I'm sure that Zenverge wants to appeal to the traditional cable company architecture and cost structure, making the existing cableco broadcast channel(cable broadcast, that is) QAM/MPEG-2 streams relevant to IP devices. But honestly, the simple and elegant solution is to eliminate those broadcast QAM/MPEG-2 channels entirely, and convert the cableco PON networks to two-way IP pipes more effectively and completely.
For instance, instead of retaining the vast majority of the PON's 6 MHz channels for one-way broadcast QAM/MPEG-2 TV streams, you distribute these 6 MHz channels to individual households in each PON neighborhood, as their IP broadband link. Making much more efficient use of the PON's capacity to provide two-way IP pipes.
Having done that, each household would have enough capacity to watch two, three, maybe more, HDTV streams that are IP encapsulated. For those live events that "everyone" wants to watch simultaneously, the cableco can deliver the streams as IP multicast, within the confines of their network, to save on unicast sessions.
The Zenverge solution could be a short-term transitional solution maybe, although I'm not even sure how necessary it is today. I thought a lot of cable companies were already providing IP streams for their customers "anywhere/anytime." Seems to me that there are no show-stopper technical obstacles anymore, in this regard. The only obstacles are that the cablecos want to retain their tiered bundles and pricing structure. And even there, ultimately they could do this over an all-IP network anyway.
"the simple and elegant solution is to eliminate those broadcast QAM/MPEG-2 channels entirely."
Simple and elegeant yes, but also expensive. Replace all the edge QAMs in every neighborhood as well as replace every subscriber's STB. That's billions of dollars worth of simple & elegant.
What they need is an econmically viable transition plan that shifts the network allocation to less QAM/MPEG2 and more DOCSIS/IP without bankrupting the company. Perhaps Zenverge's transcoding strategy can be part of such a plan. If so, I expect it will happen more in the headless gateway realm rather than in the STB realm.
Since most cable providers are already offering broadcast and Internet to consumers along the same residential coax, shifting the balance between them would seem to be a very manageable task. It would be much more of a problem if the two could not coexist in the same wire or if devices didn't already exist to split the feeds at the residential end.
There are many paths cable companies can take to eventually get to all IP infrastructure. I think a "transcoding" at home option Zenverge is offering is an interesting one.
Actually, a real story is how Zenverge managed to sneak into the STB market -- traditionally a stronghold of Broadcom and STMicroelectronics.
If things are transitioning to "headless" media home gateway architecture as Zenverge claims, what cable guys will need is not an overly complicated STB, but just a thin client and a MHG integrated with a residential gateway. That transition, in my opinion, is interesting to watch.
Cable companies don't need anything they don't already have, Junko, other than reconfiguring their existing infrastructure to more IP and less one-way streams. That's my point. (Also DrQuine's point.)
If you're going to have to install new STBs anyway, to make use of this Zenverge scheme, why not instead install STBs in which most or all of the 6 MHz channels on the cable can be allocated to Internet broadband access? That being done, now the cable company can offer more Internet bandwidth to their customers, and more, or even all, of their TV content over IP. Which they are already doing anyway, for access to tablets and smartphones.
In other words, nothing new needs to be invented. Just reapportion what you already have.
It seems to me that Zenverge would have more appeal to the satellite TV companies, where the two-way Internet pipes are far more difficult to provide.