[FFmpeg-devel] dash encoder. Correct generated manifest for MPEG-DASH MPD Validator and calculate current bandwidth for eath chunk. Now works correctly with dash.sj

Michael Niedermayer michael at niedermayer.cc
Tue Apr 18 21:14:36 EEST 2017


On Fri, Sep 09, 2016 at 11:15:32AM +0300, Ligverd Haer wrote:
>  
> > missing commit mesage / not a proper git patch
> ...
> > this looks unrelated, unrelated changes need to be in separate patches
> > 
> > please explain in the commit messages of the patch(es) what each
> > patch does, why it does it and if you have at hand any reference to
> > specs feel free to include them
> > 
> > floats make the binary output code platform dependant, theres no
> > need to use floats here
> 
> Thanks for the positive comments Michael.
> 
> 0001-libavformat-dashenc.c-fix-validation-manifest.mpd.patch
> 1. If the MPD is dynamic the Period element shall have an id.
> 2. Attribute ‘bandwidth’ must appear on element ‘Representation’
> 3. Attribute [profiles,width,height,sar,frameRate,audioSamplingRate,mimeType,segmentProfiles,codecs,maximumSAPPeriod,startWithSAP,maxPlayoutRate,codingDependency,scanType]
>    Common attributes for AdaptationSet and Representation shall either be in one of the elements but not in both.
> the tests were conducted on http://www-itec.uni-klu.ac.at/dash/?page_id=605
> 
> 0002-libavformat-dashenc.c-calculation-bandwidth-for-mani.patch
> When stream encoding is not counted bandwidth which makes it impossible to properly switch to the desired stream.
> Use float I removed. Now the calculation accuracy decreased to kilobits.
> Is it possible to supply test division by 0 ?
> 
> 0003-libavformat-dashenc.c-change-moveflags-works-better.patch
> Replaced flags movflags. These flags input rtsp stream looked better.

Sorry for the late reply, i was hoping someone who knows dash well
would look at this and then i forgot

patch1 still applies cleanly
patches2 and 3 dont

patch1 removes the storage of the framerate this leaves several
unused variables.
Most likely the framerate should be stored differently instead of
not storing it anymore but
certainly no unused variables should remain

thx

[...]

-- 
Michael     GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB

Those who would give up essential Liberty, to purchase a little
temporary Safety, deserve neither Liberty nor Safety -- Benjamin Franklin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 181 bytes
Desc: Digital signature
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20170418/7e39585b/attachment.sig>


More information about the ffmpeg-devel mailing list