Fibre infrastructure costs, FTTC vs FTTP vs FTTrN vs FTTdP

There is oft talk about FTTC being cheaper than FTTP to deploy, which is why Openreach and the BDUK programmes have rolled out very little of the latter; the sheer time and effort (by the time customers have been connected, if not in laying out the DPs) may be another reason, as there simply aren't the engineers available?

However with suggestions of ~£30k for an FTTC cabinet, on the face of it I do wonder why FTTP wouldn't be cheaper in areas that are relatively compact and have new/good ducting (eg/ new build developments). Are there any ideas of cost around?

And should developers not give Openreach a guarantee on the ducting they've installed, ie/ that they've not gone and blocked it, so Openreach should take advantage of that to roll our fibre to new-build sites whilst they have recourse to the developer for problems?

It confuses me why cabinets serving new-build sites still seem to be targeted with only FTTC, even when FTTPoD is (was) available in that area, so the capability to provide FTTP must be available at the exchange level. There are 4 outstanding in Basingstoke, all new(ish) build, which BT have indicated will be implemented under the commercial programme, but which are still outstanding (out of only ~20 commercials outstanding in Hampshire, 1600 delivered).

Openreach were engaged with us, but have now gone quiet, we assume as they're prioritising BDUK work (but won't say so). Do also wonder if they've decided to cease deploying FTTC where they're paying, with G.Fast potentially around the corner, and in places it should be more straightforward to implement?

Costs also vary between brownfield and greenfield, which is where developers come into the mix.

No sign of FTTC deployment ceasing in commercial footprint still seeing what look to be the odd cabinet come online, and even some signs of EO changes in London too.

Analysis Mason did the main stuff on C versus P costs, and recent like Gigaclear in Essex show the difference that can arise. Using B4RN as a guide is difficult since volunteer labour and free wayleaves makes a massive difference.

Remember FTTP includes all the fibre costs of FTTC but if serving say 40 homes, the extra metres of fibre splicing/chambers/manifolds to reach them. In many cases the cost of the fibre cabinet outweighs this by some margin and is faster to deploy too.

Some new build estates have seen FTTP appear, so it does happen but conditions need to be right or its just luck.

FTTrN and FTTdP is difficult to judge but will provide different price points. The dP trials are looking at using pluggable fibre so probably more like gigaclear where the final drop to each home is easier and cheaper to do hence the talk of a new FoD2 product. Using connectors with fibre introduces more attenuation but for residential size runs if probably not a problem.

The big elephant in the room is things like cost of power (FTTP does not need it), if a duct needs to be laid crossing a road and cost of returning road surface to standard.

Have seen some FTTP installs take multiple days to get up and running, due to building issues and finding suitable locations for the fibre ONT that needs power.

Size of engineer workforce is an issue.

On the BDUK work, I doubt there are many infrastructure projects that draw so many people into go over the data and reach many varied conclusions.

From a BT board viewpoint my expectation is that they've run the figures and decided that FTTC now is quick to deploy and gets take-up quick, and that with lots of fibre re-use possible a mix of dP and FTTP over the next decade will get us to an almost 100% fibre scenario around 2025 for roughly the same cost as starting national FTTP roll-out now, but without the slow roll-out scenario.

I do wonder why BT are going with G.FAST. I was expecting they would just extend FTTPoD to all exchanges and let demand drive the fibre. But they clearly had a lot of problems with FTTPoD and have decided that it just isn't a viable solution. Like you I wonder why they had so many problems with it.

But there is one note on duct availability. Not all properties have ducting all the way to the front door as 'twere. I know mine doesn't. So to roll-out fibre on my estate BT would have to micro-trench everyone's garden or driveway. I suspect that's the biggest stumbling block.

But I don't really know. It'd be good to hear what other people think. Is G.FAST really the best practical next step? And why?

g.fast avoids the need to cross roads for every property for example.

I think some have got confused by the first stage g.fast roll-out likely to be just cabinet based rather than actually on the dp for properties.

Basically BT long term plan is a stepped approach, pushing fibre closer and closer, and as a firm with an existing network in place this makes sense.

For the new entrants going for FTTH from the start makes more sense, it will be interesting to see how CityFibre and Sky's own FTTH stuff plays out. If Sky and TalkTalk get good signup rates that affect openreach then BT board may change things up a gear.

In reply to a post by olicuk:

And should developers not give Openreach a guarantee on the ducting they've installed, ie/ that they've not gone and blocked it, so Openreach should take advantage of that to roll our fibre to new-build sites whilst they have recourse to the developer for problems?

The Planning Departments of all Local Councils should mandate such access is made possible in any new housing and business park planning applications.

Actually just needs no legislation at all, just people to do some research before committing to a 25 financial year mortgage.

If people stop buying new builds that do not have decent broadband available already then developers will quickly change their ways.

The business scenario is different, since leased line and Ethernet available almost everywhere, just for a small 3 person firm that is not a video edit studio consumer type services provide enough speed, just need to stock up on the SLA options.

Yes, a lot of people thing that FTTC somehow debars moving to FTTP later (or maybe g.fast of course). The point about FTTC is that it's relatively cheap and fast to roll-out as it uses a lot less manpower. The cheap bit could be dealt with by throwing money at the project (if it was available), but I cannot see the manpower situation easing. The idea that we could somehow train and employ a new army of field staff to get anything approaching the speed of roll-out of FTTC is a non-starter.

