>> I'm not sure if it's actually working, but feeding the supposedly
>> H.263 video stream from any of the samples to ffmpeg's decoder results
>> in a bunch of nasty error messages about various invalid values; this
>> could be because my code is broken (likely :) or because the video is
>> not actually standard H.263 (also possible; anyone have more
>> information?). ?The mplayer demuxer uses the binary Win32 codecs to
>> decode both the video and audio, so it's not too much help here.
>>
>> FFmpeg also doesn't currently support either of the possible audio
>> codecs (G.723 and Siren), so I can't check if those work either.
>>
>> I'm probably not going to work on it any more unless someone's got a
>> revelation about the H.263 problems, so feel free to do what you want
>> with it...
>>
>> Thanks,
>> -- Daniel Verkamp
>>
>> [...]
>
>
> I am very glad to hear from you that you are doing the vivo demuxer and have
> made great work,
> so, any related specs for vivo format? I am familar with the muxer mechanism
> of FFmpeg but know little about vivo.
>
> According to the link:
> http://wiki.multimedia.cx/index.php?title=Small_FFmpeg_Tasks#Implement_a_Vivo_demuxer_for_FFmpeg
>
> Is the file demux_viv.c of MPlayer main reference?
>

The MPlayer demuxer is what I have used as a reference so far (as well
as inspecting sample files).  As far as I know, there is no
specification available.

> thanks
> zhentan
> --
> Best wishes~