
Network access provider Openreach (BT) has this morning announced that they intend to expand their ongoing pilot of 10Gbps capable XGS-PON based Fibre-to-the-Premises (FTTP) broadband ISP technology, which over the next 12 months will be made available to “more than a million premises” across the United Kingdom. But a third of its full fibre network could be covered over the next 5 years.
Openreach’s existing full fibre network is currently still largely based off older Gigabit Passive Optical Network (GPON) technology, which places limitations on how fast it can go before capacity becomes an issue. For example, GPON supports a capacity on each trunk line of up to 2.5Gbps (Gigabits per second) downstream and 1.24Gbps upstream, which needs to be shared between several premises. The fastest consumer broadband download speeds on this network currently reach up to 1.8Gbps (120Mbps upload).
In response the operator has been busy running a customer pilot of a new 10Gbps (10,000Mbps) capable Passive Optical Network (PON) technology, called XGS-PON – the ‘X’ stands for 10, the ‘G’ for Gigabits’ and the ‘S’ for Symmetric speed. This is being offered at maximum symmetric speeds of up to 8.5Gbps and at present EE still seems to be the only broadband ISP able to offer such packages (here).
Advertisement
The problem is that Openreach’s pilot area for this has so far been limited to the Guildford and neighbouring Woking areas – available across around 40,000 premises, which also catches some nearby locations like Brookwood, Puttenham, Clandon, Shere and Worplesdon. But Openreach now plan to expand its availability to more than 1 million UK premises by the end of August 2027.
Over the longer term, the business said they’re looking to expand the future-proofing technology to cover around a third of its Full Fibre network over the next 5 years, through a programme of further build, network upgrades and equipment modernisation and will “continue to review further expansion opportunities based on market demand, and feedback from its Communications Provider customers“.
Trevor Linney, Openreach’s Director of Network Technology, said:
“Our Full Fibre network already offers Gigabit capable broadband to more than 23.4 million premises, and more than 9.4 million customers have upgraded so far. As demand for data continues to grow, our investment in XGS-PON technology will enable us to offer even higher speeds over the same network, and the additional capacity needed to support the UK’s digital future.
By continuing to invest in upgrades, we’re delivering a platform that will power the next generation of technology, from AI-enabled services, smart homes and future mobile networks, to cloud-based business and public services.”
The upgrade is necessary for Openreach to remain competitive with their many rivals in the wider UK market, most of which have been using XGS-PON technology for years and some are already going beyond that. But the operator’s limited XGS-PON availability and slow deployment may yet hinder its adoption. Part of this might reflect the fact that their engineering resources are still more focused on new FTTP build than upgrading their existing network.
On the other hand, it’s worth noting that it will also take time for the operator’s many retail ISP partners to fully adapt, due to the need for new capacity links (trials of 100Gbps Cablelinks aren’t expected until 2027 – here) and investment to support such speeds. In addition, second tier wholesale suppliers are often slow to adapt (e.g. BT Wholesale), which is another reason why some retail ISPs can lag behind.
Advertisement
Consumers who adopt this service, even if they’ve already had an Openreach based FTTP line installed, will typically require another engineer visit to install one of the operator’s latest 10Gbps capable Optical Network Terminals (ONT) inside their homes (i.e. this is the optical modem that goes on your inside wall, before being wired to a router from your ISP).
The classic challenge with packages this fast is that most consumers would struggle to harness those top speeds, usually due to Wi-Fi/device limits and any limitations of the online servers you’re connecting to (Why Buying Gigabit Broadband Doesn’t Always Deliver). But if you’re happy to pay for it, then why not. The rest of the internet will catch up eventually, as it always does. The power of marketing is often in its ability to sell you something that you don’t strictly need, yet.
UPDATE 9:48am
In terms of the deployment plan for XGS-PON’s future expansion, Openreach has informed ISPreview that they’ll deploy it across “various locations“, including, but not limited to, large towns and cities (as we’d expect for obvious economic reasons – lots of customers in a smaller area). The first phase will be a selection of headends in handover Exchanges in Southern England. In later phases they intend to expand coverage to Exchanges across all four nations.
