[FFmpeg-devel] Why the tag “DURATION” is a necessity in Matroska files?

wm4 nfxjfg at googlemail.com
Mon Jun 19 11:28:46 EEST 2017


On Mon, 19 Jun 2017 09:38:19 +0200
Tobias Rapp <t.rapp at noa-archive.com> wrote:

> On 19.06.2017 02:04, 21naown at gmail.com wrote:
> > Le 18/06/2017 à 00:57, Sasi Inguva a écrit :  
> >> Hi,
> >>
> >> I was the one who made that change. We needed a way to obtain individual
> >> stream durations from a  Matroska file without reading the entire file. I
> >> don't think it adds to the file size greatly (< 50 bytes) compared to the
> >> size of video and audio data, hence I didn't hide it under an option.
> >> However if u really need it I can add an option.
> >>  
> >
> > Hello,
> >
> > Yes the size required for the tag “DURATION” is insignificant, but it is
> > preferable to have a choice with this tag like mkvmerge partially does
> > (because by default “--enable-durations” is enabled, maybe because it is
> > a part of “statistics-tags”, but we can disable them with
> > “--disable-track-statistics-tags”) and because the end user and his
> > software do not require it.
> >
> > Yes if you can move the tag “DURATION” as an optional tag—so an argument
> > is needed to enable it, I will be grateful to you.  
> 
> I disagree to this tag writing being opt-in. It requires marginal 
> runtime and storage space resources to write a duration field. The 
> application used for playback might change or even a new version of the 
> same application might make use of the tag. So it would be better to 
> enable it by default.
> 
> And for the tag writing being opt-out: Maintaining code also has some 
> cost. As far as I understand this option would just be required for some 
> aesthetic reason, thus I don't think it's worth adding that maintenance 
> cost by implementing code to make this tag opt-out.

But there's also this argument that writing these tags is silly, and
that mkvmerge should never have started it. Instead, there should be
proper EBML elements for this.


More information about the ffmpeg-devel mailing list