AT&T announces $336 million video TV plan, other upgrades

[ticket]

(Hartford,CT.-AP, June 22, 2006 3:52 PM) _ AT&T says it's nearly ready to roll out its new technology to provide television programming over phone lines.

The company announced today that it's spending $336 million to launch the service. Officials say it'll be ready for the first Connecticut subscribers by the end of the year.

The service provides television programming, movies on demand, sports scores, stock quotes and other information.

It shows up on a regular TV screen . The difference is that it's sent through phone lines, rather than through cables, as current cable TV providers offer.

AT&T says that gives the company the chance to offer more variety, and an alternative to being locked in to cable company rate increases.

Several cable companies and Attorney General Richard Blumenthal have asked the state's Department of Public Utility Control to prevent the service for now.

They say AT&T should have to get a cable franchise. But DPUC commissioners ruled earlier this month that the company is exempt from that requirement, because the technology is so different than regular cable TV.

There's no word yet on how much the service will cost subscribers and what will be included in the packages.

http://www.wtnh.com/Global/story.asp?S=5066758

 

[Brewer4]

Sweet. Cant wait to see which packages they offer. There is so much unused fiber and lines when SNET had that cable TV service. I am interested in what they offer especially if they can offer more high speed internet choices. I would give up my Cable Modem and Vonage if something is comparable. I would also like to get my Mom off basic analog cable.

 

[hancox]

Gimme gimme.

 

[red hazard]

One HDTV Channel

The problem with AT&T's broadband implementation plan is that it will provide only 1 HDTV channel and 2 or 3 so-called high quality (read: standard definition)
channels. That is a horrible business model as many households in a few years will have more than one HDTV set as the prices continue to drop. Folks will not be satisfied with watching the same HDTV program on all sets. Until I see test results on bonded twisted pairs that indicate sufficient bandwidth for at least 3 HDTV channels, I will be watching my AT&T stock very closely. It's doubtful they can increase the bandwidth on twisted copper pairs my much since it was initially designed to only carry 3 KHz. It looks like a terrible business model IMO and cannot compare to Verizon's FiOS which will bring fiber to the residence and use the coax within the residence.

 

[jegrant]

I might be in the minority at this type of site, but HD is not much of a concern in my household. We just want a wide selection of SDTV channels at a good price.

While I'd love to have FiOS, living in an AT&T/SBC/Ameritech ILEC area, I doubt I'll get FiOS anytime soon. (I also don't live in a big city that Verizon might choose to "overbuild", either.)

It is also possible that by the time more people / more TV sets need simultaneous HDTV signals, that compression technologies may have improved by then, at least enough to provide HDTV for 3 or 4 unique streams to each home through copper. (Or, by then, AT&T may have simply replaced that copper and gone entirely fiber.)

 

[Brewer4]

They are planning on upping the HDTV. Its the initial rollout that is only 1. I would rather they go slowly and get it right then try to do everything at once. I watch my HD OTA for about 70% so 1 HD stream might not be a limitation in my house. I just hope their stuff works with my Windows Media Center or it works with HD OTA or it is sunk before it gets in my house.

All I know is its coming to Connecticut, the DPUC has made it easier, its another choice and player and it should put some price pressure on the others. Its good and I will be glad to see it available whether I use it or not.

 

[ticket]

heard from a good source about AT&T U-verse He said.....

I think we're launching it here in the fall. I know Norwalk,CT. is one of the
first towns we're turning up with the new IP TV and we already have a trial
going there. Basically you'll have a 20 meg pipe going the your house that
delivers high speed data, cable tv and phone over the phone line like DSL

 

[cheer]

The problem with AT&T's broadband implementation plan is that it will provide only 1 HDTV channel and 2 or 3 so-called high quality (read: standard definition)
channels. That is a horrible business model as many households in a few years will have more than one HDTV set as the prices continue to drop. Folks will not be satisfied with watching the same HDTV program on all sets. Until I see test results on bonded twisted pairs that indicate sufficient bandwidth for at least 3 HDTV channels, I will be watching my AT&T stock very closely. It's doubtful they can increase the bandwidth on twisted copper pairs my much since it was initially designed to only carry 3 KHz. It looks like a terrible business model IMO and cannot compare to Verizon's FiOS which will bring fiber to the residence and use the coax within the residence.

First...as others have pointed out, that's 1 HDTV stream initially; more to come.

Second...I'd guess the number of residences with multiple HDTV sets in use at the same time is rather small.

Third...it isn't just about the bandwidth, and comparing bandwidth between FIOS and AT&T's product is comparing apples to concrete. FIOS is using the same basic technology that cable is -- so the line into the house has to carry all channels at once. IPTV only needs to carry active streams (channels you're watching) so the bandwidth requirements are different. Remember too that as compression technologies improve, they can fit more...all they have to do (more or less) is push new codecs down to the boxes and whoomp, there it is.

