Verizon DSL → G.fast and XG.fast

Anyone been watching the development of this?

»www.zdnet.com/g-fast-1-g ··· 0031575/

I know Verizon has pretty much killed off FIOS expansion, but I wonder if G.fast might give them an incentive.

Sounds like it would require FTTdp (fiber to the distribution point) which is better than FTTH (fiber to the home) from their standpoint. But worse than FTTN (fiber to the node) which is what is done for from DSL variants like VDSL2.

I know my area of Verizon is ADSL2+ and I manage to get 7Mbs/768k out of the service, so I don't see this taking shape before 2015 or 2016 if at all, but I wonder if it would be a game changer for them to financially upgrade some copper, or if they'll forever stay on their LTE bandwagon.

I'm surprised I haven't seen any mention of this on the forums. I know there are some providers currently doing trials, but most seem to be on the other side of the ocean.

I wonder if NYC will ever see this. Sure would like to get some more bandwidth. Wonder how much they would jack up the monthly billing.

Unfortunately, under the current VZ administration (McAdam/Shammo) the focus is on making $31 billion of revenue in a quarter than upgrading their aging infrastructure and reducing those juicy profits a little to ensure the long term future.

The construction budget has been shrinking, not growing, along with the maintenance budget.

VZ still has many Central Office's with no broadband of any sort, but with the whole "Lets replace wireline with LTE" thing I don't see much hope for those stuck with nothing now. The only expansion I have witnessed (locally in VA) is when a new subdivision is built in a no-broadband copper area, they bring out FIOS, but just for that subdivision and they ignore all the addresses passed to reach that location. A big shame really.

PA BFRR regulation seems to be working though for those in PA with no DSL, VZ must hate being forced to put down RT's and provide DSL to their neglected rural customer base.

Its clear that some more life can be squeezed out of the copper plant with VDSL2, G.fast and XG.fast. I can see G.fast and XG.fast working out well in subdivisions with young copper from the 90s or newer, where the houses are close together and drop lengths from the ped's are short. Going by that graph, the new technologies will offer great speeds out to 250m or 820ft.

I hope Verizon will at least do some trial testing (unlikely with CEO MCwireless) Verizon must understand the average family can not afford LTE as their primary internet. I don't understand why Big Red won't invest into a new technology that could possibly improve their current copper network. I understand that fiber cost less to maintain ETC...

I hope that they at least trial the technology and see if it can be financially worthwhile to keep a fixed landline based service alive by breathing new life into it.

LTE and wireless is great, and the bandwidth caps have created a cash cow for them, but they seem to be completely ignoring the need for home-based broadband for traditional PC's and other home applications. They focus mainly on mobile devices and pretend that outside their fiber-deployed areas there's no market for home-based internet applications.

Sure, there are other providers to pick up the slack, and they seem to push people towards them purposely for lack of a competing product, but if new technologies like the G.fast development can satisfy both the stockholders with the lower build-out cost and still provide a quality service at a reasonable price that will fill their pockets with more money, then what is the issue?

Jim

The issue is keeping the highly regulated Copper Plant alive. They want it to die - CLEC's are a pain in VZ's side as CLEC's get the priority for anything.

See this:

»www.infotelsystems.com/c ··· ues.html

I would love to see this technology trialed - a townhome subdivision seems ideal for XG.Fast due to the short distances between homes and the plant.

Its weird how Investors balk at spending on telecom infrastructure when they invested in a telecom company. The infrastructure is what lets said company make money. I don't see Comcast investors whining about DOCSIS3.0, node splits or IPv6 deployment.

Perhaps the wrong crowd picked up VZ stock...

The fact that it only works 800 feet and less means there's limited types of places it would work. Even in a city, 800 feet could be less than a whole block.

This seemed like a way to limit the copper plant to me. For example, deploying fiber to the neighborhood distribution box at the end of the street, and then copper is maintained to the individual homes. The 90's brought on a lot of quality copper lines, sometimes multiple pairs to homes for analog modem use. The aging copper infrastructure could be replaced without the major expensive of digging up yards and running fiber directly to the home.

Agreed 800 ft. isn't much, but looking at my area, every two blocks has a box at the corner of each alley that serves about four city block halves. I'm in about the middle of the street with one of those boxes, so I'd likely be in range. I bet you could have a large percentage of homes in some areas where 800 ft. would be enough.

»www.cedmagazine.com/news ··· st-trail

There's also work being done to allow for VDSL2 and G.fast compatibility where VDSL2 can extend a few thousand feet at still decent speeds.

In FIOS areas, Verizon already uses VDSL2 in some complexes like apartment buildings when running fiber to individual units isn't easily accomplished.