As far as current costs go, it would be nice to know what the breakdown is between the actual costs of installing FTTC cabinets and that spend on getting the "E" side fibred up. In principle, the latter is a springboard for whatever comes next.

Personally I have some doubts about putting g.fast nodes very deep into the "D" side network due to the power issues as much as anything else. (And all those line-powered approaches, whether forward or reverse are unlikely to be light on resources). That said, I'm acutely aware that if there are a large number of properties like mine with direct-buried phone cable, the costs of running fibre to the home will be very high.

So I think the country will look like a patchwork quilt when it comes to broadband provision. There simply won't be some grand scheme which will unify technical standards in the manner of the phone network when it was brought together by the GPO in the early 20th century.

What, of course, we will have is increasingly insistent pressure groups to deal with those left behind. I don't think they will be pacified with satellite solutions.

Thanks for your insight Andrew in the other responses, and encouraging to hear there is still some commercial deployment ongoing elsewhere; all of the roadworks.org documented work in Hampshire at the moment is BDUK related.

On the subject of developers, I don't think consumer power is the answer; it's a nice ideal, but for most people there will be significant financial, school, environment, or other considerations that will outweigh the broadband status. A couple of years ago when I bought, I investigated, the council said they'd had agreement with Openreach that we'd be getting FTTP. That coupled with the proximity of two FTTC cabinets (~30m and ~100m), and the expected widespread deployment of 4G (which we have now from most networks, but it's not stable enough, nor is pricing/bundling good enough, to make it better than 3-4Mbit ADSL) ticked the 'ok' box, and for new build there weren't really any better options. It was a good time to buy, and because of the mortgage arrangements and offers from developers, new build was the only option for us. As much as my wife and I could have committed to commute another 30 mins each way a day to find a fibre-enabled development, or bought a 1-2 bed older rather than 4 bed new build property, understandably fibre lost!

Notwithstanding that, I don't understand why housebuilders don't commit to offering superfast on all of their developments, as I'm sure they could do so with none or very little additional outlay (far exceeded by 'discounts' etc they offer). But with the government backing housebuilding schemes to the benefit of developers, I do find it hard to justify why they don't mandate superfast provision to order for a development to qualify for these schemes - problem sorted - and to save them having to pick up the pieces later.

Even now a new Persimmon/David Wilson development for 1000+ properties has started nearby, on former council land, and still seems to lack any superfast provision... but houses there seem to be selling like hot cakes.

Given that we have a rapidly increasing population (up by over half a million last year alone) and have a housing crisis on our hands, it's a seller's market for the builders.

However, there's no excuse for planning guidelines for larger developments not to include proper broadband provision. The incremental costs are trivial compare to the price of a house when installed at build time. I don't know if local authorities have the power to insist on this, but they can certainly require developers to contribute to the costs of enhanced infrastructure in the form of community centres, schools, roads and much else.

To be blunt, if the UK Gov threw £30bn at the problem today, actual money in the bank account it will still take 5 to 10 years to do FTTP to everywhere.

Double the money and you could afford to hire lots more new staff and tempt over installers from Europe to work on a project knowing its only good money for 3 or 4 years.

Not sure firms like Virgin Media, City Fibre would allow this to happen either, the FTTC option from BT meant that there was enough of a gap to what they can do/plan to not panic them in their chosen markets. A system like the connected city vouchers would not work that well for FTTP building in rural areas, as you need a certain number to make it worthwhile.

BT could do a lot better though, the problems where they've left partially built FTTP is recurring theme to mail bag, but BDUK contracts take precedence due to penalty clauses.

The development that I'm buying on is getting native FTTP, but you can't find any mention of the fact anywhere. You'd think the builders would use it as a selling point, but if I hadn't asked about it I don't know if it even would've mentioned to me until I got to the point of moving in!

At £30k for a cabinet that serves 300 homes, that's only £100 per home. It is the sharing of costs that helps.

If you had a £100 budget, that might buy you 1m of duct in the road, 2m of duct in the path, or 3m of duct in the verge. Digging stuff into the ground is expensive ... and as you get fibre closer to the cluster of premises, there are fewer premises to share it all around.

There's an interesting graph of the total cost of a rollout over at the bottom of this article:
http://www2.alcatel-lucent.com/techzine/the-numbers-...

Otherwise, the most detail comparison of FTTC vs FTTP for the UK comes from:
http://www.analysysmason.com/PageFiles/5766/Analysys...

A tech think-tank did a study into the costs & benefits of deploying FTTP, and came to the conclusion:

In reference to www.nesta.org.uk/publications/exploring-costs-and-benefits-fibre-ho...:

The discounted cost of FTTH deployment in 2016

is the same as

The discounted cost of deploying G.fast in 2016 and overbuilding with FTTH in 2023.

In other words, if speeds beyond the 700 Mbps G.fast offers are not required until after 2023, then it will be better to first deploy G.fast, even if a later upgrade to FTTH is necessary.

You might also like this one:
http://www2.alcatel-lucent.com/techzine/5-myths-abou...

For me, the fact that FTTC can be deployed 4 or 5 times quicker than FTTP, and yet provides a jump-off to FTTP anyway, becomes the most compelling feature.

With just FTTP, most of us would still have nothing!

