Openreach Confirm G.fast Broadband Rollout Paused Until 2021 UPDATE
Openreach (BT) has confirmed to ISPreview.co.uk that they do not intend to build their G.fast based “ultrafast broadband” (100Mbps+) ISP network to any more UK homes until April 2021. The service has been on pause since last year, when the operator re-focused toward gigabit-capable Fibre-to-the-Premises (FTTP) technology.
In a hybrid fibre G.fast setup the fibre optic cable only goes as far as your local PCP street cabinet (similar to FTTC / VDSL2 technology) and then runs to your home via an existing copper line. G.fast can make use of 106-212MHz of spectrum and thus delivers its best speeds for those within c.100-300 metres of a cabinet (variable speeds of up to 300Mbps), but it’s fairly useless for those on longer lines.
By comparison FTTP takes a high-capacity optical fibre cable all the way to your home, which removes the reliability problems of copper and makes gigabit (1Gbps+) speeds a possibility. The advantage of G.fast is that it was quick and cheap to deploy, but Openreach’s cabinet (side pod extension on PCP) based approach hasn’t always produced the best speeds and suffered as the political impetus switched to “full fibre” and “gigabit” networks.
Advertisement
Suffice to say that few were surprised when Openreach announced, in September 2019, that they would review the plans for G.fast build beyond March 2020 (here). Unfortunately the conclusion of this review ended up being delayed by the COVID-19 crisis, but the operator has just now begun to inform UK ISPs about their latest position.
An Openreach Spokesperson told ISPreview.co.uk:
“We recently announced that we’ll be building Full Fibre broadband technology to 20m homes by the mid-to-late 2020s, so that’s our priority.
We do still expect to use Gfast in some circumstances, but we don’t expect to be building Gfast to more premises this year.”
In other words, G.fast deployments will officially remain on pause until at least April 2021, but in reality it’s unlikely to ever play a big role in the future of UK broadband again. Openreach does however still see a potential role for G.fast in some specific scenarios, such as via in-fill in certain hard to reach areas (including certain types of MDU / apartment blocks).
The decision makes perfect sense in light of the greater focus on full fibre (right across the market), the desire to withdraw copper services further down the line and the fact that take-up of the G.fast service by consumers hasn’t been particularly strong (here).
However the decision to keep G.fast on pause may create some problems. For example, ISPreview.co.uk is aware via the London Grid for Learning (LGfL) that some schools in certain authorities are being promised G.fast this Autumn 2020, which judging by the above announcement may not happen (they may go with slow VDSL2 instead or wait much longer for FTTP).
Advertisement
UPDATE 4:32pm
We’ve had a small update from Openreach.
A Spokesperson for Openreach added:
“Although we don’t expect to build Gfast to more premises this year, the technology is available to 2.8m homes and businesses across the country. Anyone who has already placed an order for Gfast will still be able to get a service, and we will continue to accept Gfast customer orders on the 2.8m premises so customers can benefit from faster speeds. People can see what broadband services are available to them by using the Openreach fibre checker.”
Mark is a professional technology writer, IT consultant and computer engineer from Dorset (England), he also founded ISPreview in 1999 and enjoys analysing the latest telecoms and broadband developments. Find me on X (Twitter), Mastodon, Facebook, BlueSky, Threads.net and Linkedin.
« 44% of UK Online Multiplayer Gamers Still Complain About Lag
TUC – Invest in UK Broadband and Create 40,000 Telecoms Jobs »
Latest UK ISP News
- FTTP (6035)
- BT (3643)
- Politics (2722)
- Business (2442)
- Openreach (2407)
- Building Digital UK (2331)
- Mobile Broadband (2150)
- FTTC (2084)
- Statistics (1907)
- 4G (1822)
- Virgin Media (1769)
- Ofcom Regulation (1585)
- Fibre Optic (1470)
- Wireless Internet (1463)
- 5G (1411)
Good as G.fast are useless and a waste of time! Should never go ahead in the first place!