Advertisement
Advertisement
Seen some of the combo cards in a few sites, but a long way to go
Pretty slow rollout I must say! Only 1/3rd coverage over 5 years. I guess they will focus on areas where GPON’s more limited shared capacity is already becoming an issue.
> This is being offered at maximum symmetric speeds of up to 8.5Gbps
Is this right? I thought they were only offering asymmetric speeds.
Symmetrical speeds will br offered at outrageous prices so that their leased lines aren’t hit.
Consumer packages are asymmetric speeds, so far as I can tell from how Openreach are positioning it. But their network profiles do support symmetric speeds (likely as premium business options). This is a bit like how you can get 220Mbps uploads on their existing FTTP network, but that’s only intended as a business plan (for consumers it’s 120Mbps).
Question is whether there’s any evidence that their GPON network is having any congestion issues. Personally I’ve never run speedtest and not got my advertised speed (900/109). I doubt whether many consumer lines even get close to maxing out their connections very often unless running speedtest. I bet 150/30 is probably more than adequate for 90% of consumers who just want to watch Netflix etc & the improved stability & lower latency of FTTP are bigger factors.
Yeah I have no evidence of it, of course, since they don’t publish anything. But since they sell 1.8 Gbp/s over this and the 2.5 Gbp/s is split between up to 30 premises (I believe) it’s not hard to imagine in high take-up areas that they start hitting the cap at peak times. 1.8 Gbp/s is very easy to max even with some consumer services e.g. through a Steam download or whatever.
I suppose offering 8 Gbp/s over XGS-PON might run into similar issues but unlike 1.8 Gbp/s it’s pretty much impossible to find a consumer service that can deliver at anything like that speed.
I suspect in practice the chances of a port having 30 active customers, a 1.8G customer and the 1.8 customer and the rest (probably mostly 100M or less) maxing out their connection is low. Probably half or more of them are almost idle, and most those that are active will be using sub 50M
It’s worse than it appears. Assuming the trial is successful, I expect that going forwards new deployments will be exclusively XGS-PON. That “third” includes the new deployments. So the pace of upgrades of existing GPON areas will be very modest.
I don’t understand why Openreach is so stubborn to provide symmetric speeds especially since all the others are offering it. If I want symmetric Hyperoptic is currently my only option.
BT has been waiting on for the market for such services to both exist and to be profitable.
Openreach and BTwholesale always protect the business first than the residential for better upload and symmetric speeds for some reason.
I do remember BT isp say to me most peoples at home don’t need more than 1Gig FTTP and don’t need faster upload.
“If I want symmetric Hyperoptic is currently my only option”
Then you have the symmetric option – what difference would it make to you if Openreach provided it as well?
That’s what the market is about – you have choices, and you choose the combination of performance and price that matches your needs most closely.
Hyperoptic seems to have frequent nationwide outages and being an EE Mobile customer I could probably get a discount with EE One. If EE offers 1000/1000 at a reasonable price I might consider switching.
>> what difference would it make to you if Openreach provided it as well?
An Openreach fibre allows you to pick from a wide range of ISPs. So far as I know with Hyperoptic you’re stuck with their vertically integrated ISP. Glad I have CityFibre here – best of both worlds!
Not so click of a switch like the BT Fanboys on here been chiming on about. Looks like a handful of exchanges have combi cards then!
So, back to reality, another 5 years of waiting for Brokenreach to do something that competitors have already done, and even then, its still asymmetric and much higher priced.
Shambles.
The term “Exchange” is so inappropriate these days.
and doesn’t matter. Some are in exchanges that aren’t going to be closed and some are handover points, headends, cabibets, pit chambers, whatever terminology.
I can’t see “Optical Handover Point” coming into common parlance in the same way “Telephone Exchange” has.