AT&T is actually (IMO) using a smarter business model: instead of making a huge investment all at once, you make a decent investment and go after the mass market, then as you're doing that you can continue to do the physical upgrades necessary for the higher-end customers.

Finally...I hear what you're saying about the capacity of twisted-pair...but then, I can remember when a 2-wire phone line would "never" exceed 9600bps.

 

[cheer]

I think we're launching it here in the fall. I know Norwalk,CT. is one of the first towns we're turning up with the new IP TV and we already have a trial
going there. Basically you'll have a 20 meg pipe going the your house that
delivers high speed data, cable tv and phone over the phone line like DSL

Yeah, they're doing the initial rollout with VDSL technology which gets somewheres around 20-25 mb/sec. And yeah, I think there are bits of CT that are part of AT&T because AT&T (or SBC) bought some small LEC up there ages ago.

Anyway...you can see roundabout pricing already, because the service has "launched" officially in Texas. Looks like it starts at $69/mo for 3 receivers (one DVR), around 100 channels (incl. locals), and 1.5mb/sec internet. The top package seems to be $114/mo for 3 receivers (one DVR), HBO, Cinemax, Showtime, Starz, etc (around 30 movie channels total), some extra sports stuff, and 3mb/sec internet.

It's certainly competitive. Now I'm sure they're going to push bundles with local phone service, but IMO they'd be smarter to push bundles with AT&T Callvantage (VoIP). And once the Bellsouth merger is complete (should be around the end of the year), expect bundles that include AT&T Wireless (which is what Cingular will be renamed).

I know some folks at AT&T; I'm going to try and get on the beta team for the Chicago area. If I do, I'll post my thoughts.

--c

 

[red hazard]

First...as others have pointed out, that's 1 HDTV stream initially; more to come.

Second...I'd guess the number of residences with multiple HDTV sets in use at the same time is rather small.

Third...it isn't just about the bandwidth, and comparing bandwidth between FIOS and AT&T's product is comparing apples to concrete. FIOS is using the same basic technology that cable is -- so the line into the house has to carry all channels at once. IPTV only needs to carry active streams (channels you're watching) so the bandwidth requirements are different. Remember too that as compression technologies improve, they can fit more...all they have to do (more or less) is push new codecs down to the boxes and whoomp, there it is.

AT&T is actually (IMO) using a smarter business model: instead of making a huge investment all at once, you make a decent investment and go after the mass market, then as you're doing that you can continue to do the physical upgrades necessary for the higher-end customers.

Finally...I hear what you're saying about the capacity of twisted-pair...but then, I can remember when a 2-wire phone line would "never" exceed 9600bps.

I disagree.

First DSL technology is limited to 20 - 30 MBS. That is insufficient for two HDTV channels plus there other simutaneous services.

Second there are many members here with more than one HDTV satellite receiver. It won't be long before multiple HDTV sets in a houseshold will be as common as multiple standard color TVs are now. When color TV prices came down, B&W TVs were a thing of the past. That scenario will be repeated.

Third IT IS about bandwidth. Putting several HDTV channels on DSL is like trying to stick a dump truck down a gopher hole. Shannon's law shows the relationship between channel bit rate and available bandwidth in addition to background noise. (C = W log2(1 + S /N )). TWP is very bandwidth limited compared to RG6 cable. Since there is little one can do about the intrinsic noise, one must increase the bandwidth to increase the bit rate.

I don't see how this can be a smart business model as AT&T is not competing against Verizon - - they are competing against the cable companies. That is the comparison one should be concerned about. Short distance coaxial cable runs have huge bandwidth compared to TWP.

Regarding your 9600bps comparison, it's just not valid. Shannon's Law indicated that over a carrier system voice channel (about 400 Hz to 3300z) that the maxium bit rate would be about 40 KBS. Voiceband modems eventuall achieved about 38 KBS -- close to the theoretical maximum. The so-called 56KBS modems surpassed this on the download side only because there was no bandpass filter in the download channel's profile from the ISP. The upload channel was/is approx 31 KBS. Higher bandwidths obtained by DSL over TWP is due to the fact that the bandwidth of TWP is far more than that of a voice channel derived from a carrier system.

 

[grendyl]

I read somewhere the cable companies are asking for a stay w/ the argument that they could cherry pick good areas (even though their local govt approved state franchise agreement) Some guy at at&t said they may pull the plug in Conneticut if the stay is granted, saying something like they have limited funds and they only want to be in states that want them. That was like the 28th or so. Is there an update on that sitution? The fact they could drop the state after such a huge investment was surprising to me.

