SI
SI
discoversearch

We've detected that you're using an ad content blocking browser plug-in or feature. Ads provide a critical source of revenue to the continued operation of Silicon Investor.  We ask that you disable ad blocking while on Silicon Investor in the best interests of our community.  If you are not using an ad blocker but are still receiving this message, make sure your browser's tracking protection is set to the 'standard' level.
Technology Stocks : LAST MILE TECHNOLOGIES - Let's Discuss Them Here -- Ignore unavailable to you. Want to Upgrade?


To: MikeM54321 who wrote (2067)9/26/1998 10:38:00 AM
From: Kenneth E. Phillipps  Read Replies (1) | Respond to of 12823
 
Article from Redherring Online re Cable vs DSL:
IN BROADBAND WARS,
WILL DSL SERIOUSLY
CHALLENGE CABLE?

By Peter Jerram
Red Herring Online
September 23, 1998

The rise of the Internet has fueled a tremendous demand
for faster access by both business users and consumers.

On the home front, Internet users are maxing out on
56K modems, with many poking along at half that
speed, or less. Businesses have access to higher speed
leased lines, but pay steep tariffs to telcos for the
privilege. Many feel the constrained bandwidth is
stunting the development of Internet content and
applications.

"Once we enable high speed, we'll enable all kinds of
new applications, and no one really knows where that's
going to take us," says Lisa Pelgrim, senior analyst with
Dataquest in San Jose.

But where it's taking us is a question that's ahead of
itself; where we are right now in terms of competing
broadband opportunities is still at issue.

Worth the wait?
Several companies, led by @Home (ATHM) and Time
Warner's (TWX) Road Runner, are touting Internet
access services using the television cables already
installed in a majority of American homes. By using this
existing infrastructure, fast, cheap Internet access could
be available to millions of homes.

After two years of trials, however, cable deployment is
spotty, with fewer than 400,000 subscribers in the U.S.
and Canada. In the headlong rush of the computer
industry, progress has seemed glacial at best.

"People I talk to in Silicon Valley don't believe [cable]
exists because if it did it would be there first," says
Michael Harris, an analyst with TeleChoice in Phoenix.

Deployment was delayed in part as cable operators
waited for modem standards, and for the price of
modems, to fall to a retail price of under $300.
Operators also underestimated the infrastructure
challenges in rolling out Internet access.

"They got into the field and realized it was going to be a
little harder than they thought," says Harris.

Going two-way
Industry leader TCI (TCOMA) feeds about 20 million
homes, but only about 1.2 million of those are now
two-way capable, according to TeleChoice. Despite the
condition of TCI's network, AT&T (T) paid $28 billion
for TCI to get at it. "AT&T purchased infrastructure,"
says Ms. Pelgrim.

With interoperable modem standards finally in place and
prices dropping into the consumer comfort zone, TCI
and other cable operators are now on the cusp of
large-scale deployment -- a central proviso of AT&T's
acquisition.

"The entire merger was predicated on pervasive
deployment of cable-modem technology," says Henry T.
Nicholas, chairman and CEO of broadband chipmaker
Broadcom (BRCM). "AT&T paid a premium with the
understanding that two way upgrading was underway."

Mr. Nicholas says that TCI will have its network 50
percent two-way enabled by next year, and he should
know. In a recent announcement, Broadcom revealed
that it has integrated all of the cable-modem media
access control (MAC) and physical layer transmission
functionality into a single chip instead of three.

The company said this will mean that next-generation
cable modems will be smaller and cost less for
consumers while being able to deliver data, digital video,
telephony, and Internet access at speeds up to 56
megabits per second, or 1,000 times faster than
standard 56k voice-band modems.

Moreover, Broadcom's new chip has the circuitry for
phone connections through TV cable, which would
allow cable companies to offer Internet access at rates
more competitive with phone companies. According to
The Wall Street Journal, the new chips are expected
to cost around $50 each in small allotments --- about
the same as the current three-chip set -- but maybe half
that or less in large orders.

