[FFmpeg-devel] [RFC] libavfilter-soc and variable-frame-size video streams

[Vitor Sessak]

Michael Niedermayer wrote:
> On Fri, Dec 12, 2008 at 04:54:26PM -0800, Jason Garrett-Glaser wrote:
>>> But maybe I'm just confused and there is a far simpler solution, or
>>> maybe we could just decide that supporting variable-frame-size video
>>> streams is not worth that hassle.
>> This sounds reasonable to me: thinking through what every filter would
>> have to do to support such a thing, it sounds completely
>> unmaintainable, especially for filters that act temporally: how will
>> they deal with the fact that cached temporal data from previous frames
>> (such as for temporal denoisers) is no longer the correct size
>> matching the current frame?!
>>
>> Avisynth gets away just fine without variable frame size support: I
>> have actually never seen such a thing as a feature request either.  I
>> don't think libavfilter needs it either.
> 
> I think variable size is pretty much a requirement for libavfilter.

While I understand and appreciate FFmpeg policy of "not in SVN until 
nobody knows how to improve it more", in this case the more requirements 
we put in lavfi, the later it'll get in SVN. What I propose is to work 
in getting lavfi commited but not to make lavfi part of the public API 
until all those things really needed are supported (variable frame 
sizes, audio and subtitle filters, memcpy-less padding, etc). We now 
have an API with a pretty complex colorspace negotiation and memory 
allocation, but no real world testing since it is not in svn. I'm afraid 
that adding more architecture to it may make it even more unlikely to 
get committed any time soon (and harder to spot bugs once it is).

-Vitor