[Kdenlive-devel] AVCHD seeking committed into FFmpeg repository

[Ivan Schreter]

Hi Dan,

Dan Dennedy wrote:
> Not sure why this did not get to the MLT list.
>
> 2009/8/22 Ivan Schreter <schreter at gmx.net>:
>   
>> Hi all,
>>
>> FYI: since there were no objections from FFmpeg since a week, I've just
>> committed my seeking patches to FFmpeg repository (revision 19681). MPEG-TS
>> (and thus also AVCHD) seeking now works, at least for AVCHD movies from my
>> Panasonic camcorder. I know that Canon camcorders had some issues with
>> proper key frame detection (files are missing appropriate SEI messages), so
>> they might still have a problem.
>>
>> As for MLT, I've updated my patch for the newest MLT revision (as of today,
>> 2009-08-22). I didn't test it too much, though, but here it is attached. So
>> feel free to test it.
>>     
>
> That's great! I will apply the patch, but in a way that requires an
> explicit configure flag to enable it.
>
>   
>> MPEG-PS movies (e.g., DVD VOB) don't work correctly with MLT patch only. Two
>> patches for FFmpeg are still needed, one which introduces proper seeking for
>>     
>
> I think I will name the flag --avformat-break-mpegps to make people
> acknowledge the risk. My clients and I use a lot of non-VOB/DVD MPEG-2
> Program Streams.
>
>   
I'm working on another patch for MPEG-PS, so that will be supported in 
the near future correctly as well, I hope.

Since the changes I made are in if() conditions, you can eventually use 
this to disable it for MPEG-PS for now. Configure flag with name 
"*-breaks-*" doesn't sound particularly good :-)

>> MPEG-PS, one to fix a bug in mpegvideo parser. These patches are not yet
>> accepted by FFmpeg, but I attached them here. They also cause seeking
>> regression (patch NOT attached).
>>     
>
> Please explain further the last line. I had first read it to mean:
> applying the patches breaks seeking on ps when the app has not made
> some adjustments. However, I do not understand what patch you refer to
> when you say it is not attached.
>
>   
Only the patch with regression is not attached. The two FFmpeg patches 
for parser fix and seek update were attached, so you have complete code.

Applying the seeking patch on FFmpeg doesn't break apps. It just seeks 
correctly. So the current workaround in MLT with seeking one second 
backward will work with the patch in FFmpeg as well, you'll just get 
less performance (as it is today) and off-by-2-3-frames. Only MLT patch 
and no FFmpeg patches will break seeking of MPEG-PS.

I hope it's clear now...

Regards,

Ivan


-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.kde.org/pipermail/kdenlive/attachments/20090824/d96110c6/attachment.html>