[MPlayer-users] Seeking in MPEG2/MPEG4 streams

Barnabas Hajas hajbazer at gmail.com
Mon Jan 4 23:18:36 CET 2010


Thanks for your help, -demuxer lavf gives me huge seeking performance
improvements, but -key-fifo-size 2 seems to do nothing. Also the
working option above causes warning messages at almost every decoder.
I find this weird, because this did not seem to be present at older
versions:

ffodivx-handled content: [NULL @ 00E05314]Invalid and inefficient
vfw-avi packed B frames detected
libmpeg2-handled content:
[mpeg @ 00DF7EB8]invalid dts/pts combination
[mpeg2video @ 00E05314]Missing picture start code
[mpeg2video @ 00E05314]Missing picture start code
[mpeg @ 00DF7EB8]invalid dts/pts combination
[mpeg2video @ 00E05314]Missing picture start code

Greets
Barnabas Hajas

2010/1/4 Reimar Döffinger <Reimar.Doeffinger at gmx.de>:
> On Mon, Jan 04, 2010 at 06:57:16PM +0100, Reimar Döffinger wrote:
>> On Mon, Jan 04, 2010 at 06:53:30PM +0100, Reimar Döffinger wrote:
>> > > I hope this has something to do with libmpeg2 or an option
>> > > (e.g. -lavdopts) which default value was changed during the releases.
>> >
>> > Probably -key-fifo-size 2 gives the old behaviour (if you knew where to
>> > look it would would have been in the manual :-) )
>> > However now that I try it there seems to be another, rather strange, issue.
>> > -demuxer lavf seems to avoid it for me.
>>
>> Actually it seems your guess was even better, for MPEG files -vc mpeg12
>> works far better (compared to not at all without) for seeking with the
>> default demuxer.
>
> Fixed for at least some cases with r30217 even though I am not sure I
> know 100% why. I currently suspect that this "fix" only works
> "accidentally" because/if the PC can decode faster than your key repeat rate...
> _______________________________________________
> MPlayer-users mailing list
> MPlayer-users at mplayerhq.hu
> https://lists.mplayerhq.hu/mailman/listinfo/mplayer-users
>


More information about the MPlayer-users mailing list