dslreports logo
site
spacer

spacer
 
   
spc
story category
Deflating 700-900 Mbps DSL Hype
Promising, but years out and limited in scope
by Karl Bode 09:47AM Wednesday Nov 03 2010
As we noted last week when discussing some of the new, ultra-fast DSL services being hyped by hardware vendors, consumers should probably temper their enthusiasm for 800 Mbps arriving anytime soon. There's a long list of caveats with these services. Many telcos are upgrade phobic, many homes lack quality copper or extra pair, much of these technologies haven't left the lab yet, and like all DSL these new incarnations are distance constrained, meaning most users won't be able to get them whenever they do arrive. Industry analyst Dave Burstein does a nice job separating fact from fiction and noting that most of these new DSL approaches from Ikanos, Huawei, Alcatel Lucent and Nokia will first be seen in Europe several years from now, though most users will remain out of range.

view:
topics flat nest 

Swordfish II
Watching A Dream
Premium
join:2002-05-12
Cloud 9

To sum it up

Essentially cable is still and for the near future will be the most cost effective and fastest service.
--
I'm not going there to die. I'm going there to see if i really am alive.

ArrayList
netbus developer
Premium
join:2005-03-19
Brighton, MA

Re: To sum it up

no fiber is the most cost effective.

Duramax08
To The Moon
Premium
join:2008-08-03
San Antonio, TX

Re: To sum it up

Either way copper sucks.

antdude
A Ninja Ant
Premium,VIP
join:2001-03-25
United State
kudos:4
Reviews:
·Time Warner Cable

Re: To sum it up

said by Duramax08:

Either way copper sucks.
Yeah, dial-up sucks!
pepe7

join:2003-08-25

1 recommendation

Incorrect, Mr. Gore (ROFTL).

In the context of what's already available in the market, fiber is faster but not more cost effective. Cable HSI already covers larger swaths of the country, even areas that fiber will likely never see. For now at least that makes it very cost effective.

ArrayList
netbus developer
Premium
join:2005-03-19
Brighton, MA

Re: To sum it up

up front costs are less than the cumulative upgrades you will need with coax.

vpoko
Premium
join:2003-07-03
Boston, MA

1 recommendation

Re: To sum it up

But cable companies cannot pay you on Tuesday for a hamburger today; there's a big difference between paying now and paying later. Of course we all agree fiber is more future-proof and will have lower maintenance costs in the long term, but tearing up all existing coax and replacing it with fiber would certainly not be economical.

ArrayList
netbus developer
Premium
join:2005-03-19
Brighton, MA

Re: To sum it up

why would you tear it up? leave it in place and put fiber in as you can. at the rate they go right now, they don't do any.
pepe7

join:2003-08-25

Re: To sum it up

Geez. You live in the north shore like I do. Apparently you have forgotten entirely about the labor costs involved. For this reason around Chicago it is currently extremely cost prohibitive to put in a FIOS-like system.

ArrayList
netbus developer
Premium
join:2005-03-19
Brighton, MA

Re: To sum it up

of course the captial involved in putting in a project like that will be a lot, but unless they figure out how to break the laws of physics they will never have to upgrade after the fact.
Sammer

join:2005-12-22
Canonsburg, PA
said by pepe7:

Cable HSI already covers larger swaths of the country, even areas that fiber will likely never see. For now at least that makes it very cost effective.
Actually some cable companies have already realized that RF over Glass is sometimes the least expensive way to expand beyond their existing cable lines so fiber has already started to serve some homes that are likely to never see the external copper of HFC. Admittedly RFoG may be a niche technology but it points to the fact that even the cable companies realize that fiber to the home is the eventual future.
pepe7

join:2003-08-25

Re: To sum it up

I read about RFoG at one point. It looks like it will still support DOCSIS too. The following might be of interest to someone here-

»www.lightwaveonline.com/mso-opti···082.html

maartena
Elmo
Premium
join:2002-05-10
Orange, CA
kudos:3

1 recommendation

Fiber is probably be able to deliver the fastest speed. But for the next 10 years, the current cable networks are probably more cost effective.

First, the cables are already in place, and fiber has been installed to the neighborhoods. (If you have digital television and any HD channels, you have a TV company fiberoptic cable less then 2 miles away, likely less then a mile).