Useless and a waste of time? I wonder how many of those subscribers now on G-Fast would want to see their connection speeds revert back to pre-upgrade levels?
I find the faster speeds and lower ping times of my g.fast connection rather useful when working from home compared to the previous fttc connection. I’ve had it for a year and a half now.
City fibre are currently installing cables in the nearby streets though probably be some months until activated. Well the service come with a static IP and IPv6 for me to make the jump?
Remember when you bought a G.fast modem and ordered the service, Sir?
It wasn’t the greatest decision ever but someone who bought it as it was significantly faster than VDSL perhaps isn’t the greatest judge.
G.fast was slower for me than VDSL but it is a long line.
I didn’t even meet the EE minimum guaranteed speed so they let me end the contract without penalties
Upload is still awful.
FTTP or die.
I am sure for some G.fast is good. For you not so much. Last month you said you got plusnet with money back for referrals and last week it was £10 a month, which one is is Max?
Anyway with Gbps starting to creep into homes and Virgin going up and up and up. I am sure G.fast will be as dead as FTTPoD in the near future.
I might be wrong – but then again..
Useless and a waste of time?
I was looking forward to it in my block of flats because I will never get FTTP.
They suspended it just weeks before it was due in my postcode and now I’m stuck with FTTC and can never get better.
You might think it’s useless but for others like myself it would have been a welcome speed boost,especially when I’m in a town that really should have better speeds by now.
I bet if you offered them the money it would be in soon ish – seems they dont like to spend any money which is theirs
Sandra
I am happy with FTTC 80/20 line only for £10 a month! 🙂
I get 290Mb down and 48 to 49Mb up on my G.fast. I’m paying barely more than what I was for VDSL2. Not sure how that’s useless. The upload speed, for instance, is much better than what you can get from the majority of VM DOCSIS packages, unless you’re happy to pay an extortionate amount to a business that wilfully oversubscribes and has no scruples.
For those who can get G.fast I would recommend it and now that the rollout has been postponed, possibly forever, you can get back to quietly waiting for an FTTP solution giving you symmetrical gigabit which the majority of households will not come close to wanting, let alone needing, for at least a decade but probably much longer.
It’s not a waste for us, we got about 50Mbps before gfast and now we generally connect at 200Mbps and the line is way more stable than before, we used to drop connection every few days before and now we stay connected for months at a time, obviously you can’t stop things like updates to the hub etc but all in all it’s a great service.
Not saying no to FTTP though as either way it’s the future.
@James
I would tend to agree.
If you can get good GFast go for it – we do at home and we get 289/48-9.
I do t think it is any more or less stable than FTTC.
I tend to be fervently anti GFast on here mainly because it is a dead end and also because of the lack of generalisability – you could do a PCP and discover that half the street can get it and the other can’t which is a bit useless.
I honestly believe OR would have been better off forgetting GFast and implementing VDSL 35a profile – which would have been even quicker and even cheaper – new line cards in existing DSLAM or on the newer ones a firmware upgrade. It would have roughly doubled national headline speeds.
@Shaun McDonald:
> City fibre are currently installing cables in the nearby streets though probably be some months until activated. Well the service come with a static IP and IPv6 for me to make the jump?
Cityfibre don’t provide any service to end-users. It’s an open-access wholesale network, but right now, Vodafone have an exclusivity deal for residential properties. The Vodafone service is called “Gigafast” and as far as I know doesn’t offer static IPv4, or any IPv6 at all. But it’s cheap. You could use the AAISP L2TP service over it (for £10/month) to get static v4 and v6.
Over time, expect Sky and Talktalk to offer residential services on the Cityfibre network. Currently Sky provides IPv6, Talktalk does not.
