Marcus Metzler wrote:
Nico writes:
> > can you please post a backtrace of
> > mplayer -v recorded.ts ?
> I can't use the Athlon 2500+ right now, but it also works on a headless 2400+
with mplayer rescaling and the DVB-S card as output. I also tried it
on a P4 2.2 GHz with X11, but this one seemed too slow and the sound
stopped after a while. I used the same CVS version on all three Computers.
I used mplayer -vo mpegpes -ao mpegpes -vf scale=480:480 /usr3/MPGTEST/hd.ts
-v and got the following:
ADDED 1440140 bytes to video fifo, PTS=53287.886719
have you seen what PES they are shooting? They are stored in a demux packet until they are complete,
then passed (without memcpy()ing) to demuxer->{video|audio|sub}.
Maybe adding some cache (8192) will help a little, but I don't bet on it: the data to decode is too much.
-vc ffmpeg2 should help, too.
When I scale down the image to 640x480, the playback is acceptable on my Athlon xp1500+ with a Nova-S.
Yet, as far as I can see there's no segfault here .
I'm searching your segfault because I have to understand if it's another bug in the TS demuxer or in libmpeg2.
Nico
uninit video: libmpeg2
DEMUXER: freeing demuxer at 0x85324b8 DEMUXER: freeing sh_audio at 0x854b0e0 DEMUXER: freeing sh_video at 0x854b048 vo: x11 uninit called but X11 not inited..
-- Info: To unsubscribe send a mail to [EMAIL PROTECTED] with "unsubscribe linux-dvb" as subject.