The use of the term “Exchange” is quaint and dated given that the last manual operator was retired in 1976.
Looking for a prize for quibbling about the word exchange? Even the article says that word.
Quaint you mentioned, just like Openreach’s legacy GPON product 😉 and asymmetric broadband even on XGS-PON……..
Again, the term”Exchange” is both quaint and dated.
Look like BT Wholesale will be very slow ahead (changed from 1000/220 to 1800/220 next August 2027)
In my exchange, barely anyone offers anything over 900Mbps. Zen’s top package is 900/100. We’ll get XGS-PON in about 2038. It does feel like even in the FTTP world, there are different tiers of service for different areas. We have Netomnia, but YouFibre is pants. My hope is that the Virgin/Nexfibre deal goes through and we start to get a choice of ISPs over the Netomnia XGS-PON lines.
What a joke. Still can’t get anything other than a 70mb connection in parts of Cheshire, suffering with 2g (edge) mobile service, and incompetence from alt nets who took the money from the government then did half arsed jobs installing to a couple of houses in some villages.
Yep. Similar story where I live too (Hampshire). FTTC is the only option as Openreach has cancelled rollout plans. Mobile network not worth mentioning.
We’re still waiting for fibre in large parts of London if it’s any consolation. Meanwhile it’s getting common to see at least 3 or 4 providers in adjacent areas.
The whole fibre rollout has been a bit of a disaster really with Openreach being too slow and providers just overbuilding each other instead of working to get fibre to everyone in the UK.
BenInLondon :Don’t make me laugh…. Openreach slow rolling full fibre out, build rate is among the fastest in Europe, passing roughly 1 million new premises every quarter and still going strong to reach targets.
Whilst I would agree with you in that many Alt-Nets seem pathetically slow and incompetent in their BDUK roll-outs, you are incorrect in saying that they “… took the money from the government then did half arsed jobs installing to a couple of houses …”.
They only get the money afterwards and then only for the properties that are RFS and audited as such.
Openreach has the fastest build out of any operator. If they wanted to build faster they would have to employ more installers, buy/hire more plant/vans etc and then get rid at the end of it. There is only so much resources they can deploy at any one time.
I’d be happy to take 70mbps over the current poor 20-30 at best offered via FTTC here in Copthorne!
@beinlondon
Communityfibre cover large parts of london, hopefully that is/will be an option. in case you didnt realise, Openreach will only be covering upto 30million UK premises, theres currently around 33.5m uk premises, so 3.5m premises ( or more ) wont be getting Openreach full fibre. guessing thats why EE are partnering/offering starlink in a few months
@delta
lots of copthrone has openreach full fibre ( as per think broadband map ), looks like an alt net is there also, prob hey broadband
@tyler @Big Dave
Openreach may be faster now. But that’s like saying you are running faster when everyone else is about to finish or has finished. Most of Europe is ahead of us in coverage and adoption. Countries like Spain haven’t just finished rolling out FTTP, they’ve decommissioned all their copper – something Openreach will still be doing for many years in the future. And what will happen once OR reach their current goal – are they going to keep going at the current rate.
@graham I’m aware of all the providers – I even recommend CF to others since my family have it. But they only cover about 40% of London properties. Near us, Openreach are overbuilding them as well as other alt-nets. Hence the problem – lots of overbuild leaving significant numbers of the population without fibre.
That’s the thing. Openreach’s build speed is not the issue – it’s /where/ they are building. For instance, they will happily overbuild to a property that already has 2 full fibre options while ignoring the property 100 metres away that has none, because it’s more expensive to build to that one.
That is exactly what is happening where I live.
The rollout focus should be based on coverage, not overbuild – at least at this stage. Openreach may have built to 1 million properties or whatever this quarter but if 90% of that was overbuild, it’s hardly helpful is it.
The ignored properties stay ignored because of slightly higher build costs. Government has a pot of money set aside (allegedly) but those funds are not made available easily or only made available using the most convoluted programs using very long and complicated processes.