G,Fast useless, like hell, I’m only 35m from the cabinet as the crow flies and the phone line follows a fairly direct path. I’m on the 330/50 package and being so close to the cabinet my router (Asus DSL-AC88U) is syncing at 330 down with a true download speed of 37Mb/s, coming from a 80/20 connection I think G.Fast is awesome.
I work for a MAT that looks after a few schools and they all have Virgin Media connections.
Virgin took over the schools Broadbands back in 2012-2014. Bit concerned which schools have not had there broadband upgraded to fibre. Surely the LGFL should be doing more to get these schools onto a fibre connection.
They even have 1gb (Download and Upload) available for secondaries. Which they are actively trying to get secondaries onto. I believe primaries are supposed to get 200mb (D and U)
LGfL have made a pledge2020 to upgrade all broadband for free to their participating schools. They are committing to 100Mbps for Primary and 500Mbps for secondary and promising that equipment will support 1Gbps. As you say they have a long relationship with VM and much will be filled by that relationship in the short term. The reality is that schools profiles remain primarily asymetric (O365, Gsuite, LGfL apps etc) and dedicated leaselines are expensive so the cheaper business options over HFC may suffice. LGfL will utilise with their customers any LA schemes including LNNF funded links between public buildings many of which are underway. I think it is sensible to consider G.Fast (if available and technically confirmed across playground) where there is no alternative is available and if they can consolidate to get a deal through an ISP so much the better. Especially as LGfL is losing some schools and gaining others. Its being pragmatic.
We know the history but its there so lets use it for the next 5 years if we need it.
LFFN
Guy i know who runs an ISP has authority consent to use one of the many spare dark fibres going into the Primary school. So they are about.
@Mark Do you know/have an idea of when they due to start offering G.Fast where they have already installed it?
Watch the G.Fast deployments start up again once interest rates rise and the cost of capital increases…
FTTP is great when interest rates are low and the Investment Banks are looking for long term, secure mortgageable assets…not so attractive when they can get higher rates of return elsewhere.
I doubt it, the market dynamics are now firmly based around FTTP and if Openreach were to spend big on G.fast again then it.. well, let’s just say we’re past that point.
G.fast is now all but dead its never coming back and fttp will happen regardless of interest rates – which are remaining low anyway…
Gfast always was a short term, keep up with VM, solution.
Now VM are going full 3.1, downstream, and therefore 1G – GFast no longer cuts it speedwise.
As GFast has a lifetime window of maybe 10 years, at most and in some areas that are getting FTTP zero years, then that is not long to get a payback. Nobody is interested in investing in GFast commercially. There is plenty of cash chasing FTTP deployments.
With both the government, opposition and private investors pilling into FTTP the will & momentum is there to get the job done.
As @MJ says retiring the copper network is a priority and that does not fit with GFast.
I suspect that existing pods will be made live, except where the area is already earmarked for OR FTTP, and that will be that.
GFast isn’t investible and that is what killed it in the end.
@ Mark Jackson
May I remind you of a discussion in 2013.
Had the BT Group invested in ANY real replacement maintenance their costs wouldn’t be anywhere near as high as they are now. Small wonder they need so many engineers which everybody can see whenever they drive a few miles – there’s swarms of them.
One of the last real engineers with a practical understanding and a healthy disrespect for bean counters is Dr Peter Cochrane. Below you will see his comments to Neil Fairbrother’s G-Fast condemnation dated 22 August 2013.
https://neil-fairbrother.squarespace.com/blog/2013/7/1/gfast-a-high-speed-cul-de-sac
All this was so obvious way back in 1986….but ‘real engineering and economics’ has been driven out of the telecoms industry. You can’t beat physics (loss and crosstalk) and you can’t stop Moore’s Law! Mini-DSlams are an insane option! To get network reliability and resilience you have to take out electronics not put more in! To get a ‘Green Network’ you have to reduce the amount of material used and energy consumed! And Mbit/s are not enough for an obvious future rushing towards us. We have to start talking Gbit/s. But if you want sub-optimal industries and a population who just sit and watch sport on TV….just keep installing copper!
