In these days when we (including yours truly) obsess about YouTube, Facebook, and every tiny feature upgrade from Google, it is fairly easy to forget and perhaps overlook the fundamental technologies that are changing the networks, and making all this fun stuff possible. One such announcement comes today from San Jose, Calif.-based start up, Broadlogic Network Technologies.
The company which has received funding from the likes of Cisco Systems, Intel, Time Warner, and August Capital is well known for developing a chip that would theoretically allow about 600 megabits per second over cable company’s pipes. Today, the company has announced a new chip, the TeraPIX video processor that allows cable operators to better utilize the spectrum inside the coaxial cables, triple the digital spectrum and basically give the plain old coax capabilities to take fiber to the home (FTTH) technologies head on, at least for now.
Typically every cable operator ends up offering 80 channels (give or take a few) of broadcast and cable channels that can be received without a set top box. These channels are delivered in an analog fashion, and consume about 500 MHz of the total 750 MHz capacity in a cable network.
Rest of the capacity is shared between various services – it can accommodate 200 standard digital channels, 30 HD channels, Video on demand channels, Internet Data (DOCSIS) and VoIP. In short, these are the channels of the future, and the big money makers, yet they live in 30% of the network capacity. One way to overcome this problem is deploy a digital set top box for every cable-ready device. Doable, but an expensive approach.
To get this bandwidth back, operators can deliver every channel in their lineup in digital format – but conventional approaches to viewing these digital channels would require their subscribers to install a digital set-top box and remote for every cable-ready device in the home. This solution is both costly for operators and undesirable to many subscribers.
Broadlogic says TeraPIX is the other option. A massively parallel chip, it is capable of decoding dozens of digital video streams. The chip, sits inside a new kind of residential gateway that sits at the edge of an all-digital cable network. The chip recreates the 80 analog channels inside the consumer homes, and yet keeps the network all-digital giving the cable operators ability to offer more current (and profitable services such as Internet access, voice and HD video.
The chips, and other such solutions might just help the cable companies blunt the speed advantages of fiber, at least in the near term.
Sounds really impressive! So this one chip does the work of 80(!!) Mpeg (2? 4?) decompressors? This could be a great step, for the transition to more bandwidth on cable networks.
By the way: Do you, in the US, get 80 analogue channels? Here, in the Netherlands, we have around 30, and I tought that was around the technical limit?
I guess the big question will be if this chip is cheap enough: If it’ll cost $400, it will probably be cheaper to just bring a settop box to every TV, instead of this decoding beast in a gateway
Sorry, it’s hard to imagine that this wonder-chip is cheaper than a few $50 single-chip digital STBs.
Wes, the issue here are the many customers that have multiple TVs but don’t want any STBs. These customers make the least amount of money for the cable company (don’t buy the digital tier or DVR) but are also the largest drain on capacity (need all those analog channels). Puting one box based on this TeraPIX per house allows the network to be all digital and still support multiple TVs/house with no STB. So, the TeraPIX makes sense even if it costs as much as 3 or 4 cheap STB. Also, since you need fewer TerPIX than you would STB, you need less warehouse space and have fewer devices to maintain.
Pingbacks don’t seem to be working, so I’ll post this as a comment.
My thoughts on TeraPix can be found here.
http://www.nyquistcapital.com/2006/11/06/broadlogic-and-cables-analog-ghetto/
If a company created cheap a Digital legacy to analog legacy multiplexer. Cable companies could deploy TeraPix without disrupting the analog base customers.
Great post Om. This almost sounds like Dense Wavelength Division Multiplexing but for Cable? I would be very interested in seeing Broadlogic’s initial pricing around this. As is mentioned above the solution is both costly for operators and subsequently it may undesirable to many subscribers given high subs and expensive DTB’s they may need to purchase but its definitely one to keep a eye on.