Openreach were engaged with us, but have now gone quiet, we assume as they're prioritising BDUK work (but won't say so).

I suspect this fits somewhat. They're probably at the squeaky-bum point of the project, trying to meet the deadlines for most projects; things might return to normal in 7-8 months time.

In reply to a post by libf:

The development that I'm buying on is getting native FTTP, but you can't find any mention of the fact anywhere. You'd think the builders would use it as a selling point, but if I hadn't asked about it I don't know if it even would've mentioned to me until I got to the point of moving in!

Yes, the fact that this sort of thing isn't publicised more by people such as developers is ridiculous. We're halfway through 2015, this isn't 1998 any more - a good internet connection is fast approaching "essential" for most people. That might seem like hyperbole, but I don't think it is, especially with BT themselves saying they expect everyone to be completely IP-connected with even POTS by 2025. I don't think I believe that, i.e. it's just an extremely optimistic argument they're using as leverage to escape from their basic requirements [i.e. the POTS requirements now enforced by OFCOM] which have been standard since something like the 1940s, but still, things are moving quite fast now. At last.

Edited by deleted (Sat 27-Jun-15 06:31:42)

In reply to a post by gazzyk1ns:

BT themselves saying they expect everyone to be completely IP-connected with even POTS by 2025. I don't think I believe that, i.e. it's just an extremely optimistic argument they're using as leverage to escape from their basic requirements

I'm not so sure.

From one side, I think BT realise they now need to make sure their business is built on data, not on voice ... and that needs a network to match, and regulation to match. This is definitely the carrot.

The stick perhaps comes in the forms of their existing voice network - the System X and AXE-10 exchanges installed in the 80's and 90's. When we come into the '20's, those exchanges will be getting elderly, and the cost to maintain will be rising, and parts will be harder to find. There's going to become a pressing need to swap all this equipment out.

They halted the rollout of the voice component of 21CN a long time ago, and I suspect they won't go forward with plans looking like that (essentially VoIP from the exchange).

With the direction things are going on the data side, where D-side copper remains in place, and the final parts of the copper (incl the drop wire) are going to remain a critical part. The E-side, however, can be largely ignored (at least in 10 years, or so).

I think we're heading in a direction where voice services will be VoIP from the cabinet, or VoIP from the home; the latter is probably the proper end-goal, with analogue <-> VoIP conversion performed in a home gateway.

[i.e. the POTS requirements now enforced by OFCOM]

I couldn't find the time when a universal service was actually obligated on either BT or the GPO, or exactly what was being obligated. In the mid-nineties, Oftel said this:

In reference to www.ofcom.org.uk/static/archive/oftel/publications/about_oftel/cons...:

Universal service is the concept that anyone asking for a basic telephone service can have it at the same price wherever they live. We define it as: Affordable access to basic telecommunications services for anyone reasonably requesting it regardless of where they live.

And the definition of 'basic' service is: Individual access to the telecommunications network via exchanges capable of providing voice telephony, with free services of itemised billing and selective call barring, and some supplementary services available (such as call diversion and call waiting).

Going back to basics, there is little doubt that a VoIP gateway in the home, an all-IP access network (deployed using a mix of methods), and softswitch IP exchanges would fit the bill.

which have been standard since something like the 1940s,

I'm really not sure the idea of a "universal" service has been around for quite that long.

I can find the following figures on households with phones in the UK:

- 1964 = 2%
- 1971 = 38%
- 1972 = 42%
- 1981 = 75%
- 1988 = 85%
- 1993 = 90%
- 2000 = 95%

The sixties and seventies were obviously a great boom time in telephone access (supported by mobile exchanges to help with demand) ... but I'm not sure anyone really thought in terms of a universal service, at least as an obligation.

In the reading I just did to glean these figures, the most striking thing about the studies on universal access was that none of it was about access problems in rural areas. It was all about access to the low-income.

http://www.ofcom.org.uk/static/archive/oftel/publica...
http://www.ofcom.org.uk/static/archive/oftel/publica...

The earlier figures obviously refer to just fixed-line phones. Once beyond the mid-nineties, however, the availability of mobile play quite a part. One of those 1999 Oftel links surmised that 3 in 5 of those households without a fixed-line phone did have a mobile.

<sub>Edit: It just struck me: Those figures for homes with a telephone ... they show a nationwide rollout not unlike the one we're now getting for FTTC.

Back then, they progressed at a rate of about 4% of households per year through a 20 year peak, which looks spookily like the rate that an FTTP deployment might progress at today.

Our (mostly) FTTC rollout has been running at around 4% per quarter instead.

Edit 2: Those figures also mean that, on average, copper has only been in the ground for about 40 years</sub>

Edited by deleted (Sat 27-Jun-15 13:20:29)

I finally found the last part that I had wanted to add previously ... some analysis of deploying G.fast and FTTdp nodes.

Last year, there were some presentations floating around about how different deployment choice could be made in Amsterdam. One of these presentations is here:
http://www.joepeesoft.com/Public/DSL_Corner/_Index.html

Look for the document titled "Hybrid FTTH deployment in dense city areas (Amsterdam case study for G.fast)"

There is also something closer to a research paper, giving the theories and maths behind those deployments:
http://publications.tno.nl/publication/102774/Y67gP3...

