Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[WWWombat]

VM have made room in the spectrum for more channels than they are using right now. In theory they could go to 500Mb+ in some areas relatively quickly, potentially within days.

But would it be hard to market until the "some areas" turns into almost "all areas"?

They have nothing to fear from G.fast in terms of headline data rates. By the time G.fast is in its 3rd incarnation they'll be happily ticking over on DOCSIS 3.1 with the 1.2-1.7GHz networks to match.

Very true. If Openreach really feared VM, they'd have FTTP in much more of the pipeline.

Kitz - I mentioned Virgin, however VM don't put Openreach off, quite the opposite. Openreach get lower take up, yes, however when they get a customer back from cable due to FTTC they are making both the FTTC charge and line rental where they weren't before so there's considerably more incremental revenue there than where they are a monopoly.

This difference is one reason why I think, in terms of locations, why BT might choose to cover Lightning areas.

Openreach have no political axe to grind as far as getting ultrafast to match Virgin's coverage goes. Even if VM were to stop Project Lightning now Openreach still wouldn't have as many homes passed with, as the definition now seems, >=300Mb.

It is funny how the government continues to use 100Mb, while the supposedly non-political Ofcom have chosen 300Mb for what appears to be entirely political reasons.

My own area should be interesting. Very high FTTC take up, Project Lightning just rolled in, going live soon, and, at least in the rest of the area, showing very high take up. On an exchange level Lightning has just enabled nearly half of all premises passed by the exchange, none of which have especially good ADSL speeds, some of which didn't have FTTC, Openreach will have seen their customer numbers take a hit.

That's why I think that, if they choose to go deeper anywhere, it will be in Lightning areas.

That there is someone on the ground to observe progress is useful!

If VM uptake in my local area specifically is very high Openreach may be left with an empty or almost empty DSLAM with little prospect of getting many customers back on it. All properties are also fully ducted so there could even be a question mark over whether the maths would 'work' with regards to FTTP, potentially even using the existing 2nd DSLAM. OLT cards, 10Gb backhaul, done.

It would allow options, for sure.

[WWWombat]

I'm just thinking from a different angle here, but I wondered...

What's the feel for how a G.Fast node will be backhauled?
- Its own GbE fibre, pretty much like an FTTC cabinet
- Its own 10GbE fibre - as above, but 10Gb optics
- Part of a GPON
- Part of a 10-GPON
- Backhauled into an FTTC cab, then sharing the backhaul from there?
- Backhauled into an FTTC cab, then overlaid on the backhaul from there with WDM?
- Something else?

[S.Stephenson]

Why not just feed both the FTTC cab and the G.Fast cab off a single 10GbE link surely the could just swap the equipment at both ends and not even have to run new Fiber.

Then when they run nodes swap to a single 40/100GbE link.

[Chrysalis]

ignition the big question is for your area, is 40mbit speeds not good enough for the customers on the openreach cabinet?

The openreach cabinet filled up quickly because it served properties with poor adsl lines (As evidenced by your unstable 2mbit adsl), however it is now providing speeds in the 10s of mbits per second, which unless you are doing lots of heavy downloads, is enough for most use today.  So the question here is how many people on your cabinet are like yourself who just want the highest headline speed (and dont care about anything else) or will only change if there is something they cannot do on their existing connection?

Part of BTs decision to ignore VM will be based on the fact only 10% of VM customers are on the top tier, and the vast majority of BT customers are on the lower tier FTTC product.

[WWWombat]

Why not just feed both the FTTC cab and the G.Fast cab off a single 10GbE link surely the could just swap the equipment at both ends and not even have to run new Fiber.

Then when they run nodes swap to a single 40/100GbE link.

That assumes the two nodes are in the same location,and probably interlinked on the backplane - which suggests an alternative is to fit G.Fast boards to an existing DSLAM. Deeper G.Fast nodes will need a different solution.

In addition, it would also not work with the hardware as described in the pilot - because that uses a separate L2S/OLT to terminate the connections at the head-end. I have no idea why they want this restriction, nor any idea whether it will stay in place once livel.

[niemand]

I'm just thinking from a different angle here, but I wondered...

What's the feel for how a G.Fast node will be backhauled?
- Its own GbE fibre, pretty much like an FTTC cabinet

- Backhauled into an FTTC cab, then overlaid on the backhaul from there with WDM?

Edited to reflect the two methods as of right now, depending on availability of fibre. There are some FTTC cabinets that are 'parented' from others via WDM already.

[PhilipD]

Hi

If G.fast isn't bringing fibre any closer to premises it seems a waste of effort and money.  G.fast mainly improves on speed by using higher frequencies, but that also brings issues with interference to FM radio stations, so these higher frequencies will have chunks taken out of them to mitigate interference limiting the potential increases in speed.  They could just turn up VDSL2 now to use 30MHz and boost speeds, although as vectoring has been a failure and the higher the frequencies get, the more cross talk on decades old telephone cable, speeds aren't going to be that much improved unless you are just metres from the cab.

