Re: [codec] possible issues to track

Just to clarify my last email, I think we're actually mostly in 
agreement here. We already know that we can do stereo just by using 
independent channels, so that would be the basic requirement. From 
there, it should just be up to the working group to decide whether any 
proposed stereo (or even >2 channel) coupling is worth the extra 
complexity it adds. If there's a way to get large savings from a simple 
enough coupling, then we should use it. If it turns out that we can't 
save more than a few bits, we just drop the idea. Makes sense?

	Jean-Marc

On 2010-03-25 18:30, stephen botzko wrote:
> I am thinking that we really ought not have a requirement for joint
> encoding.  Instead it would be better to have quality requirements for
> multi-channel.  If a codec candidate achieves better quality though
> joint encoding, then that is great.  If it achieves the desired quality
> through independent channel encoding, than that is fine also.
>
> Ideally, we should avoid creating requirements that force candidates to
> use specific technology or techniques.  I think its better if we stick
> with defining the quality, delay, birates, etc that is needed for the
> application.
>
> Stephen Botzko
>
> On Thu, Mar 25, 2010 at 3:09 PM, Michael Knappe <mknappe@juniper.net
> <mailto:mknappe@juniper.net>> wrote:
>
>     Thanks Ben, my understanding had come from a discussion with
>     Jean-Marc and
>     some others involved in that discussion a few days back, apologies
>     if that
>     misrepresented consensus on this topic.
>
>     Jean-Marc, can you discuss your thoughts with respect to joint channel
>     coding?
>
>     Ben, how would you approach channel counts larger than 2? Joint coding
>     between any two L-R pairs, with other channels (like the center channel)
>     left singular?
>
>     Cheers,
>
>     Mike
>
>
>     On 3/25/10 3:01 PM, "Benjamin M. Schwartz" <bmschwar@fas.harvard.edu
>     <mailto:bmschwar@fas.harvard.edu>> wrote:
>
>      > Michael Knappe wrote:
>      >> we need to decide if we will support any kind of joint channel
>      >> coding (current consensus so far is to keep all channels separable).
>      >
>      > Could you clarify this?  I think we've had several discussions where
>      > people wanted jointly coded (i.e. non-separable) stereo.
>      >
>      > --Ben
>      >
>
>     _______________________________________________
>     codec mailing list
>     codec@ietf.org <mailto:codec@ietf.org>
>     https://www.ietf.org/mailman/listinfo/codec
>
>
>
>
> _______________________________________________
> codec mailing list
> codec@ietf.org
> https://www.ietf.org/mailman/listinfo/codec