The guy who wrote that document finished off his thesis on the topic in general, which is *way* more theoretical:
http://publications.tno.nl/publication/34609914/IW6u...

His summary includes this quote:

Analyses led to the conclusion that a migration of Full Copper via FttCab (VDSL-based ) and FttCurb (G.Fast-based) to FTTH is both cost effective and retains the customers.

In reply to a post by WWWombat:

I'm not so sure.

From one side, I think BT realise they now need to make sure their business is built on data, not on voice ... and that needs a network to match, and regulation to match. This is definitely the carrot.

The stick perhaps comes in the forms of their existing voice network - the System X and AXE-10 exchanges installed in the 80's and 90's. When we come into the '20's, those exchanges will be getting elderly, and the cost to maintain will be rising, and parts will be harder to find. There's going to become a pressing need to swap all this equipment out.

They halted the rollout of the voice component of 21CN a long time ago, and I suspect they won't go forward with plans looking like that (essentially VoIP from the exchange).

With the direction things are going on the data side, where D-side copper remains in place, and the final parts of the copper (incl the drop wire) are going to remain a critical part. The E-side, however, can be largely ignored (at least in 10 years, or so).

I think we're heading in a direction where voice services will be VoIP from the cabinet, or VoIP from the home; the latter is probably the proper end-goal, with analogue <-> VoIP conversion performed in a home gateway.

I certiainly don't disagree with any of that - i.e. I don't disagree that the country/nation needs it, and I think that BT genuinely wants it to happen, too. I'm just not sure they can do it by 2025.

...[i.e. the POTS requirements now enforced by OFCOM]...

I couldn't find the time when a universal service was actually obligated on either BT or the GPO, or exactly what was being obligated. In the mid-nineties, Oftel said this:

In reference to www.ofcom.org.uk/static/archive/oftel/publications/about_oftel/cons...:

Universal service is the concept that anyone asking for a basic telephone service can have it at the same price wherever they live. We define it as: Affordable access to basic telecommunications services for anyone reasonably requesting it regardless of where they live.

And the definition of 'basic' service is: Individual access to the telecommunications network via exchanges capable of providing voice telephony, with free services of itemised billing and selective call barring, and some supplementary services available (such as call diversion and call waiting).

Going back to basics, there is little doubt that a VoIP gateway in the home, an all-IP access network (deployed using a mix of methods), and softswitch IP exchanges would fit the bill.

which have been standard since something like the 1940s,

I'm really not sure the idea of a "universal" service has been around for quite that long.

I can find the following figures on households with phones in the UK:

- 1964 = 2%
- 1971 = 38%
- 1972 = 42%
- 1981 = 75%
- 1988 = 85%
- 1993 = 90%
- 2000 = 95%

The sixties and seventies were obviously a great boom time in telephone access (supported by mobile exchanges to help with demand) ... but I'm not sure anyone really thought in terms of a universal service, at least as an obligation.

I think the agreement was to provide a POTS sevice to anyone who ordered it, although obviously I could be wrong. I remember 999 calls being a factor, although of course BT weren't responsible for new builds unless anyone asked (usually the developer I assume), or people who actively opted out. BT used to even provide the telephone in the house, and if you wanted a different one then it was tough.

OFCOM have only existed since fairly recently of course, but this article is what reminded me of it at the beginning of the weekend:

Provision of universal service in the UK

1.2 Up until 25 July 2003, anyone running a telecommunications system in the UK
is required under the Telecommunications Act 1984 to do so under a telecoms
licence. Universal service is ensured through conditions in the licences of
telecommunications companies.

So you're right, obviously, the 40s was a bad guess by me. But still; since 1984 is a relatively long time.

Edited by deleted (Sun 28-Jun-15 19:38:45)

for an developer to gain FTTP you have to sign an FTTP contract -- this some misinformation -- that development is serviced by cab Basingstoke PCP 102

Tell tale sign if a green box at entrance to your new development no FTTP conversation and no fibre conversatkion only a voice conversation

interesting Broadband is now one of the deciding factors in a house purchase how -- this has changed in the last 6 minths or so -- much more focus on new homes and broadband -(or in most new homes lack of it)

Hi,

Not sure I follow.
There are new developments in Bedford (although I previously lived in Basingstoke!) who have PCP's at the entrance to the development and have FTTP.

Fibre only developments via openreach are rare.

Thank you for the in-depth research and links WWWombat, I'll have a good read. From the couple of your links I have already looked at, it's interesting to see the third party research effectively aligning with BT's approach.... though perhaps not surprising for a corporation responsible to its shareholders.

PS. Fastman2, we knew the development wasn't enabled for FTTP by the developer, so obviously would have a standard cabinet, but the local council officer responsible for promoting superfast broadband in Basingstoke and liaising with Openreach (this was pre the BDUK programmes), provided the following response in June 2012:

... we have secured FTTC for PCP101, we don�t have which phase of Openreach�s roll out it will be included in so can�t say for sure when it will be installed. What will interest you is that at the same time we managed to secure FTTP for PCP102, and this will be the first fibre to the premise in the whole of Basingstoke, again we don�t know which roll out phase it will be included in.

Obviously Openreach have significantly scaled back their FTTP plans (for the time being) since then...