As far as compatibility with current ADSL2 technology that's coming from the CO or remote terminals, I think they'd be throwing much of that away.

Yep, there is a lot of infrastructure from the Bell Atlantic days allover. Fiber, remote terminals. Things were being set up for a FTTN future I think, until the switching of gears to FIOS.

Regarding the copper, your right that the infrastructure is overbuilt thanks to that modem and fax boom. Perfect for pair bonding VDSL2.

Here is an interesting graph from the New Networks Institute/TeleTruth showing Verizon's construction budget over the past:

Interesting change over the last couple of decades, at one time we had 3-POTS lines one of which was reserved for dialup and later a fourth circuit for early SDSL. We are now back to a single circuit for combined POTS and ADSL.

VDSL2 is an effective way to utilize existing copper but it needs remote terminals to keep loop length under a few thousand feet. Vectoring, while complex, has shown great promise to increase typical VDSL2 speed. Unfortunately all VDSL2 circuits in a given cable need to be included in vectoring, pretty much eliminating VDSL2 as a business opportunity for Competitive Local Exchange Carriers (CLEC).

The higher speed telephone copper loop technology is really optimized for multiple dwelling units (MDU), not dense suburban environments.

In the early days of residential broadband I was very bullish about DSL compared to Cable. But the reality is that the Cable companies have blown the socks off the Telcos, and the Telcos have shown little incentive in upping the anti by deploying FTTP.

/tom

This got posted in the CenturyLink forum too.

It's a really interesting idea, but in essence, you have to put these things so close to a home that you're better off just running the fiber all the way to the house. It would avoid the situation of either needing to power all of these things by beefing up span power insanely, or by installing power meters on every single pole or at every single pedestal in the neighborhood, because essentially what this would be doing is placing extremely small DSLAMs in-line, sort of like the old FITL deployment. (And, we all know how that went.)

Though it's neat to see "DSL does gigabit if your DSLAM and modem are in the same room!" I just can't imagine this being practical outside of industrial situations already using LRE, SDSL/SHDSL or VDSL2 for whatever reason, such as hospitality, outfitting old offices or MDUs with faster Ethernet service, etc.

Although this is neat and I'm sure somebody will buy it for some industrial purpose, I believe the true way forward for the telcos is to get fiber into an ONT on the side of your home or on your desk. VDSL2 would make a lot of sense if fiber itself were somehow impossible to install and we could reasonably rip out all the old infrastructure and design the physical plant with DSL service in mind, instead of just sort of inserting it after the fact. Smaller VDSL2 service areas nodes using fiber-fed DSLAMs like the Adtran TA1100 series (Lucent has an equivalent) would be easier to maintain, faster (and closer to symmetrical) speeds would be available for more people, etc.

On the other hand, and this is where the real rub lies -- even if your loop is 17 miles, the telco can lay fiber along it, put an ONT on your house, eliminate or minimize electricity usage between those two points, and provide you with the same service as the guy whose loop is 200 feet.

In a perfect world, we'd all have a gigabit fiber connection to the home. This technology is something that I could see the telcos looking at and seeing if they could balance a better service with a low cost deployment. When they can install additional hardware into a box at the end of the street for maybe 20 homes in close range, send people a self-install kit in the mail and have them up and running on the cheap, it might be feasible.

Now there's no digging up ground and/or stringing cable and having professional installations at each home which is where the majority of the cost is involved.

Fiber to an ONT at the home is obviously the best choice, but when you're dealing with Verizon (and the like) mentality who are looking to please shareholders more than the customer base, then it may make more sense.

Laying the backbone for what would become FIOS began long before the GTE/Bell Atlantic Merger.

Yep and that's clear based on what infrastructure is out there, existing right now.

I heard VZ has an inane policy where anything for FIOS must be new construction, they cant reuse existing dark fiber. This seems silly to me.

Fiber to the building is clearly the best choice for future deployments of telecom infrastructure. Costs have come down in recent years, the services provided are not limited by distance nearly as much, etc.

GPON can handle I think a 30 mile distance from the CO without active equipment - in contrast to DSL's partly 2 mile distance +/- with hacks like loop extenders and DSL friendly load coils.

I was the coordinator for my area and I couldn't get the number of people they required to sign up. They gave me a number that seemed way too large and I don't know how they calculated it when taking into consideration several streets in were able to get DSL and a small development had cable. Even if they did bring it out to us, I'm sure it would be shitty 3M from an Adtran 12xx. Thank goodness I got the new cable company to run a line to me, I'm the last house on the street to get it.