[FFmpeg-devel] [PATCH] another stab at improved dv probe

[Michael Niedermayer]

On Sun, Sep 27, 2009 at 02:00:50PM +0200, Reimar D?ffinger wrote:
> On Sun, Sep 27, 2009 at 01:27:54PM +0300, Maksym Veremeyenko wrote:
> > i had a mov file where DV frames chunk start from 17-th byte and
> > probe test found that its a raw DV stream with ratio 0.75. In that
> > case truncated raw DV stream will have 0.5 but normal 0.99
> 
> Your code does not really work well for that, first it can be mod 12000 for
> mov, too, while for badly cut DV files it won't be at offset 12000.

yes


> _Reliable_ detection of DV in mov would check that the distance between "primary"
> DV headers is exactly the same as the frame size, but that is completely
> independent of single-frame detection.

thats not reliable, one can store dv in mov so that the frames are exactly
by that amont apart. Checking for the lack of mov atoms in addition would
be more reliable (iam not suggesting that to be done of course ...)


> Now another step "up" would be to extend the frame reading to support DV
> in mov without mov,

that would make dv be at the level of every other major video decoder in
ffmpeg in that respect ...


> but that is likely not such a great idea due to possibly
> breaking things too much.

i dont remember breakages in any other video decoder or parser that searches
for a startcode if one is missing at the expected pos ...

[...] 

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

The greatest way to live with honor in this world is to be what we pretend
to be. -- Socrates
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 189 bytes
Desc: Digital signature
URL: <http://lists.mplayerhq.hu/pipermail/ffmpeg-devel/attachments/20090927/5ad79454/attachment.pgp>