Interesting analysis!
Just a comment on the POTS requirement - the big factor here is that the mobile phone is fast becoming ubiquitous. If OFTEL/technology existed to allow "porting" of landline home/business numbers to the mobile network then the requirement for POTS would likely diminish fairly quickly.

Not everyone can yet receive a mobile signal (nor everyone fibre broadband) therefore some of the technologies utilising VoIP perhaps at a cabinet level may be required for the foreseeable future.
Again the handset could be mobile based and and potentially allow switching between landline and mobile networks where both networks are available. Businesses and call centres are already standardising on VoIP handsets so POTS with the traditional in-house PABX is also on the way out.

In summary, there are two types of network today - landline and mobile - both which can offer voice and data services to varying degrees. Fibre is the technology under discussion here but is not the only game in town (and country).

GC

If you get FTTP I will surprised as signs are that commercial roll-out is to deliver FTTC

Council of course used the wrong wording, words like secure make people think its certain when plans can change.

I did a rough back-of-the-envelope calculation of the idea of spending £60bn in four years as a sort of blitz building approach. That's about £15bn a year and if about 80% of that was spent on labour, and allow a rather generous £100k per worker per year, that implies a workforce of about 120,000. Of course that's not all network technicians. Much of it is basic street-work activity.

I wonder if there are that many people who could be enticed to do this, especially where skills are required? I also wonder if people would be prepared for the amount of disruption. Personally, I think if the government has £60bn to spare it would be better spent on easing the housing crisis.

Of course, this is all fantasy land stuff. There are all sorts of reasons why this would never happen.

nb. In looking at that £30bn then doubling it number, it's interesting to compare this with the Australian National Broadband Network (now largely de-scoped to fibre/copper hybrid). Last time I checked the budget for that, if adjusted for population differences, would be the equivalent of about £44bn in the UK).

In reply to a post by GC007:

Not everyone can yet receive a mobile signal

Not even close, I'd say. I live in a rural town that's only a minute from the M40 and has a population of over 12,000 and FTTC (I get over 65Mb/s). But my mobile can't see a reliable voice signal unless I go out into the garden. It's been known to drop off the network entirely and even fail to get text messages. It wasn't always that bad but when EE went official they seem to have merged some masts and left my part of the town with patchy coverage.

Edited by Andrue (Mon 29-Jun-15 13:54:39)

In reply to a post by TheEulerID:

nb. In looking at that £30bn then doubling it number, it's interesting to compare this with the Australian National Broadband Network (now largely de-scoped to fibre/copper hybrid). Last time I checked the budget for that, if adjusted for population differences, would be the equivalent of about £44bn in the UK).

..and they don't have as much history to dig around. Or over

Consider the actual physical efforts of getting fibre to every existing phone user.

Although I can readily point tothe PCP about 50 Metres away, across the road in a very obvious location to the north; and I also know specifically that my phone wire pair pass through a small footpath junction strip about 125 Metres to the east of my house, thus taking at least 250 Metres to cover 50 metres, I have been unable to positively define the route in between.

There is a definite but unproven possibility that the pair actually pass my front door (and the line entry point to the NTE) on its way to that junction strip; but that would require opening up u/g chambers, tracing cables etc.

So to replace all/most of around 60 million (I think) "D Sides", would cause a tremendous upheaval in cities, towns, villages, hamlets etc.

-------------------------

The PCPs are clear-cut locations, close to most end-points; and it is known that subject to a few exceptions such as EO lines, that access to most lines is achieved very readily at the PCPs.

===================

About 10 years after this estate was opened, with all the houses pre-wired for phone back to the PCP equivalent, there was an electricity failure at a phase joint "somewhere".

To locate its position, the Electricity Board had to call on the services of a retired worker, who came along, took a look at the sub-station, then walked about 40 paces south-westwards; and announced he thpought he was standing over the phase joint.

No chamber or anything else to show it.

Dug down; and there was the phase joint.

It still is not positively identified, by, say, having a chamber built around it.

Only the patched tarmac.

---------------------------

Apply that experience to the many millions of "D-Sides".

Edited by deleted (Mon 29-Jun-15 14:45:35)

In reply to a post by Andrue:

In reply to a post by GC007:

Not everyone can yet receive a mobile signal

Not even close, I'd say. I live in a rural town that's only a minute from the M40 and has a population of over 12,000 and FTTC (I get over 65Mb/s). But my mobile can't see a reliable voice signal unless I go out into the garden. It's been known to drop off the network entirely and even fail to get text messages. It wasn't always that bad but when EE went official they seem to have merged some masts and left my part of the town with patchy coverage.

It's definitely quite network dependent in some areas. I have no problems on O2's network at work in the Oxfordshire countryside (or so far in Brackley that I've noticed), but colleagues on other networks really struggle.

Will definitely check my signal next time I'm up in Brackley though as I wasn't planning on getting a landline when I move up there!

Hi Andrue,

I also live in a rural area and have battled with poor broadband and mobile signals.

If you have broadband at least 1Mbs up and down then an option, although not cheap, is to install a device offered by some of the networks which extends the 3G network into your house via the broadband. EE used to have one called Signal Box. Vodafone market one called Sure Signal.

Happy to take this discussion to another forum if more appropriate.

GC

Afternoon Andrew

I wonder of any of the posters have direct experience of the robustness and resilience of a single fibre "cable" on both underground and overhead situations, corresponding to likely "D side" locations.

