Should I be at all surprised that the ffmux_mpeg generates what seems to be garbage?
gst-launch-0.10 -v filesrc location=/home/dvds/thoth/myth-mirror/The\ Drinky\ Crow\ Show\ -\ Tunnel\ Girls.mpeg ! decodebin name=db ! videoscale ! "video/x-raw-yuv, width=352, height=480, pixel-aspect-ratio=1920/1056" ! ffenc_mpeg2video ! ffmux_mpeg ! filesink location=/tmp/x db.! fakesink
I'm trying to put some reencoded video into a program stream to
preserve the PTSs. It's not working so well.
The program stream pack header seems to be short and contains
nonsensical values. ISO 13818-1 table 2-33 leads me to believe that the
smallest PS pack header is (including start code) 112 bits, but based on
the stream, I think it's 2 bytes too short.
The mux rate on the packets appears to be uniformly 336870400bps.
All but one of the marker bits are wrong.
The time code appears to be monotonically increasing, but bears no
relation to the SCRs or PTSs in the source stream.
The PES headers also appear to be flawed.
I found a couple of instances where the PTS_DTS_flag was 01, which isn't
even permitted by the syntax of table 2-17. The PTS values themselves
were quite random.
The stream DOES play in mplayer, but this is probably more a function
of mplayer not caring about clocks or bit rates that are mostly of
interest to multiplexers.
Is there someone I can work with to help improve the quality of the
packets?
------------------------------------------------------------------------------
Open Source Business Conference (OSBC), March 24-25, 2009, San Francisco, CA
-OSBC tackles the biggest issue in open source: Open Sourcing the Enterprise
-Strategies to boost innovation and cut costs with open source participation
-Receive a $600 discount off the registration fee with the source code: SFAD
http://p.sf.net/sfu/XcvMzF8H_______________________________________________
gstreamer-devel mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/gstreamer-devel
Locked
