[Live-devel] Where can I trap and debug response to signalNewFrameData() ?

[temp2010 at forren.org]

Ross,

Thanks.  Cool... I figured out this exact same problem on my own yesterday,
perhaps 5 hours before your email.  I got up in the middle of the night
tonight and implemented a fix.  I just got it working... yeah!... and went
to email you back.  Then I found your email.  I was proud of myself for
figuring it out.  I had to add or re-enable a bunch of tracing, comparing a
working file-based case to failing my-device-stream case (sorry, I had to
temporarily punch a bunch of code to do this).  But this helped me find it.
 After all, there were no positive indicators (aka warning messages) about
this absence, so it could have been any one of a number of problems.  You,
of course, are very familiar with the software, so you nailed it on the
first suggestion!

The problem was my live stream starts before the [VLC] client connects, and
so the SPS and PPS NAL units have long passed by before Live555 code begins
looking for them.  So, I figured out how to add a callback function pointer
to my device, so that when it's created it calls this callback, which posts
a message that starts up my live stream.  Thus, the live stream won't start
until after the [VLC] client connects.  This fixed the problem.  The
Live555 code sees the SPS and PPS NAL units and begins working.

Thanks for your assistance nevertheless.  If I hadn't had a sudden
breakthrough on my own yesterday, your advice now would have made it happen
for me today.  And your prior assistance got me up to the pre-breakthrough
point as well.

NEXT, I'm back on to my horrible video quality.  You may recall an old post
from me about it, from my first otherwise-working cobble, then my wasn't
really double-encoding confusion.  It simply MUST BE a problem with my
stream origination.  I'm going to go search for some H.264 analysis tools,
and I'll have to better educate myself on it.  FYI, I'm using Microsoft
Media Foundation in the stream origination, and using their H.264 encoding
function.  Something must be wrong with the setup or use, because it seems
the quality can't be on the Live555 side (including my code and fault),
because the from-file version works beautifully.



On Tue, Feb 19, 2013 at 9:26 PM, Ross Finlayson <finlayson at live555.com>wrote:

> I suspect that your input source does not contain SPS and PPS NAL units.
>  Our server code needs to see these (to generate stream 'configuration'
> information) before it can initialize completely.
>
> Therefore, you should either (1) make sure that your input stream contains
> SPS and PPS NAL units (preferably at or near the start), or (2) change your
> "createNewRTPSink()" implementation to use one of the alternative forms of
> "H264VideoRTPSink::createNew()" that take SPS and PPS NAL unit data as
> parameters.
>
>
> Ross Finlayson
> Live Networks, Inc.
> http://www.live555.com/
>
>
> _______________________________________________
> live-devel mailing list
> live-devel at lists.live555.com
> http://lists.live555.com/mailman/listinfo/live-devel
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.live555.com/pipermail/live-devel/attachments/20130220/9aa58f81/attachment-0001.html>