Re: [rtcweb] RTCWEB needs an Internet Codec

[Harald Alvestrand <[e-mail]>]

On 08/29/2012 06:13 AM, Roman Shpount wrote:
> Not that I have anything against Opus, but what exactly makes Opus an 
> internet codec? What is internet codec anyway? Makes this all kind of 
> a meaningless argument.
>
> I would argue that for my broadband internet connection G.722 is a 
> perfect internet codec. I do not care about bandwidth savings of Opus, 
> and quality wise, for the voice conversation, I cannot hear any 
> difference.
>
> I would argue G.711 should be the MTI codec. The rest can be left up 
> to browser implementers. We can argue all we want, but the best 
> royalty free low bitrate codec available will be the one everybody 
> supports. We can force it to be Opus, but even if we don't, it will 
> still be Opus on its merit alone. G.722 will probably end up being 
> supported as well, since it is free, pretty good quality, and easy to 
> implement.
>
> P.S. Not that I am arguing for it, I am surprised no one made a case 
> for iSAC, since it is also royalty free, low bit rate, and very high 
> quality. It is event named "internet Speech Audio Codec".

Google, after considering both, would prefer that Opus be the standard.

At high bandwidth / bitrate, Opus is operating in CELT mode, which 
outperforms iSAC.