How does "fibre" react to almost continuous wind whip, particularly at and near suspension points?

How does it react to simple striking by a spade in underground locations?


Copper is malleable and from over-a-century's world-wide experience can absorb quite a lot of such episodes; but will fibre have similar survivability, as it clearly lacks the malleability?


Anyone know the answers?


Also bend/corner radii, are they more critical from a fracture aspect?

Edited by deleted (Mon 29-Jun-15 14:46:38)

Fibre if struck will break simple, even spare pairs are likely to.

Hence the ducting and sub duct. FTTH people though say the maintenance is lower, and South East Asia has a LOT of overhead fibre in the cities linking small apartments

The fibre is bendy, but beyond a certain radius will break.

Wind whip as its held in a tube should be find and curve radius is maintained at the poles to avoid kinking.

Agreed with your analysis; and to add to that, assuming that 25% of that work-force would require specific fibre training - how long would that take; and what work would the trainers be diverted from, thus halting or certainly slowing down present progress?


When that FTTP installation programme has been carried out, what would that large, experienced, narrow skill-base work-force do, especially if it has received salaries of £100K, compared to the present average of around £25K?

Edited by deleted (Mon 29-Jun-15 14:56:14)

Very much doubt a fibre splicer can get paid £100k per year, maybe if they got paid overtime and worked every day of the week for 16 hour days maybe.

Fibre splicing needs more care than crimping two cables together, but not that difficult.

my undersranding is that PCP 102 is FTTC

Agreed - but that £100 K was not my estimate.

Agreed fully about the splicing of fibre etc.

A company gave me a private demonstration about 1990, including the polishing etc, when making up connectors with fibre tails, to make it easier to terminate long fbre runs, out on site.

"Things to Come"!

In reply to a post by GC007:

If you have broadband at least 1Mbs up and down then an option, although not cheap, is to install a device offered by some of the networks which extends the 3G network into your house via the broadband. EE used to have one called Signal Box. Vodafone market one called Sure Signal.

Yeah. Unfortunately I'm actually with Virgin Mobile (which piggybacks on EE's network). VM do have such a box but it's only available to Virgin Media customers. Vodaphone was my first ever provider and I'd never go back to them :-/

In reply to a post by libf:

Will definitely check my signal next time I'm up in Brackley though as I wasn't planning on getting a landline when I move up there!

I'm in the south west corner, just off the bottom of Pavillons Way. I think the problem is that we're shielded from the main town transmitter by the central hill and are reliant on a more distant transmitter. I have an app on my phone that uses cell strength to enable disable wifi. It reckons the two strongest transmitters are 650-15395 and 650-8850592. I've noticed that my Google device history shows my phone's location forever bouncing between home and Evenly.

I was being as generous as possible. Also, cost of employment is a lot more than basic salary with pension contributions, NI and so one. If I took a figured of £50K then we'd be talking 200,000+ people. Simply enormous (about 5 times what OR currently employ).

They do, however, have a much larger area to cover albeit that most Australian's live in urban areas. It should also be noted that the more remote areas will be served by satellite (albeit within that same budget).

The party overhead BT Openreach metal line to my house failed regularly. There was one year when it failed 4 times, once caused by the pair being allocated to someone else. I am hoping the underground fibre connection will be more reliable. Time will tell.

And they've launched their own bird birds?

Where did the £60bn figure come from? What is the total number of premises to be connected?

If you look further up the thread, you'll see Andrew was speculating on what might happen if the government did have £30bn for a full scale 100% FTTP rollout (a not unreasonable figure) and even then it could easily take 10 yrs or so due to manpower resources. Further, he speculated if you could double that figure you might be able to attract a large number of European workers for a 3 year blitz effort. I was just turning that into the size of workforce it might mean. Perhaps 150,000. Then allow 600 days each over 3 years and that's 90 million man days, or about 3 man-days per premises on average.

So it's just trying to work out. even if there was no budget constraint, what such a mammoth effort might look like (not that it ever will of course).

They are (or are going to). I seem to recall it's something like 3 satellites.

Evening Michael

I take it you are on FTTH or FTTP; and how long have you had that?

As you state that the fibre is underground, how was this achieved?

-------

To try to answer your other questions, I generally assume that there is almost 1 fixed, traditional copper line per head of population, although I suspect that what-ever that Total Figure is/should be, about 25% to 35% are likely to be to non-domestic premises.

I have done a brief search for such a Total; but nothing jumps out.

This page seems to offer a clue-

http://www.ispreview.co.uk/review/top10.php

Table headed-
Top 10 UK ISPs By Subscriber Size


However, this is likely to include ADSL as it comes to approximately 24 million.

Back in January, BT OR reckoned that only about 15% of the lines in the Fibre "Coverage" areas, had been upgraded, mainly to FTTC.

More recently a later figure of 19% appeared; and both certainly tie in with the progress I have noted on "my" FTTC-

March 2014 Released for Service with 48 Filter Links fitted, out of a total back-plane for 288.

April 2015 an additional 48 Filter Links fitted, taking it to 96 present Maximum, out of 288 potential.

So just before that additional set was added, it looks as though the highest figure attributable of 48 of 288, amounts to 16.6 %, sitting nicely between the 13% and 19% OR figures.


I have not got absolute Line figures for "my" PCP; but by checking against various aspects, it appears to have approaching 300 lines.