Surely there is a better way to handle this?
Alt-nets – yeah don’t get me started. Complete and utter shambles, the lot of them.
@sonic what part of hampshire are you based?
@sonic – totally agree. I’m perhaps a little more accepting of the alt-nets, some are good and some are bad. But the land-grab that we have just incentives poor build / install quality and doesn’t incentivise getting as many properties with fibre installed as possible.
Are we saying here that Openreach should have to roll out to all, even if out of budget, but altnets can continue to pick and choose ?
I think the government should step in and cover uneconomical properties, but not too quickly(or they’ll be giving out handouts for commercially viable builds), but I don’t think it should be on one provider
It bears repeating that we are not only in a competitive era, but one where Openreach is required to give away its crown jewels for pennies. That will be part of any decision of whether or not to reach an “expensive” property as part of the commercial rollout. Would it be in Openreach’s interests to do that hard work only for an altnet to have the right to use it on a parasitic basis?
It was a no brainer for the GPO or for the monopoly era BT to do such jobs knowing that they’d be guaranteed a permanent return. Not so now.
I don’t think it would be reasonable for Openreach to be forced to rollout anywhere or even to be blocked from upgrading their existing network in a given location. Some people will gripe because it hasn’t got to them yet but the statistics don’t lie – they have the biggest and fastest rollout in the UK.
It would make no sense for Openreach to have concentrated on the hardest and most expensive to reach areas while altnets cherry pick and win Openreach’s customers over.
@Jennifer90s – Winchester
@Martin – I’m not singling out Openreach. All I’m asking for is a more sensible approach to the fibre rollout. Openreach and all the other builders.
Funds (£5 billion) have been set aside to help with the more expensive installs. I hate the term ‘rural’ and ‘hard to reach’. It just means ‘expensive’. By rebranding as such, there is a better chance for urban areas to receive a portion of these funds.
If a property/street/postcode is “too expensive” to roll out to, the installer should be able to present the evidence and unlock funds. Surely that can’t be that hard to work out? That will ensure no one gets missed out.
Another thing is the incentive part to ensure ‘coverage’ and to discourage overbuild. I don’t know, maybe some stricter regulation can help here. But as I said above, unlocking additional (government) funds will definitely help with that.
Some properties having 4, sometimes up to 5 separate networks made available to them while millions of properties (many urban) have zero choice 10 years after the start of the build programme simply doesn’t make any sense to me. How is that beneficial to anyone? How does that make commercial sense?
1. Incentivise installers to not skip any streets/properties by making additional funds available quickly from “the pot” for properties that go over the threshold
2. Discourage overbuild (point 1 should help with that) or other sweeteners. Or tougher measures until blanket coverage is reached.
3. Ease planning regs (for digs, new poles etc.)
@Martin round here, OR seem to have started with the more uneconomic properties! Installing new infrastructure past properties that already have a choice of 2+ providers, but missing our relatively new ducts and no competition.
But in general I would have looked at other models. Instead of paying alt-nets to compete with Openreach, let them compete to build network for Openreach. We might even have got a point-to-point network like the Swiss are building, with the ability to have your fibres (they get 4 per home) connected to your chosen operator. It’s a higher cost, but then we also wouldn’t be building multiple networks to each house instead of competing at the ISP level.
Your ignored property is not ‘slightly’ higher build cost from what you’ve said, sonic. No space in footway for duct, nowhere to put a pole. Closing a street and digging in the carriageway for nearly a hundred metres isn’t slightly more expensive than regular PIA £200 a pop stuff.
It’s sometimes astonishing how far behind we are, despite the current fibre frenzy. I was just travelling out in Malaysia with my wife’s family. Her home town has had full fibre for years, long before here, and I had solid 4G/5G coverage everywhere I went, even out in quieter areas. Never a blip in mobile data coverage. I don’t know how Britain can’t seem to manage what the rest of the world already achieved years ago.
