[vlc] Re: [streaming] Trying to emulate VLC streaming in a test equipment, but failing

[Chhaya, Harshal]

Hi,

I found some more details and so am following up to my earlier
post in the hope that this additional information can help
someone diagnose the problem and suggest a solution.

I analysed the messages generated in each case and I found
that when I stream using VLC or when I stop and re-start the
client when streaming from the test equipment, the client 
displays this:

ts debug: pid[68] unknown
ts debug: PATCallBack called
ts debug: new PAT ts_id=31975 version=23 current_next=1
ts debug:   * number=1 pid=66
ts debug: PMTCallBack called
ts debug: new PMT program number=1 version=25 pid_pcr=68
ts debug:   * es pid=68 type=3 fcc=mpga

but when I stream from my test equipment in a loop, the 
client displays only:

ts debug: PATCallBack called
ts debug: PMTCallBack called
ts warning: discontinuity received 0x0 instead of 0x4 (pid=69)
mpgatofixed32 debug: libmad error: bad main_data_begin pointer
main warning: buffer is 71700 in advance, triggering downsampling
mpeg_audio debug: emulated startcode (no startcode on following frame)
main warning: backward_pts != dts (40000)
main warning: backward_pts != dts (40000)

There is no "new PAT..." or "new PMT..." line.

What causes the client to detect a new PAT and PMT? Is it
something that the server sends?

How can I ensure that the client VLC detects when the video
stream has looped around?

BTW, I found the 'use sequential numbers instead of 
timestamps' option in vlc 0.8.6 and enabled it but it didn't 
make any difference.

FWIW, mplayer can play the stream even from the test equipment
but the video quality gets progressively worse. It also can't
play the audio which is in mp3 format.

Thanks for any and all suggestions.

Regards,
- Harshal

-- 
This is the vlc mailing-list, see http://www.videolan.org/vlc/
To unsubscribe, please read http://www.videolan.org/support/lists.html