------------

For FTTP and FTTH, assume that advantage is taken of the FTTC Fibre, leaving only about 300 "D-Side" lines to have Fibre added, to reach individual houses, premises etc.

Bearing in mind Andrew's comments on the fragility of fibre, and that it is very unlikely that any of those homes and premises have ducting suitable for blowing fibre through, it does suggest a massive task to convert "the whole country", particularly if that includes getting "D-Side" fibre to every flat in blocks of flats and similar.

It may be that running a fibre line over existing poles for several kilometres to a distant farm and such-like, could be a lot cheaper than to a house in the middle of a city.

He is I recall on a Gigaclear FTTP service.
They direct bury in the ground their fibre cables.

Edited by deleted (Mon 29-Jun-15 22:02:53)

I have had a Gigaclear FTTP connection for 2 months. The cable terminates in a pot on the boundary of my property, and is likely in a duct with other Gigaclear cables about 40cm below the surface of the lane. It takes a slightly strange route due to the desire to minimize the number of joints. Gigaclear refused to use a nearby BT duct.

Well Talk Talk are selling FTTP for £21.7 p/m which is about half what Gigaclear or Hyperoptic charge. What do you think it costs them to provide a connection?

In reply to a post by Michael_Chare:

Well Talk Talk are selling FTTP for £21.7 p/m ...

Curious ... I just put my landline number (where FTTP has been available for just over 2 years) into the TalkTalk Fibre checker and got the result 'Fibre broadband isn't available in your area yet' although it did try to sell me TV, SIMs etc.

Isn't Talk Talk's FTTP offering part of the joint venture with Sky and City Fibre using their own infrastructure and fibre to the premises?

It is and CityFIbre are clear the residential is an incremental revenue opportunity, i.e. core business use from mobile, council and large business pay for most of the network, with basically what TalkTalk and Sky doing being the icing on the cake.

One very small corner of York, which seems to be just a Tesco store and maybe a road around it is like at present.

Also doing things like migrating people from existing network onto the NBN to the point that people have been warned to ensure they migrate before the forced move as they will lose services otherwise.

Also have I believe bought Telestra FTTC network.

It is the nationalised Telco dream really.

In reply to a post by TheEulerID:

I did a rough back-of-the-envelope calculation of the idea of spending £60bn in four years as a sort of blitz building approach. That's about £15bn a year and if about 80% of that was spent on labour, and allow a rather generous £100k per worker per year, that implies a workforce of about 120,000. Of course that's not all network technicians. Much of it is basic street-work activity.

I wonder if there are that many people who could be enticed to do this, especially where skills are required? I also wonder if people would be prepared for the amount of disruption. Personally, I think if the government has £60bn to spare it would be better spent on easing the housing crisis.

Of course, this is all fantasy land stuff. There are all sorts of reasons why this would never happen.

nb. In looking at that £30bn then doubling it number, it's interesting to compare this with the Australian National Broadband Network (now largely de-scoped to fibre/copper hybrid). Last time I checked the budget for that, if adjusted for population differences, would be the equivalent of about £44bn in the UK).

Something I've been 'banging on' about for almost ten years,

BT's figures in late 2007 suggested a figure of £27billion for a national FTTP rollout, using just BT's engineering resources.

As a Government funded project - leaving aside the inability of such projects to deliver on-time, within budget, and within scope - this could be completed within 24-36 months, and virtually end unemployment during the project time-frame; use the existing unemployed. There are a variety of skills required, from labouring through plant operation to splicing to project management.

Such a project delivers several knock-on effects; the reduction in social welfare spending from central govt (JSA, Working Tax Credit, Child TC) and local govt (Housing benefit, CTax Benefit), and increases in economic spending (from the increased retail spending natoinally, and the benefit to small business inherent wherever workmen are working - sandwich shops and newsagents).

What is missing is the political will to engage in such a massive national project, and the will to overcome the objections from the teleco's who have invested in existing infrstructure. To mitigate, the Govt could buy-and-reuse much of the existing plant.

Regards making fibre access a condition of housing regulations, I have to chuckle. When in 2001 the then Cheshire Council asked GX how we could improve comms in the area, we stated just that; define broadband as an essential utility the same as power, water, and sewerage, and thus all new builds required it for planning permission. For us, it was 10 times cheaper to install fibre during the construction phase of a development as opposed to coming back and laying fibre once complete.

With the subsequent postings, do they point to that Gigaclear FTTP connection being an unusually cheap method, but very effective for him - with very restricted application as things stand, for the majority?

In reply to a post by TheEulerID:

I did a rough back-of-the-envelope calculation of the idea of spending £60bn in four years as a sort of blitz building approach. That's about £15bn a year and if about 80% of that was spent on labour, and allow a rather generous £100k per worker per year, that implies a workforce of about 120,000. Of course that's not all network technicians. Much of it is basic street-work activity.

As you note later, it is a valid thought exercise, but not so likely to happen.

The other "£60-odd billion" project being considered - HS2 - can only really be done by spending at a rate of around £2-4bn per year. I suspect the same limitation exists there - the number of people with the right skills. With careful juggling of the same resources that will be in demand for Crossrail and Crossrail 2.

Of course, the starting point today is a little easier, with fibre spines in position. But it has taken 6 years to get that sorted. And that is probably 20-25% of the civil engineering effort.

