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

[Baptiste Coudurier]

On 9/29/09 3:17 AM, Reimar D?ffinger wrote:
> On Sun, Sep 27, 2009 at 10:17:56PM +0200, Michael Niedermayer wrote:
>> On Sun, Sep 27, 2009 at 10:10:09PM +0200, Reimar D?ffinger wrote:
>>> On Sun, Sep 27, 2009 at 10:02:10PM +0200, Michael Niedermayer wrote:
>>>>> 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 ...
>>>
>>> Video maybe not, but see just the current situation where mpeg files with AC3
>>> are recognized as AC3 files. That kind of thing I mean.
>>
>> so there is a bug in the ac3 probe, did you report that?
>
> No, because I saw a bug report about it. Can't find it anymore though...
>
>> i would guess copy&  pasting the thresholds and scores from mp3 probe
>> might fix it ...
>
> Maybe, if I manage to find the report I'll look into it (does anyone
> else seeing the report? It was about ffmpeg-created files not recognized
> by ffmpeg).
> And I finally applied this patch, at least for what I can tell it
> doesn't make anything worse.

FYI One user reported a problem in -user:

And here's a sample video clip:

http://mcfrisk.kapsi.fi/temp/ffmpeg/small_test2.dv

It was taken from a Sony dv camera with dvgrab. Same problem comes up
with clips that have been deinterlaced with ffmpeg and packaged to an
.avi file.

One more sample to test.

-- 
Baptiste COUDURIER
Key fingerprint                 8D77134D20CC9220201FC5DB0AC9325C5C1ABAAA
FFmpeg maintainer                                  http://www.ffmpeg.org