Second, fiberoptic cables are expensive. This is why Verizon isn't doing a massive deployment and just hook everyone in a city up to fiber, regardless of whether they have TV, Internet or Phone or not. They did this with telephone back in the day, and just connected all homes to copper regardless of whether they actually had this newfangled invention called a tele-phone. Instead, they do it on a case by case basis.

I agree COMPLETELY that fiberoptics is the future, and can't wait till all houses are connected with a 10 Gbps pipe.

But as far as cost-effectiveness goes, the DOCSIS system has allowed for easy upgrades and downwards compatibility. A cable company can take DOCSIS 1 and DOCSIS 2 equipment out, and replace it with DOCSIS 3 equipment, and still serve all of their DOCSIS 1 and 2 clients....with the new equipment whereas both Verizon with FIOS and ATT with UVerse have to maintain 2 separate networks and 2 separare systems. One for their old DSL clients, one for their new FIOS and UVerse clients. And that.... is very cost in-effective.
--
"I reject your reality and substitute my own!"
Kearnstd
Space Elf
Premium
join:2002-01-22
Mullica Hill, NJ
kudos:1
Fiber is the best and should be aimed at for the future. but the HFC networks are much more expandable than the telco lines. and the cable company has another advantage. they already have fiber to the nodes, as such if they want to switch to FTTH I am sure they have less work ahead of them.
--
[65 Arcanist]Filan(High Elf) Zone: Broadband Reports

CylonRed
Premium,MVM
join:2000-07-06
Bloom County
But not the best where I live - so it is DSL for me...

n2jtx

join:2001-01-13
Glen Head, NY

DSL and Distance

Short of rolling out DSLAM's to the neighborhoods so that the total length is only a few hundred feet to the premises, I seriously doubt this technology will see the light of day. Considering what they would have to spend to do it, fiber would be the most cost-effective solution in the long run. Plus, as the article points out, poor quality copper lines would be an issue as well.
--
I support the right to keep and arm bears.
elray

join:2000-12-16
Santa Monica, CA
Reviews:
·Time Warner Cable
·EarthLink

Re: DSL and Distance

said by n2jtx:

Short of rolling out DSLAM's to the neighborhoods so that the total length is only a few hundred feet to the premises, I seriously doubt this technology will see the light of day. Considering what they would have to spend to do it, fiber would be the most cost-effective solution in the long run. Plus, as the article points out, poor quality copper lines would be an issue as well.
Fiber-optic probably is the most cost-effective, if you're looking out 30 years, but telco can't use that calculus.

Verizon took the chance and is paying the heavy costs of installing fiber-optic @ $4k/household, and so far, many people are willing to pay the monthly freight at $50-70 for internet. But if the FCC takes a Left turn, ends forbearance, and starts dictating wholesale rates, Verizon's balance sheet would suffer - they therefore have to make their money in the short run.

Am I making the case for regulated rates of return on a natural monopoly? Some might suggest so - and that isn't the worst possible outcome* - but it would cost a lot more than we pay today, given the cozy nature of government with regulated utilities, and the creative accounting that follows.

*(Worst possible outcomes are no service, or USPS-SEIU-IBEW Broadband-For-All.)

n2jtx

join:2001-01-13
Glen Head, NY

Re: DSL and Distance

Years ago I had DSL at home from COVAD. It was run on a dedicated line and I was lucky to get about 640K/128K as I am on the fringes of the service from my area's CO (which is odd considering this is a heavily populated suburban area of Long Island). It was only when a thunderstorm knocked the line out that I went with Cablevision as COVAD refused to have the line fixed (thankfully in retrospect). In my office, we use DSL as it provides more options to us than Cablevision does. The DSL carrier is Verizon but we use a third-party ISP that provides us better service.
--
I support the right to keep and arm bears.
AndrewW

join:2009-03-07
Toronto, ON
kudos:1

1 recommendation

The $4k/household was the cost when VZ started to deploy Fios. The costs have come way down since then. Its about $700 per house passed with about another $500 to actually connect a house to the network.
sonicmerlin

join:2009-05-24
Cleveland, OH
kudos:1
4k/household? Why do you lie? Is your argument not solid enough to hold its own?
elray

join:2000-12-16
Santa Monica, CA
Reviews:
·Time Warner Cable
·EarthLink

Re: DSL and Distance

said by sonicmerlin:

4k/household? Why do you lie? Is your argument not solid enough to hold its own?
$4K/household was the original projection given in a variety of documents when FIOS was proposed, and cited for years as an ongoing cost. I would suggest that the number is even higher now, once the low-hanging fruit has been picked - rural deploying costs more. But regardless, one need only look at AT&T's choice, not to deploy fiber-optics, to infer that the cost is substantial compared to reinventing copper.