I cannot imagine that an FTTP project could possibly be handled in less than 10 years or so.

In reply to a post by GC007:

Just a comment on the POTS requirement - the big factor here is that the mobile phone is fast becoming ubiquitous. If OFTEL/technology existed to allow "porting" of landline home/business numbers to the mobile network then the requirement for POTS would likely diminish fairly quickly.

In summary, there are two types of network today - landline and mobile - both which can offer voice and data services to varying degrees. Fibre is the technology under discussion here but is not the only game in town (and country).

Interesting, of course, to spot that the same places where BT push forward their vision of an all-IP network (with voice as an IP service), they also put forward the concept of a fully-converged all-IP network where land and mobile network (I guess) are invisible.

I guess that concept brings up the questions about not just number porting, but the ability to direct your "personal contact number" via a mobile network when you are out, a fixed network when at home (though delivered over the last few feet by WiFi, in-house LTE, or DECT), or Starbuck's WiFi if you're out.

The end result of a ubiquitous network is that you just receive your service wherever you happen to be.

In reply to a post by TheEulerID:

nb. In looking at that £30bn then doubling it number, it's interesting to compare this with the Australian National Broadband Network (now largely de-scoped to fibre/copper hybrid). Last time I checked the budget for that, if adjusted for population differences, would be the equivalent of about £44bn in the UK).

As you have mentioned NBN, I have a few interesting links on recent stories, in the context of judging the cost.

First, the original budget for FTTH might have been shot. It seems the average cost has been $4,300 per home (£2,100) rather than the previous estimate of $2,400 (£1,200).
FTTH cost doubles

Second - the points made about the skilled workforce seem to be apt: NBN is facing the same kind of worker shortage ... which might mean Openreach end up with problems with emigration
Acute labour shortage

Jersey Telecom has also faced work force issues, number of them and various other pay/billing/hours disputes.

KC is a FTTP heavy roll-out but slow pace means that it is actually pulling the averages for constituencies/councils in the area.

As much as FTTC is a pain, its speed to deploy does have advantages, particularly now self-install is a lot more common.

In reply to a post by TheEulerID:

They do, however, have a much larger area to cover albeit that most Australian's live in urban areas. It should also be noted that the more remote areas will be served by satellite (albeit within that same budget).

I looked at this previously, and reckoned that 90% of the Australian population live in a way that is not dissimilar to the UK - at least in terms of population density.

It just means that last 10% get very, very, very sparse.

Don't forget that NBN also includes fixed wireless coverage for their "plain" rural areas (as opposed to the bush. I vaguely remember them doubling the target speed for this from 25Mbps to 50Mbps.

In reply to a post by eckiedoo:

To try to answer your other questions, I generally assume that there is almost 1 fixed, traditional copper line per head of population, although I suspect that what-ever that Total Figure is/should be, about 25% to 35% are likely to be to non-domestic premises.

In a presentation dated 2009 (which was a primer for the FTTC VDSL2 service they were planning to trial), one slide tells us there were:

- 5,600 exchanges
- 80,000 PCP cabinets
- 28 million line pairs
- of which 21 million were residential, and 7 million were business.

The annual results since then suggest they lost 300,000 lines in 2010 (the last of a few consecutive years of reductions in the number of copper lines), but then pulled about the same number back up to 2014, and in the last year have added 200,000 lines on top. I guess these numbers are closer to one line per property, rather than one line per head of population.

More recently a later figure of 19% appeared; and both certainly tie in with the progress I have noted on "my" FTTC-

Indeed - 19% is the figure quoted in the latest quarterly results from BT. 4.2m premises out of 22m covered.

So just before that additional set was added, it looks as though the highest figure attributable of 48 of 288, amounts to 16.6 %, sitting nicely between the 13% and 19% OR figures.

I have not got absolute Line figures for "my" PCP; but by checking against various aspects, it appears to have approaching 300 lines.

If you want the number of lines for a sample of around 35,000 cabinets, take a look at this thread. I created a table; there's a link in the middle of the first post.

particularly if that includes getting "D-Side" fibre to every flat in blocks of flats and similar.

It may be that running a fibre line over existing poles for several kilometres to a distant farm and such-like, could be a lot cheaper than to a house in the middle of a city.

It could be. It all comes back to the number of premises that can share the cost.

In reply to a post by MrSaffron:

KC is a FTTP heavy roll-out but slow pace means that it is actually pulling the averages for constituencies/councils in the area.

It is staggering to see the Hull constituencies (and some of the surrounding East Yorkshire ones) at the bottom of the list in your recent blog post.

I know that Hull council has something of a historical link to KC (though no longer a financial one), but it seems daft that they haven't chosen to run a BDUK project. KC certainly can't be telling them they're due to cover the city (and villages) within the next 3 years.

There's some strange politics going on there...

In reply to a post by MrSaffron:

As much as FTTC is a pain, its speed to deploy does have advantages,

People at the tail end of the deployment are already complaining about how long things are taking to be done. Imagine if we'd chosen FTTP ... and we'd only reached 20-25% coverage.

Thanks very much.

Gives us all, more solid figures to consider!

And if there had only been the 19% Upgrade within that 20 to 25 % Coverage, ie 2 to 2.5% would be Broadband Users!

Edited by deleted (Wed 01-Jul-15 13:06:08)