G.fast was going to bring higher speeds to more people, by bringing the nodes closer to peoples homes in addition to using higher frequencies, but that is not seemingly going to happen now. People already on 80/20 are going to be less inclined to see the need to upgrade, but these are the people that G.fast will only significantly provide faster speeds to. Those on slower speeds due to distance from the cab, will still be on the slowest speeds on G.fast, might be faster than they are now if they are lucky, but still they are suffering the effects of distance.

So we are now to see yet more street furniture, and the benefits of faster headline speeds falling in an even tighter radius to fewer people than VDSL.

I wonder if anyone has ever done a cost breakdown to see how all these hacks in order to push data through cables only ever designed for voice, compares to what it would have cost just to have gone full FTTP at the start of "broadband" as we know it. 

A small fortune must have been spent upgrading exchanges to ADSL1, then upgrading them again to ADSL2+, then installing VDSL cabinets, and now G.fast cabinets, all of which are running concurrently now needing electricity and maintenance.  It has to be false economy, of course I suspect it is us the consumer paying for it.

Regards

Phil

[gt94sss2]

G.fast was going to bring higher speeds to more people, by bringing the nodes closer to peoples homes in addition to using higher frequencies, but that is not seemingly going to happen now. People already on 80/20 are going to be less inclined to see the need to upgrade, but these are the people that G.fast will only significantly provide faster speeds to. Those on slower speeds due to distance from the cab, will still be on the slowest speeds on G.fast, might be faster than they are now if they are lucky, but still they are suffering the effects of distance.

G.fast was originally supposed to operate from the DP but BT and AT&T have secured changes in the standard so it works at much further distances than previously. Deeper deployment will come but it was always going to take longer than enabling it near the existing FTTC cabinets.

I wonder if anyone has ever done a cost breakdown to see how all these hacks in order to push data through cables only ever designed for voice, compares to what it would have cost just to have gone full FTTP at the start of "broadband" as we know it. 

A small fortune must have been spent upgrading exchanges to ADSL1, then upgrading them again to ADSL2+, then installing VDSL cabinets, and now G.fast cabinets, all of which are running concurrently now needing electricity and maintenance.  It has to be false economy, of course I suspect it is us the consumer paying for it.

You are wrong - installing FTTP immediately would have been much more expensive than installing/upgrading ADSL/FTTC/g.fast and so on. It would also have taken much longer to cover the same number of customers + at least some of the cost of upgrading to intermediate technologies makes eventual FTTP cheaper.

[Chrysalis]

its another adsl2+ situation really, a upgrade for those who have it now but not to those who dont have it.  Due to ofcom effectively letting BT off the hook, it is what it is sadly, we just have to wait some years for the fibre to get another push nearer homes, those who are desperate for more then they have on vdsl (which seems more then I realised), will have to cheer on VM and hope they start taking chunks out of BT's userbase, then the accountants will have to rethink if that happens.

[PhilipD]

Hi

You are wrong - installing FTTP immediately would have been much more expensive than installing/upgrading ADSL/FTTC/g.fast and so on. It would also have taken much longer to cover the same number of customers + at least some of the cost of upgrading to intermediate technologies makes eventual FTTP cheaper.

I don't see how it could be more expensive.  We are going to end up with fibre to our properties, or at least a copper cable designed for data at some point.  Virgin media was able to install a complete new network in my city in relatively little time with co-ax to the door (that could have just as easily been fibre, but we don't need fibre, some decent copper cable designed for the job is just as good), not sure of the costs for Virgins deployment, but it would be higher than BT as BT already have ducting and the infrastructure in place, but Virgin managed it.  Granted Virgin have cherry-picked their areas for the best return and don't cover large parts of the country, but that's no different to BT cherry picking where they deploy to, then getting government help for unprofitable areas.

I can't believe the total costs will end up working out cheaper when a real physical data connection arrives at our front doors from BT, if it ever does, as opposed to hacks that ADSL/VDSL and G.fast are, when to achieve they use several intermediate technologies which all have to be designed, procured, installed, powered and maintained.  I wonder what the total electricity bill is for VDSL a year?  A huge amount of power is wasted with VDSL (and ADSL) as you have to use a lot more power just so some signal remains after most of it drifts off into space, or worse, interferes with someone else's connection, fibre or real data cable needs much less power.

At least with VDSL, the money invested has returns in later years even if VDSL becomes obsolete, because it's laid the way to bringing fibre much closer to us, so it isn't money down the drain.  With G.fast just being another unsightly bit of street furniture to get graffiti on next to the existing boxes, with it doing nothing to extend the fibre network, investment in that technology is money down the drain, to only benefit those who don't really need it, and has a very short lifetime.

These hacks just delay, and make more expensive, the long overdue upgrade to POTs.

Regards

Phil