Slashdot Mirror
SBC and Microsoft to Provide HDTV Over IP
Chroniton writes "SBC Communications (The #1 DSL provider in the US) is announcing new plans for broadband deployment, including internet, HDTV, and VOIP service: "With today's announcement, SBC will significantly accelerate its previously planned deployment pace and now plans to reach 18 million homes by year-end 2007. Through Project Lightspeed, the company will deploy 38,800 miles of fiber - double the amount used to build out the company's DSL network - at a cost of $4 billion to $6 billion."
This comes in response to an FCC ruling which shields IP-based networks from traditional telecom regulation. Speeds are expected to reach 15-25 Mbps, enough for HDTV: "To take advantage of this new network, SBC companies and Microsoft have begun testing an IP-based switched television service based on the Microsoft TV IPTV platform. This infrastructure would enable features such as standard and high-definition programming, customizable channel lineups, video on demand, digital video recording, multimedia interactive program guides and event notifications. IP-based television services will also allow TVs to interact with other devices in the home, including computers and PDAs." More details available here and here"
A friend of mine worked at a TV station that I am not permitted to reveal (but is right in MS's backyard somewhere). They had a multimillion dollar pilot project to use Microsoft software to deliver digital signals between the studio and the transmitter (and cable distro point) with dedicated, unlimited bandwidth digital circuits. Microsoft threw millions of dollars of research money into the project, it was to be their showpiece, to demonstrate how MS could provide end-to-end digital infrastructure for TV stations.
It was an utter failure. Despite the use of supposedly uncompressed video, everyone started complaining the picture was fuzzy and the audio didn't sync perfectly. The station abandoned the project after millions of dollars of their own investment, MS lost even more money.
And this was plain old NTSC video, not even HDTV. If MS couldn't get this project to work with the entire company behind it, what in the HELL makes people think they could succeed at HDTV?
Yes, it's called ADSL2 and I have it at my house in St. Louis, MO right now. Granted I'm on SBC's trial program for the next 3 months, but the speeds can reach as high as 24Mb/s which is enough for HDTV. They just installed the new DSL modem last week and are going to be ramping up speed over the next few weeks. Currently I'm at about 5Mb/s.
Celebrate the finer things in life
And they announced it back in 2003 "We plan to hit about one million lines by the end of 2003".
And they announced it back in 2002.
Stay tuned for another announcement in 2005.
This time they're paying back the Bush adminstration for the FCC deal that permits them to keep third-party ISPs from using their lines. The telcos have been lobbying for this for years, so that consumers don't have a choice of ISPs. It's an election year move, not a new development.
SBC has talked up a few fibre-to-the home trials, but even the small scale trials never seem to happen.
15-25 Mbps is enough for one HDTV stream to one location.. I would hope that a jump in technology like this would at least support something equal to current technology (many people have 2 or three cable or satellite receivers today).
Hopefully they are using multicast.. it would almost have to be, otherwise the network and server resources for streaming a huge number of HDTV streams would be immense.
Obviously, there is a possibility for a bunch of restrictions to be placed on what would otherwise be an incredibly powerful concept. But, they could also make really powerful changes that could change the competitive landscape.
If they look at it as a general communications platform, with additional services sold on top of it (Internet, VOIP, TV, PPV, etc.), ala-carte pricing might be more feasible. I would be very happy if I could just pay for HBO-HD, EPSN-HD, Comedy Central, and my locals. And, that would be a big blow to their cable/satellite competitors.
Actually, they tried to do this about 4 years ago with Project Pronto.
The problem was that they would invest billions of $$ on the fiber upgrades, but SBC was forced to share that with competitors. What's the point in investing billions and having your payback stolen by people who don't have to maintain the network?
They held off on the largest portion of the project until after this ruling..
As someone who lives in Korea (and enjoys the fine speeds u talk about ;) I think you will find that the reason you have lower speeds there is because of the geography of the states. I know that the reason why BB is so crap in Australia... there simply isnt enough ppl per square Km to justify lighting billions in fibre. Here i live in the pockets of another 5000 people, so its a bit more viable to have the 100+Mbit connections.
-- Fuck the grammar police.
Since the muxing is generally done over a fiber line this is actually a situation that can lead to the highest speed DSL being available. All that is needed is for a remote shelf DSLAM to be put into the neighborhood box to do the DSL->fiber conversion. SBC called their project to do this Project Pronto, it's since been mostly abandoned as a large scale goal but they still do it in places where demand warants it. Get your neighbors to sign something saying they are interested in DSL and take it to your telco, there's a good chance you can get a remote shelf put in the local box if enough people want the service.
There are 4 boxes to use in the defense of liberty: soap, ballot, jury, ammo. Use in that order. Starting now.
Cause that fiber runs from CO to CO or Data Center to Data Center. This is going to have to be fiber to the curb, or at the very least, fiber to the SLIC that services the neighborhood.
HDTVoIP with its far bigger hunger for hbandwidth.
(disclaimer : I spent 3 years as the dev lead/manager for a large streaming media company)
The bandwidth for streaming is never as high as people think. Once you start to control the whole network it gets a lot easier. If you can place caching servers in each major subscriber area and most importantly enable multi-case (which you can finaly do because you control all the routing and switches) it will drop a lot. Sure movies that are truely "on-demand" will have to be served on an individual basis, but again, local caching servers would reduce bandwidth requirements to just the last 1-2 hops.
The problem is that the output streams are all coming from a single point. Although a post below suggests that caching at various points on the network could go a long way to alleviate this problem. Why not distribute movies via P2P?
Some major broadcasters are considering this method already. Have a look at the BBC's interactive Media Player (iMP).
OK, so this may not be true streaming in the generally accepted sense, but it will still produce pretty much on demand titles that you source from a node close to you. Set your PVR to download the titles you want and away you go.
The size of HD files are possibly the biggest issue here. Some recent investigation I did revealed that WMV 9 HD files will end up at about 16 MB for 20 seconds of 24fps at a resolution of 1280x720. That's about 4GB for a 90 minute movie. You'll need a pretty big pipe into your home no matter where the file is served from.
Brought to you by the author of such childrens' classics as "Some Kittens can Fly!" and "All Dogs go to Hell."
When you make up numbers (40:1), you get wrong results. Over the air HDTV is 18 Mbps. WMV3 HDTV is under 10 Mbps.