I am not one to believe telco's numbers at face value - $4K could be inflated, or it could be understated. Based on the activity I've seen with FIOS rollouts - taking months to a weeks' worth of work, with contract crews imported from the opposite coast, and the nascent failure issues with some of the installs and the STBs, I could see where VZ might see even higher costs.

If you have data which supports a different figure, please present it. I have nothing vested in $4K.

battleop

join:2005-09-28
00000
"Short of rolling out DSLAM's to the neighborhoods so that the total length is only a few hundred feet to the premises,"

The only way this will happen is if the DSLAMs are very compact in size (like a fiber splice box or lunchbox) and the cost of each DSLAM becomes very cheap.
sparc

join:2006-05-06

delete the hype and it's still not bad....

"Many homes will be able to get 50 and 100 meg downstream."

the same article mentions that as well. Most people would be happy with even half that at a reasonable price.
mlcarson

join:2001-09-20
Los Alamos, NM

Re: delete the hype and it's still not bad....

Is many defined as more than 10?
sparc

join:2006-05-06

1 recommendation

Re: delete the hype and it's still not bad....

same goes for all of karl's claims and the use of many.... "Many telcos are upgrade phobic, many homes lack quality copper or extra pair"

i'll take Karl's word on the telcos phobia for upgrades, but i haven't seen any data about the poor quality copper and lack of an extra pair.

I could just as well say many homes have good quality copper and have an extra pair available.
sonicmerlin

join:2009-05-24
Cleveland, OH
kudos:1
said by sparc:

"Many homes will be able to get 50 and 100 meg downstream."

the same article mentions that as well. Most people would be happy with even half that at a reasonable price.
And how much do you think they will lower it? DOCSIS 3 has to share 160 mbps downstream with a large number of users. The more high bandwidth consumers, the more node splits the cable company has to do, which costs them money. If the number of users per node reaches 32, it makes much more economic sense for them to just lay fiber out all the way.

In other words cable companies have a huge incentive to prevent the mainstream user from accessing those premium tiers. They are meant solely to extract exorbitant prices from rich power users willing to sacrifice thousands of dollars a year on speeds that fiber could prevent at a fraction of the cost.

Packeteers
Premium
join:2005-06-18
Forest Hills, NY
kudos:1
Reviews:
·Time Warner Cable

UF.DSL perfect for apartment buildings

if you could deliver 100 Mbps over two pair of untwisted copper withing 100 yards of the main building hub - you could light up 80% of apartment buildings within a year, and only need pull one line of fiber between the CO and each building. people who only have 2 pair coming into each apartment, could simply use VoIP or their celphone, which is probably what they are doing already.
bogey7806

join:2004-03-19
Here
kudos:1

Re: UF.DSL perfect for apartment buildings

said by Packeteers:

if you could deliver 100 Mbps over two pair of untwisted copper withing 100 yards of the main building hub - you could light up 80% of apartment buildings within a year, and only need pull one line of fiber between the CO and each building. people who only have 2 pair coming into each apartment, could simply use VoIP or their celphone, which is probably what they are doing already.
Pretty much. The telcos could have leveraged stuff like this if years ago they started actually making their copper plant friendlier to high frequency systems. But alas, they'd replace a 26ga cable with a 26ga cable. So now the only real people benefiting from xDSL advances like this are recent apartment dwellers who usually have at minimum cat3 wiring.
tmc8080

join:2004-04-24
Brooklyn, NY
Reviews:
·ooma
·Optimum Online
·Verizon FiOS

missing the point

the shmucks who develop this stuff up don't seem to ge the fact that the real world deployment has to go through antiquated copper networks. pristine copper can do all sort of acrobatics with speed.. why the heck do you think Ethernet has been around for so long and not exactly replaced by fiber just yet? The problem is sending the singals LONG DISTANCES... and for that, it gets MORE expesive per mile than FIBER builds, in this day and age where labor costs are about the same. Fiber has the futureproofing that no other technology does in the real world. This is the proven case in the underground transoceanic deployments of fiber which will last for decades if not centuries (barring accidents, seismic damage, and other acts of man and nature). Try trusting DSL to a transoceanic deployment where the costs are 10000 fold more expensive... you know what the response would be... same for coax.