With a copper network you need over 6000 switch sites in the UK. If you install optical fibre this number drops below 70. 20,000 man in van crews goes down to 1,000, and all water ingress related faults just go away. Now redo the economic argument. Go figure!
Peter Cochrane
Hence why Openreach are installing FTTP.
Well, semiconductors have been going the put more route in for quite some time, abd we won’t be able to get much of what we’ve got from copper today without software based signal processing (e.g error correction). We went from 1500 baud to add several zeroes.
So your engineer may be smart and good indeed, but maybe not open-minded enough. bean counters are useful in keeping engineers from endlessly pursuing the holy grail.
‘Had the BT Group invested in ANY real replacement maintenance their costs wouldn’t be anywhere near as high as they are now.’
Replacement or maintenance. Which? Openreach still can’t replace copper with fibre to this day.
‘All this was so obvious way back in 1986….but ‘real engineering and economics’ has been driven out of the telecoms industry.’
So obvious the rest of the world was falling over itself to build full fibre access networks and everyone else is at 100% now having started 2 decades ago.
Wait. No. They aren’t. CATV was built with a fibre pair per 5,000 premises passed in many cases, didn’t start using full fibre until the mid-2010s, apartment-heavy cities were using VDSL in the mid-late 2000s and fibre build in anger didn’t start until the 2010s in most places with 2000s deployments being heavily cherry picked and almost universally either municipal or totally vertically integrated and bound to an incumbent – no open access.
‘But if you want sub-optimal industries and a population who just sit and watch sport on TV….just keep installing copper!’
The vast majority of new build gets fibre now and Openreach have entirely removed the option for copper. This after initially making it free for certain sizes of sites and developers continuing to ask for copper.
In 2017-18 more customers were purchasing 40 Mbit VDSL than 80 Mbit where the option existed. Most FTTP customers were on 55 Mbit or lower. What does that tell you?
‘With a copper network you need over 6000 switch sites in the UK. If you install optical fibre this number drops below 70. 20,000 man in van crews goes down to 1,000, and all water ingress related faults just go away. Now redo the economic argument. Go figure!’
Dr Cochrane: meet Ofcom and Openreach’s regulatory obligations. Ofcom: Dr Cochrane. Sky, TalkTalk, Vodafone, Virgin Media, SSE Telecomms, etc, etc, have no objection to Openreach shutting down 90% of exchanges and rendering equipment colocated in them serving 90% of UK homes and businesses redundant on a whim, right? You guys are cool with waiving the regulatory obligations on MPF unbundling, right?
It has taken years of negotiation and a substantial move to FTTC, swear word as that is to ‘real’ engineers, to permit the possible reduction to about a thousand ‘switching sites’.
‘One of the last real engineers with a practical understanding and a healthy disrespect for bean counters is Dr Peter Cochrane.’
On behalf of every engineer that isn’t a real one with any practical understanding thanks very much.
I would point out, however, that ‘real engineers’ that don’t care about the people holding the purse strings tend to achieve exactly 0% of what they want as the people who hold the purse strings are the ones paying.
Engineers want to engineer. Bean counters want to count the beans. Project managers and others sit between the two of them, everyone thrashes it out and a solution is found that hopefully goes some way towards making everyone happy.
That’s how the real world works.
G.fast had potential as a stop gap, which was all it was ever going to be and no-one ever thought otherwise, however the manner in which Openreach deployed swung too far to the bean counter side. This happens sometimes and engineers that actually work in the real world can either resign, retire or do the best they can with what they have, which they did.
The comment:
‘And Mbit/s are not enough for an obvious future rushing towards us. We have to start talking Gbit/s.’