@BenInLondon
I think one of the main reasons why the UK is “behind” in copper removal is due LLU in the UK. Those operators didn’t want to decommission their DSLAMs.
@Martin – also, my basic logic tells me that rolling out to properties that have zero full fibre options (even if builds are more expensive) is more likely to pay dividends than overbuilding to properties that have 2 or more existing full fibre options.
This is why the whole alt-net build programme leaves me utterly perplexed. Why only overbuild? Competing with Openreach is hard enough. Add to that, in many areas they build to areas where another alt-net already present. It makes zero sense to me.
Nexfibre is another classic – they have basically left their copper to rot in places like Winchester while rolling out FTTP to areas where other fibre builders are well established for many years. Many households are leaving the traditional TV/triple-play providers and simply going for fast broadband and streaming services/Freely. With an ISP choice of one on their network, once people leave, they are never going back (especially given their poor track record of customer service and insane pricing).
Make it make sense.
@sonic nexfibre have no copper or customer services: they don’t provide to end users. You’re thinking of Virgin Media. The plan last I heard remains to complete the overbuild by 2028. Winchester is always going to be tricky and relatively expensive due to the original duct build, geography and conservation area.
@Polish Poler – I don’t know what the actual problems are – it was pure speculation on my part. No one will tell me what the issues are or what the build costs are for our street. I have tried to find out but no one will tell me. I even offered to foot part of the costs or to organise a community effort but no one will even listen or engage.
Also, I’m aware of Virgin Media and Nexfibre. Ultimately it’s all the same parent. Winchester city centre might be more challenging but we live outside the city in a bog standard residential area that has nothing to do with any conservation issues.
“my basic logic tells me that rolling out to properties that have zero full fibre options (even if builds are more expensive) is more likely to pay dividends than overbuilding to properties that have 2 or more existing full fibre options.”
On the contrary. A property that has only Openreach copper available provides no return for Openreach to deploy fibre: they are capturing their own customers, and most of them will be paying only pennies more for FTTP than FTTC. The prices are regulated, and almost everyone takes the slowest service. Happy customers, but no financial return.
A property that has one or more altnets is bleeding customers away from Openreach. Building fibre there helps to retain or recapture them: a direct financial return.
Two massive issues, the g-pon and physical roll out is very expensive, which if alt nets weren’t doing xgs-pon then it would be very profitable for or.
But xgs-pon is being used by alt nets – thats a danger for or, add in that most will skip 25gbit pon (28gbits) and jump to 40gbit pon (50gbits).
OR will face the double problem of rollout costs and speed competition from alt nets – and increasing speeds.
So the balance to moving to xgs-pon is a balancing act that will see OR get it wrong……… like in this case
Andy Burnham won’t be too happy about this (why south all the times!) Nothing for North!
Hello Max,
Why start down South?
Because that’s where the mugs with the money are!
Whilst I can understand 10Gig (OK 8.5Gig) symmetric for folks that work from home on very large data files, e.g. video editing, I don’t get the point 8.5Gig asymmetric.
What is its use case(?), apart from the bragging rights of course.
I just had to shift 1.7TB of dat between S3 providers – and it was faster to do it on my connection than to use the online migration tools.
Also when I want to play the latest COD and it’s 700GB – That takes about 8 mins to get.
Next?
Hey I drop into KBS I live in the sout, where I live was removed off the list, we are not even connected to fibre , still run on copper wire connection
S3 is an Amazon product. How did you find someone else offering it?
@Polish Poler:
Cloudflare and DigitalOcean offer S3-compatible services.
So glad I live in france for fibre tech alone! The UK seems to stuck in some sort of inexplicable slow lane. can somebody explain to me what the issue is?
And yet people on Orange whinge about slow speeds on reddit- so yeah,no
You will find out if the acquisition/breakup of SFR goes ahead.
