On 2009-03-10, Ryan Leavengood <leavengood@xxxxxxxxx> wrote: > On Mon, Mar 9, 2009 at 7:24 PM, David McPaul <dlmcpaul@xxxxxxxxx> wrote: > > > > Has any thought been given to developing a Video Mixer in the same > > manner as the Audio Mixer? > > I think it sounds like a neat idea, and if it could simplify decoders > then I say it is worth looking into. > > > Also, when we want to add subtitles we will need to be able to merge 2 > > video streams (video + subtitle) > > Though I am sure François is right about the overlay, I still think > this sounds like an interesting application for a video mixer. I am > sure there are other possibilities, like maybe a picture in picture > for a TV application. PIP, Subtitles, OSD. To me they are all just mixed video. Perhaps it is easier to use overlay but overlay is not always available. > Though if we have to invent reasons for it and it is a lot of work to > do then maybe it isn't worth it. Are there other potential drawbacks > besides implementation time? Writing a basic mixer should not be too difficult especially if we just limit it to conversions. A lot of conversion code is already done in various video decoders. Putting it into the media kit might be an issue regarding R1 compatability. I am not sure if we can add a getVideoMixer call like we have a getAudioMixer without breaking things. Then converting MediaPlayer to use it would be next. Anyway, I might play with a few ideas and try and get the basics done before looking to make it a node. -- Cheers David