This announcement may just give cable the nudge, and
even the market validation, needed for cable to become
the more adopted broadband standard.

Will the telcos ever get it?
Meanwhile, the telcos are pinning their hopes on digital
subscriber lines (DSL), a means of delivering high
bandwidth digital signals over ordinary copper phone
lines.

In the past several months, most of the regional Bell
operating companies (RBOCs) have announced DSL
service. "The pressure is on the phone companies to
keep up [with cable]," says Mr. Harris.

Telcos are focusing mainly on small and medium-sized
businesses, although most carriers also offer plans for
residential customers. DSL speeds range from 128K to
8 Mbps. The high end of that range places DSL
squarely in the realm of traditional leased lines, which
are far more costly. The telcos are offering a flavor of
the technology called asymmetric DSL, which employs
higher bandwidth for receiving than for sending
information.

Early indications are that the carriers will run into some
of the same infrastructure issues that have plagued cable
operators. DSL service is only available to customers
within about three miles of a telco central office, which
eliminates more than half of otherwise eligible customers.

Telcos are also finding that the condition of local loop
wiring limits its ability to carry high speed data reliably.
Ubiquitous DSL coverage may require reworking the
copper wire infrastructure, an undertaking that was not
in the plan.

"Telcos are in a catch-22 -- do they really want to
reconfigure their plant for DSL, or do they do something
else like fiber to the curb?" say Mr. Harris.

Many in the industry have found themselves up against
the fabled telco bureaucracy as well.

"The RBOCs have moved slower than we would have
liked, it's been refreshing to work with the cable
companies by comparison," says Heidi Clark, chairman
and CEO of broadband equipment supplier Cayman.

Several analysts also questioned whether the telcos
would seriously consider cannibalizing their profitable
leased-line businesses. Competitive local exchange
carriers (CLECs), which lease lines from the RBOCs,
do not have such a conflict, and several, including
Covad Communications, are offering DSL service to
Internet service providers. However, since the CLECs
depend on RBOCs for copper wire, their fortunes are
tied to the telcos.

Disagreeing to agree
Despite the hurdles, DSL appears to have stirred up
considerable interest in the computer industry.

In June, networking-equipment vendors Cisco (CSCO),
Ascend (ASND) and Cabletron (CS) all acquired or
invested in DSL technology suppliers. And a consortium
of companies called UAWG, which includes Microsoft
(MSFT), Compaq (CPQ), and Intel (INTC), is working
on a set of DSL standards that will make it easier to
install DSL, and will require less hardware.

However, without a current agreement by all of the
RBOCs on a common flavor of DSL, and given the
recent consensus of the computer industry on modem
standards, DSL may even fall further behind cable at this
point.

"There is no widespread agreement among the telcos,"
says Mr. Nicholas, "They're trying, but it's fractious and
contentious."

Still, if the telco industry is able to reach consensus
quickly, DSL may make inroads as the cable operators
spend time building out their own networks.

"Cable has a head start, but is nowhere near ubiquitous,"
says Ken Hoexter, a vice president at Goldman Sachs.
"There's room for multiple ways of reaching the end
user."

Microsoft is one heavyweight in particular which is
characteristically mindful of those opportunities. As the
cable and telco players decamp to their respective
corners, Microsoft is hedging its bets as a DSL booster
-- who also happens to have a $1 billion investment in
cable operator Comcast (CMCSA).

Copper or cable? You make the call.



To: MikeM54321 who wrote (2067)9/28/1998 8:53:00 PM
From: DenverTechie  Read Replies (2) | Respond to of 12823
 
Mike, it depends on your definition of "threat" to TITAN's dominance..

All optical networks are the next big thing in high capacity networking, no doubt about that. The TITAN has taken the first step in addressing optical networks by outfitting its cross connect with optical interfaces as we have previously discussed. That means that no demultiplexing is needed in front of the cross connect to do an optical to electrical conversion before the cross connect can happen. But still has to do it internally, but at a much more cost efficient and space efficient rate than having separate mux and de-mux external to the cross connect.

