As Verizon rolls out its FiOS fiber optic network, AT&T is hoping VDSL can satisfy consumer demand for faster connections by offering improved speeds and IPTV at a fraction of fiber's deployment cost.
Fiber's nearly unlimited capacity gives Verizon's network a huge degree of flexibility: It can broadcast hundreds of uncompressed HD channels, provide Internet access speeds upwards of 50 mbps, and yet remain relatively cheap to upgrade as future technologies come to market.
Compare this to VDSL's high maintenance costs, lower access speeds and inability to support more than a few HD channels at any given time, and what was once considered Verizon's $23 billion boondoggle seems like a very forward-thinking investment.
But are AT&T's U-verse offering and VDSL strategy dead on arrival?
Electronics researcher iSuppli Corp.'s Steve Rago says no.
While fiber can certainly run circles around anything even optimal VDSL can provide, Rago said geographical considerations can provide a strong impetus for either solution.
In areas with aerial-fed wiring or large multi-dwelling units, particularly ones presently being built, fiber can avoid many deployment costs, Rago added. In areas with lower population densities or with below ground cabling, however, the deployment cost per user skyrockets. It is largely in these markets that AT&T finds itself competing, while Verizon is blessed with the more densely populated Northeast.
Geography is not the only source playing
"AT&T has really put a bet on two things," said Vince Vittore, a senior analyst with Yankee Group Research Inc. "That compression technology is going to improve to the point that they won't need to build fiber-to-the-home. And two, that Wall Street is going to reward them for taking the piecemeal approach."
Both are uncertain wagers. While AT&T's stock was initially buoyed on the news of the VDSL rollout, investors seem to be rethinking their assessment on the two telecom giants' strategies. Meanwhile, future compression technologies may have to improve drastically in order to make AT&T's playbook work.
HDTV currently requires about 8 to 10 mbps per channel being streamed, and U-verse's minimum standard for connectivity is 25 mbps. A second HD stream, scheduled for 2008, will test AT&T's ability to deliver as promised.
Already, The Wall Street Journal has reported that the stress of pushing U-verse's network to its limits with a single high-definition channel and regular-definition channels, plus extra features like remote recording, has lead to pixilation problems in early markets.
As HDTVs become the norm, AT&T will need to figure out how to satisfy the market demand for high-definition content while continuing to provide 6 mbps Internet access.
If HD video issues are not resolved, AT&T will still have a few options. Just as it is currently building on its ADSL2 network to start providing VDSL, it can build upon the VDSL infrastructure to expand into a fiber-to-the-home network. It is an expensive proposition but the one most likely to meet future bandwidth requirements.
Another option, currently being used in AT&T's Homezone offering, is to pair VDSL Internet access with satellite HD streaming. VDSL could be used to deliver on-demand HD content while leaving most of the bandwidth free for higher Internet access speeds.
For AT&T, the major drawback is that a rather large chunk of subscription fees would go to the satellite operator. But Wall Street has been predicting that the telecom might simply buy up an existing satellite operator.
Tom Nolle, president of consultancy CIMI Corp., said that approach is the winning combination for AT&T.
"What U-verse does is take the things cable does well, and does them badly," he said. However, he added that following Verizon's strategy of pure fiber-to-the-home would not have worked either because of the high cost-per-customer the company would have faced in most of its regions.