Re: Extending MediaRecorder to record from Web Audio node faster than real time?

Robert O'Callahan <robert@ocallahan.org> 於 2014/8/28 上午10:10 寫道：

> On Thu, Aug 28, 2014 at 1:49 PM, Robert O'Callahan <robert@ocallahan.org> wrote:
> With the above "definition", ChannelSplitterNode is not a problem; the 0'th output (the first channel) of audioNode gets recorded and the others are ignored. We'd just need to say that in the spec.
> 
> Let me clarify that:
> 
> "new MediaRecorder(audioNode)" records the first output of audioNode (where "output" is defined in the Web Audio spec).
> 
> For those unfamiliar with Web Audio: currently only ChannelSplitterNode has more than one output. Each output can have any number of channels.
> 
> BTW AudioDestinationNode has 0 outputs so I think if someone creates a MediaRecorder for that node we should just throw an exception. Alternatively we could change the Web Audio spec so it actually has an output (the mix of its inputs). I actually prefer the latter since it's more DWIM; I'll post to public-audio.

 Is it necessary to give AudioDestinationNode an output? Couldn’t we just say that MediaRecorder records the mix of inputs of the destination node for AudioContext, and OfflineAudioCompletionEvent.renderedBuffer for OfflineAudioContext?

> 
> Rob
> -- 
> oIo otoeololo oyooouo otohoaoto oaonoyooonoeo owohooo oioso oaonogoroyo
> owoiotoho oao oboroootohoeoro oooro osoiosotoeoro owoiololo oboeo
> osouobojoeocoto otooo ojouodogomoeonoto.o oAogoaoiono,o oaonoyooonoeo owohooo
> osoaoyoso otooo oao oboroootohoeoro oooro osoiosotoeoro,o o‘oRoaocoao,o’o oioso
> oaonosowoeoroaoboloeo otooo otohoeo ocooouoroto.o oAonodo oaonoyooonoeo owohooo
> osoaoyoso,o o‘oYooouo ofooooolo!o’o owoiololo oboeo oiono odoaonogoeoro ooofo
> otohoeo ofoioroeo ooofo ohoeololo.