Shows how divorced from reality the good Dr actually is. It’s 2020, 7 years since his post. The average UK home’s bandwidth consumption is a couple of megabits per second at peak time with bursts measured over 15 minute periods to the mid-high double digits. This with usage increasing at 30-50% per year. Homes that apparently needed dedicated point to point fibre 7 years ago are maxing out 100 Mbit lines for a matter of hours per year.
The new FTTP approach is far better. Belated, but better. It is to be saluted, and the pace at which Openreach are tackling the task is admirable, as is the entry of CityFibre et al.
Either way you’ll forgive those of us who live in the real world, while we may not be real engineers, for continuing to do what we do to provide better connectivity to businesses and homes.
While we’d all love to install a small rack of optical, routing and switching gear in each home and office to provide pluggable upgrades up to 100 Gbit/s at a time when the average cost of broadband is around the £25 a month mark, most customers are happy with 76 Mb/s or less even when they have the option of higher, and the incumbent has to run an open access network with regulated pricing, alongside having to pay people for civils work and installation not having an army of volunteers we do what we can.
Dr Cochrane and yourself can evangelise, and even help dig the odd soft trench and splice some fibre. The rest of us don’t have the luxury of making a living through evangelism and have bills to pay that require gainful employment. Treating the people paying you with disdain rarely maintains that gainfully employed status.
TL;DR apparently insulting every engineer that has come along in the past 24 years isn’t a good look and I doubt you have the experience or qualifications to be in a position to judge us. Neither does Dr Cochrane. He may be in a position to judge the science, academia and theoretical engineering side but has no business interjecting himself into the discussion of practicalities.
We’re given budgets. We can present a business case for more and it gets accepted, rejected or negotiated. I’m afraid the telcos are privately held businesses and that’s how the private sector works. Dr Cochrane is more than welcome to go visit CityFibre I imagine and ask them why they aren’t running point to point fibre everywhere, aren’t deploying fibre to the farm but instead spurring off existing metronet or using public sector anchor tenants as a spine, and are using every technology imaginable to get their costs down. In Leeds alone they are using duct and pole PIA, their own trenching to reach said Openreach ducts and poles, erecting their own poles, using microtrenching and standard trenching.
Either way he has no right to judge the engineers working on this today from most points of view and you have no right to judge them on pretty much any, much as I’ve no right to judge the co-operative/community/mutual projects you’re working on.
Good luck with your ongoing community projects and every success with them.
CarlT, I couldn’t have said it better myself, I swear too much.
@Carl Feeling better after a well earned vent? 🙂
Moore’s law broke some time ago. It’s perfectly possible to run a copper last mile without 6,000 exchanges once you have FTTC and VoIP because the fibre can go wherever you like.
What was the use case for FTTP in 1986? Most households didn’t have a computer and of the few that did an even tinier number had modems. There was no widespread uptake of networked devices until the late 90s.
The only conceivable case was for cable TV, which BT was specifically prohibited from doing by the government of the day.
Adding to some of the well made points above, let’s not forget that technology has matured a lot since 1986 and that decisions made then might not have aged well. For example, you may wish to check what happened with the infrastructure installed for the BT cable TV franchises in Westminster, Milton Keynes etc. (hint it became obsolete and was ripped out, even the fibre as the star topology was different to that now used by PON).
Technology for its own sake, with no credible path to any ROI is a quick way to bankruptcy when investors lose patience. A quick search will yield plenty of examples if you’re interested.
“Hence why Openreach are installing FTTP.”
More than a decade too late I might add!
@CarlT: “Replacement or maintenance. Which?”
I think he’s talking about replacing copper with fibre. Old copper infrastructure has a much higher maintenance cost than fibre.
Anyway, it’s good to see the G.Fast rollout being stopped now, by and large it was more of a GFarce!
@GNewton but thats just not up to BT.
@Joe Yes, thank you.
@GNewton: Then my point is proven. It is not and has never been up to Openreach when to replace copper with fibre. Merely taking over the broadband side via VDSL was a battle. Blame politicians for that one.