 

[jrollo]

Well, according to some posts at http://www.uverseusers.com/ , users are syncing their lines at 25 megabits, but the lines are capable of 40-50. I think AT&T prefers to use conservative synch rates for the technology, but the truth is that if they start moving the fiber closer to the homes, they should be able to start increasing bandwidth. Then with pair bonding, we might be able to approach 50-100 mb/sec. That should be enough to get them by for a few years. Then as costs for running fiber decrease, they can start deploying it all the way to the home. I'm just guessing that is their strategy (get by for now, prove the economic viability to their investors, then do it right all the way to the home when costs are lower).

 

[cheer]

I read somewhere the cable companies are asking for a stay w/ the argument that they could cherry pick good areas (even though their local govt approved state franchise agreement) Some guy at at&t said they may pull the plug in Conneticut if the stay is granted, saying something like they have limited funds and they only want to be in states that want them. That was like the 28th or so. Is there an update on that sitution? The fact they could drop the state after such a huge investment was surprising to me.

This fight has been going on for a long time.

Cable companies want AT&T to negotiate local franchise agreements. They feel that since they had to, everyone should. AT&T has no problem paying the local fees but doesn't want to take the time to negotiate all those agreements.

For the most part, AT&T seems to have won out. The US House has even approved a national franchise bill (it's in the Senate now). But AT&T will certainly focus on states with the least resistence.

 

[Brewer4]

Connecticut has in the past been a strong proponent of more cable competition. There are tons of fiber lines and existing infrastructure that should make it a good place to start the rollout. The ruling just helps the legal mess. I would expect to see service by the end of year with more widespread rollout next year. For me anything is welcome especially if it increases bandwidth for internet usage.

 

[red hazard]

Pair Bonding?

Well, according to some posts at http://www.uverseusers.com/ , users are syncing their lines at 25 megabits, but the lines are capable of 40-50. I think AT&T prefers to use conservative synch rates for the technology, but the truth is that if they start moving the fiber closer to the homes, they should be able to start increasing bandwidth. Then with pair bonding, we might be able to approach 50-100 mb/sec. That should be enough to get them by for a few years. Then as costs for running fiber decrease, they can start deploying it all the way to the home. I'm just guessing that is their strategy (get by for now, prove the economic viability to their investors, then do it right all the way to the home when costs are lower).

It makes sense moving the nodes closer to the customer premise would increase bandwidth by lowering the R, C and L in the low pass filter effect of TWP. AT&T's Chris Rice is now (June 2006) stating 30 MB/s @ 2500 Ft (still not enough for two HDTV channels and HSI). However, I got two questions.

1. How does pair bonding increase the bandwidth? After all, it essentially doubles the inter-electrode capacitance thereby increasing the low pass filter effect. (I realize twisting the pairs reduces the C to some degree). It would be very interesting to see a technical article explaining overall how it will work. I see lots of folks in forums equating pair bonding with doubling the bandwidth capacity but never with an explanation or citing an independent authhoritative source.

2. Secondly, what technology are you refering to that would decrease the cost of laying fiber to the premise in the future? What I have read is that a substantial part of AT&T's territory has buried utilities (as comparied to Verizon) and that labor is the major cost factor.

With the insatiable appetite for higher bandwidth applications, I still have doubts that anything other than FTTP is a wise business model. With deference to compression algorithms, check the HDTV forum - - there are constantly numerous complaints about PQ, artifacts, etc. Many articles also stating that MPEG-4 won't provide true HDTV. With fiber, compression won't even have to be an issue.

EDIT: Disregard the question about bonded pairs. "Bonded" is actually a poor choice of words. Bonding in the ADSL+ usage refers to a form of inverse multiplexing. For example, a 50 MB/s bit stream at the node is applied to two cable pairs in two disparate bit streams and then reconstitued at the user premise. That would almost double the capacity (more overhead required). This technique should make AT&T competitive with cable. Some outside plant guru could probably address the availability of second (or more) cable pair availability - - particularly in older neighborhoods.

 

[ticket]

Connecticut has in the past been a strong proponent of more cable competition. There are tons of fiber lines and existing infrastructure that should make it a good place to start the rollout. The ruling just helps the legal mess. I would expect to see service by the end of year with more widespread rollout next year. For me anything is welcome especially if it increases bandwidth for internet usage.

A friend of mine works for AT&T and he said Norwalk is going to be first to rollout in the fall for Connecticut

 

[jrollo]

2. Secondly, what technology are you refering to that would decrease the cost of laying fiber to the premise in the future? What I have read is that a substantial part of AT&T's territory has buried utilities (as comparied to Verizon) and that labor is the major cost factor.

I can't recall what the article stated (it was from many months ago), but I believe that it had to do with the cost of the cable, as well as certain equipment. But if labor is the major cost, then the savings would be negligible.

I would love FTTP, but I guess I'm of the opinion that I'll take what I can get for now.

 

[Brewer4]

A friend of mine works for AT&T and he said Norwalk is going to be first to rollout in the fall for Connecticut

Any idea if Marlborough is on the plate for 2007?