When I spoke to BT Openreach earlier this year they had plans to build in PO14 4 area, but then it changed to now they have no plans. I questioned why and they said duct damage, so no plans at all. So the only option in the close is Virgin.
you have to contact OR directly – if you round up your area to demand it, and get OR CEO involved things will start to shift.
my block of flats wasn’t being done until I got both sides involved and a right kick up the arse to the building owner. Needless to say I have OR GPON – ok not as good as XGS-PON, but better than a failing 70mbps line that was dying.
you’ll get there.
Aiming for full network coverage in five years would still seem pretty slow, but ONE THIRD of the network in five years seems insanely unambitious. My understanding was that only ComboPON OLTs have been installed for years now, making the rollout just a case of putting new ONTs into homes, so why does this need to take so long?
It isn’t really relevant whether or not people need these speeds yet, the fact is that if an altnet is able to offer someone 8Gbps for the same or lower price than that at which Openreach is offering 1.8Gbps then people are obviously going to go for the higher number. Surely Openreach realise that things needs to be a bit quicker than this in order to remain competitive?
I think they had plenty of network capacity left using their legacy GPON kit to use 1st. The combo kit only being installed where existing GPON capacity was unavailable.
The retail arm of BT reported a slight net increase in customers at a time when Openreach saw net losses. So it would appear that customers are not all that bothered about max headline speeds above a certain point.
All providers will face a bit of a challenge upselling to multi gig when the connection they already have feels more than fast enough. Same reason why Openreach still deploys cheaper GPON equipment and isn’t rushing to compete with the debt laden altnets.
Meanwhile, users of this website are utterly convinced that Openreach will collapse unless they offer nationwide 10 gig symmetric tomorrow.
It’s not just the speed, it’s the whole package, e.g. BT will offer a mature phone solution, Community Fibre’s has… issues as far as we found (missed calls showing up to others as being engaged) and the TV solution looked so lackluster we didn’t even try it (read: try for 24 months).
Yes you can then unbundle, but that’s not what most consumers want, they are used to a single solution and are likely to be disappointed.
Add to that most people probably don’t switch at all if it’s not broken.
I do have a use for more upstream than BT would provide at a reasonable price so I got it, and the download speeds are nice too (especially in the world of AI models and massive game downloads) but I can’t honestly say most would need or be able to use that speed.
I wonder if the issue might not actually be the ComboPON OLTs but perhaps upgrading the Cablelinks to handle the additional traffic?
It’s Great. People are buying 8Gbps on YF and have no way to use it – but it’s cheap so it supports us actual heavy users
@Ivor How many customers know the difference between GPON and XGSPON? For the vast majority if Netflix streams without buffering that’s good enough. I would bet for a large majority of customers GPON will be sufficient for the next 10 years.
Would be good if many ISP’s feedback here to let us all know what they are thinking of Openreach decision.
Sorry, what does “a selection of headends in handover Exchanges” mean?
It could mean that certain headend vendors are being selected. The handover or parent exchange is where the headend is located, some may have Nokia and Huawei so may be that only Nokia headend is looking to be upgraded
I think it means they’ll do some area covered by some handover exchanges. These exchanges might cover many hundred of thousands or more properties. That’s quite a few ports at 30 properties per port. So some ports will get done some not, likely new ports will get it but they won’t replace all the old ones in one go
inb4 Adrian complains why people need these speeds
Insufferable isn’t it.
So they’ll do it slowly, argh!
Only Hyperoptic can offer me symmmetric and then it’s capped at 1GBps on their super reliable network with their highly qualified support staff…
We need more ISPs to compete with speed at the higher end but Openreach are slowing the UK down!
Offcom should be most concerned about coverage not competition. Rural coverage through Project Gigabodge is still poor so only appeals to civil servants that work for performance not substance.
The suggestion above that whoever is building should be able to get help with expensive locations seems a lot more likely to achieve universal coverage than Project Gigabodge. All that money wasted on contracts would actually be delivering universal coverage..
Not sure what that has to do with this article but take it your place still isn’t covered.