Incidentally 10 years ago Verizon in the US were building FTTP. They ended up building to a fraction of their footprint and sold off much of the rest where it lies in all its copper goodness. There was no business case, in the regulatory and competitive environment at that time, for mass FTTP, which is why no-one else was doing it either.
Carl – on Verizon, it wasn’t so much a failure for FTTH itself, it was more a casualty of a wider policy. They sold off entire chunks of their territory outside of the north east, whether or not they had upgraded it or ever planned to.
The now former CEO tried to bet the company on cellular services (therefore, why would they need pesky regulated wires?) alongside billions spent on failed media acquisitions, most notably AOL.
Under new leadership, they’ve quietly rowed back to embracing FTTH again, by steadily wiring entire exchange areas and actively forcing customers onto fibre where it exists. I wonder where they would have been if the AOL billions had gone into meaningful infrastructure investment?
Cheers for that mate. As a pressure jointer of 20 years plus I kind of feel a bit peeved of your opinion towards me. Never. I don’t. I’m a big boy and no snowflake.
We people in Glasgow Burnhouse Street G20 have been promist by bt open reach from 2016 we would have some sort of fibre 2020 still none mmmmmm??????
I’m thinking Openreach might be the people to ask about this one Mr Hill.
G.fast is useless, they built it in my town but due to the way my network is designed all the street cabinets are on the main road and all the streets are like further away from the cabinet. Even the houses closest to the cabinet don’t see much increase in speed.
G.Sluggish
G.SNAIL
Interested in when Openreach are going to finally bring FTTP to me, I’m currently down as in scope and FTTP has been brought as far as new properties locally but despite expecting to see them around near me his week the council have now gone ahead and resurfaced the pavements so not sure what to make of it.
I’m also not sure how one can get in touch with anybody in the know with a Openreach locking off all contact with outside world.
I’m on Kings Norton exchange which is a Fibre First rollout location and a lot of upgrading has already been done to overhead lines but UG lines remain in the dark with just 0-3mbps on checkers and many providers not wanting to serve lines due to averages (I guess)
Hanging on a 4G connection setup in the loft right now.
Before lockdown came into place I checked on BT wholesale checker and G fast was available as it was on both BT & EE consumer sites. Now that is over none of the checkers are showing G Fast as available so wonder what has changed in the few weeks that Openreach stopped installs.
Bt are not good at all can not sort anything out download and upload speeds thay say you can have shit internet for the highest price please recommend your friends and family so we can screw them over to …sort out internet speeds in hackenthore s12 and restore my faith
@Dan
You need to reset the troll bot to English!
Need to get country connected with fibre first before introducing more crap they cannot maintain.
Hi All
I’m after some advice on this post please. My local cabinet has been replaced after this article was published, and a new GFast pod installed also next to the cabinet. I asked the engineers completing the works and they confirmed it was for Gfast. I check the BT Wholesale checker daily, but nothing so far. Seems pointless that they would install the new cabinet and then not take it any further. What are your thoughts? This is my only hope as Tamworth (my post code) doesnt have Virgin Media, its been missed off the current 5g plans also.
Economics of things confuse me sometimes. FTTP requires mass re-cabling of areas plus replacement of individual connections to houses. How can this be cheaper than even totally replacing cabinets? Unless they are working on the assumption that they are going to have to FTTP at some point so might as well do it now?
Expectations are strange as well. For the majority of home owners, they really don’t need gigabit internet. Yeah, I know – stupid thing to say like nobody will ever need more than 640K of RAM 😉 25Mbps is fine for 4k streaming after all.
My cabinet and a few others in the town have also just had a g.fast pod added in the last 2 weeks (confirmed to me by the installing engineers), so not sure what’s going on here!?
What about areas served by a infill dslam aka, the Huawei All In One PCP and DSLAM? Can a G.fast pod bolt to them?