Bottom line is that it is still a TDM based machine that can interface optics based networks at the port level. To date that has been the bread and butter network. Traditional DS1, DS3 networks (yes, there are still lots of them out there) need TDM based cross connects interfaced at the electrical level and SONET based networks that need cross connects interfaced at the optical level. These are the 2 prevailing network types out there today and Tellabs has the cross connect that is the best for both types of networks.

The all optical network is an EMERGING type of network that does not exist today. Today, the TITAN cannot carry light signals through its primary matrix, only electrical. Much of the talk of the demise of digital cross connect suppliers like Tellabs is that we are starting to see the advent of structures such as DWDM and OADM (optical add/drop multiplexers) within the next year or two. These are some of the building blocks to get to the all optical network, but it ain't here in total yet. And keep in mind that it is an unproven concept with untested equipment and some vapor equipment.

Having worked for large telcos in the past, they are typically reluctant to put high capacity, reliability dependent voice and data on new types of networks at the beginning. Typically, a major player like AT&T, MCI or an RBOC wait several years to see how the technology matures with early adopters who absorb all the pain of the early deployments. There is not a single OADM in operation today that I know of (could be trials of course running somewhere to test prototypes).

If I know Tellabs, they usually do not let trends like this pass them by. They were the first to have optical interfaces on their cross connect while Lucent was still thinking about doing it and to this day does not provide on certain cross connect products. It will not be long before they have products to meet the optical network demand. They wanted the Ciena deal to kick start the development since Ciena had been working in this area already. But Tellabs will get there, it will take longer than they planned with Ciena, either through acquisition or internal R&D.

In the meantime, an article in the 8/15/98 issue of America's Network on page 27 states " Despite all the talk about SONET beginning to disappear, SONET vendors are enjoying healthy growth and the carriers that are building new fiber optic networks are using the technology in those networks. In addition, many carriers already are deploying four-fiber OC-192 (10 Gbps) bidirectional line switched rings (BLSRs) in their networks." NORTEL, Alcatel, Fujitsu and Hitachi offer OC-192 BLSR technology today to provide additional reliability for long haul fiber optic networks.

Later on the article states "the SONET market in North America is continuing to grow rapidly. The market was estimated at $3.1 billion in 1996 and will be worth more than $4.2 billion this year. Growth won't flatten until 2000 or 2001, years in which the market will be worth $5 billion." Depending on how optical networking develops, I have seen estimates for 2002 as high as $6.3 billion from credible sources. Being conservative by nature, I would say Tellabs has the best cross connect product in a $5 billion market through 2001 and that's not bad compared to an all optical market of maybe $0.01 billion now.

And I bet that Tellabs will be developing the optical products that will meet the needs of networks that will interoperate on the seven layer level. I don't see the need for that in less than 5 years from now. If I know Tellabs, they won't be sitting around. There will be press releases 5 years from now from Tellabs that they have fully optical cross connects and that SONET-like optical layer protection is working in the field. Initially, I see this as an optical cross connect at 1300 nm short reach OC-48. True interoperability of optical layer protection from different vendor's gear is probably about 10 years out (I agree with the article's author on this).

In the meantime, look at the real world and real deployments. MediaOne put in SONET for its network upgrade in Atlanta using super headends and who's cross connect? Tellabs.

Who else is building SONET based networks currently? Check out this list and estimated completion investments:

NextLink Communications: investment - undisclosed
Williams Network: investment - $3 billion
Frontier Corp: investment - $500 million
IXC: investment - $1 billion
Level 3 Communications: investment - $8 to 10 billion (SONET at first)
Qwest Communications: investment - Undisclosed

The cross connect of choice for most of these OC-48 and OC-192 networks? Tellabs (although some of these operators have yet to announce their SONET infrastructure vendors).

There is a delicate balancing act between providing products currently in demand while preparing for the next wave and having products ready for it when it arrives from a practical stand point.

I would say that rumors of TITAN's death have been greatly exaggerated.