Re: [rtcweb] Tim Panton's choices

[Ron <[e-mail]>]

On Mon, Dec 30, 2013 at 04:00:24PM +0000, Tim Panton wrote:
> 
> So If I understand that correctly:
> 1) rate control on the older codecs is possible.
> 2) rate control is typically separated from the codec itself
> 3) rate control is heavily patented
> 4) a modern H261 implementation would (mostly) use the same
>    rate control techniques as H264 or VP8 

Or the same rate control techniques you'd use for audio, or any
other real time stream that might suffer congestion or loss.

You have a knob that lets you tweak the bitrate.  You (hopefully!)
have information on what bitrate your link can sustain.

You insert math between them, and hope your math isn't so blatantly
obvious that someone already patented it.  Just like anything else.


> which leaves me wondering if selecting H261 wouldn't mean we retain many
> of the same IPR risks of newer codecs but without a licensing authority
> to go to for fixes/workarounds (or a technical or legal nature). And still
> have a significantly worse visual experience at a given bandwidth than for
> a modern codec.
> 
> Seems like the worst of all worlds to me - hopefully I misunderstood.

I think you might have ...

These are orthogonal problems.  You don't get any protection from the
licence you bought with a codec.  If the rate control method you choose
to use is encumbered then it doesn't matter what codec you chose.  What
happens next is entirely between you and who owns whatever rate control
patents you tripped over.


If the risk of codec IPR is like the risk of playing on the highway,
then the risk of other unrelated IPR is like the risk of being hit by
a meteor.

You can play on the highway, and try to dodge both trucks and meteors.
Or you can move off the highway into the open space and only have to
worry about the meteors.


What Stephan is pointing out, is that even among the people who have
reached a MAD stalemate in the codec patents they hold, there's a
completely separate competition for who can screw the other with a
monopoly on effective rate control mechanisms.  And who can scare
everyone else into not even trying.


> Surely with ISDN the steps were quantized at 64kbps? 

With ISDN, when using more than one timeslot, it's not uncommon to share
the available timeslots between all of the data that needs to be streamed,
(video, audio, control channel etc.) and they don't all have to fall on
an exact timeslot boundary.

You're wasting bandwidth if you don't use all of the available timeslot
space once all your streams are added together, but individually there
is no need for them to each be in strict 64kbs units.  And for H.261 they
quite often are not.

  Ron