Re: Issue 81: Separating audio and video capabilities

RTP header extensions will often be media-specific (e.g. sender audio
level). Same with RTCP feedback mechanisms.


On Fri, May 9, 2014 at 10:57 AM, Peter Thatcher <pthatcher@google.com>wrote:

> I think it's nice to be able to get the capabilities without creating a
> sender or receiver object first.
>
> I don't really see the advantage of having separate audio/video
> capabilities.  What things will be audio-only or video-only?  The codecs?
>  The codecs already indicate whether they are audio or video.
>
>
> On Thu, May 8, 2014 at 10:19 PM, Bernard Aboba <
> Bernard.Aboba@microsoft.com> wrote:
>
>> Currently, getCapabilities is defined as follows:
>>
>> partial interface RTCRtpSender {
>> static RTCRtpCapabilities getCapabilities ();
>> };
>>
>> partial interface RTCRtpReceiver {
>> static RTCRtpCapabilities getCapabilities ();
>> };
>>
>> The implication is that getCapabilities() returns both audio and video
>> capabilities in a single RTCRtpCapabilities object.
>>
>> However, if getCapabilities wasn't defined as static, then it could
>> return only capabilities of the appropriate kind:
>>
>> partial interface RTCRtpReceiver {
>> RTCRtpCapabilities getCapabilities (DOMString kind);
>> }
>>
>> partial interface RTCRtpSender {
>> RTCRtpCapabilities getCapabilities ();
>> };
>>
>> If it were done this way, with respect to the RTCRtpSender object, it
>> would be possible to retrieve audio and video send and receive capabilities
>> separately:
>>
>> var audioSender = new RTCRtpSender(audioTrack, transport);
>> var videoSender = new RTCRtpSender(videoTrack, transport);
>> rtpAudioSendCaps = audioSender.getCapabilities();
>> rtpVideoSendCaps = videoSender.getCapabilities();
>>
>> With the RTCRtpReceiver object, it would look like this:
>>
>>    var audioReceiver = new RTCRtpReceiver(transport);
>>    var videoReceiver = new RTCRtpReceiver(transport);
>>    rtpAudioRecvCaps = audioReceiver.getCapabilities("audio");
>>    rtpVideoRecvCaps = videoReceiver.getCapabilities("video");
>>
>> One advantage of separating audio and video capabilities is that
>> subsequent calls to createParameters could be more specific:
>>
>> var audioRecvParams = RTCRtpReceiver.createParameters(
>> "audio", remote.rtpAudioSendCaps);
>> var videoRecvParams = RTCRtpReceiver.createParameters(
>> "video", remote.rtpVideoSendCaps);
>>
>
>