opus - Dec 2014 - Estimating bitrate during a real-time voip call

Hi,

Although this maybe considered out of scope here, but I'll ask anyway.

Opus has remarkable flexibility for changing encoder bitrate during a call.
Are there any suggestions about how bandwidth/capacity between the two
endpoints can be measured/estimated during a call so that the outgoing
bitrate can be adjusted accordingly?

Thanks,
Manpreet.
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://lists.xiph.org/pipermail/opus/attachments/20141215/5ebe425a/attachment.htm

Hi
You can start a VOIP call with 50 kbps bitrate and reduce the bitrate if there
is packet loss. ?You know if there's packet loss if you receive RTCP
.?Linphone does this .
Regards,Dragos Oancea
      From: Manpreet Singh <manpreets7 at gmail.com>
 To: opus at xiph.org 
 Sent: Tuesday, December 16, 2014 7:54 AM
 Subject: [opus] Estimating bitrate during a real-time voip call
   
Hi,
Although this maybe considered out of scope here, but I'll ask anyway.
Opus has remarkable flexibility for changing encoder bitrate during a call. Are
there any suggestions about how bandwidth/capacity between the two endpoints can
be measured/estimated during a call so that the outgoing bitrate can be adjusted
accordingly?
Thanks,Manpreet.

_______________________________________________
opus mailing list
opus at xiph.org
http://lists.xiph.org/mailman/listinfo/opus


  
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://lists.xiph.org/pipermail/opus/attachments/20141216/206e3a43/attachment.htm

Manpreet Singh wrote:
> Opus has remarkable flexibility for changing encoder bitrate during a
> call. Are there any suggestions about how bandwidth/capacity between the
> two endpoints can be measured/estimated during a call so that the
> outgoing bitrate can be adjusted accordingly?
The topic is quite complicated. You can find code implementing the 
algorithm that Firefox and Chrome use in WebRTC at 
<http://www.webrtc.org/native-code/development>, though it is deeply 
integrated with the rest of the media stack there. It is documented at
<https://tools.ietf.org/html/draft-alvestrand-rmcat-congestion>. The 
IETF has chartered a working group to standardize some techniques. You 
can find the other active proposals here: 
<https://datatracker.ietf.org/wg/rmcat/documents/>.

Hi Dragos,

The issue is that not all packet loss maybe congestion related. Often,
reducing bitrate seems to have no impact on improving packet loss.

Thanks,
Manpreet.


On Tue, Dec 16, 2014 at 2:09 AM, Dragos Oancea <droancea at yahoo.com>
wrote:
>
> Hi
>
> You can start a VOIP call with 50 kbps bitrate and reduce the bitrate if
> there is packet loss.  You know if there's packet loss if you receive
RTCP
> .
> Linphone does this .
>
> Regards,
> Dragos Oancea
>
>   ------------------------------
>  *From:* Manpreet Singh <manpreets7 at gmail.com>
> *To:* opus at xiph.org
> *Sent:* Tuesday, December 16, 2014 7:54 AM
> *Subject:* [opus] Estimating bitrate during a real-time voip call
>
> Hi,
>
> Although this maybe considered out of scope here, but I'll ask anyway.
>
> Opus has remarkable flexibility for changing encoder bitrate during a
> call. Are there any suggestions about how bandwidth/capacity between the
> two endpoints can be measured/estimated during a call so that the outgoing
> bitrate can be adjusted accordingly?
>
> Thanks,
> Manpreet.
>
>
> _______________________________________________
> opus mailing list
> opus at xiph.org
> http://lists.xiph.org/mailman/listinfo/opus
>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://lists.xiph.org/pipermail/opus/attachments/20141216/dd6da29c/attachment.htm