Full fibre vs FTTC vs cable vs 4G and 5G home broadband: the four UK technologies compared on speed, on latency, and on what you are owed when it goes wrong

Which technology to choose

Direct answer: For most households at most addresses the order is full fibre, then cable, then FTTC, then wireless. Speed is the smallest part of why. A fixed line from a provider that has signed both of Ofcom's schemes carries three protections, and 4G or 5G home broadband carries none of them.

Two things make that order less tidy than it looks. The first is availability: 82% of UK homes can get full fibre and 89% can get something gigabit-capable, but the remainder is a real number of real houses, and at those addresses the list above starts further down. 39,000 premises still cannot get what Ofcom calls decent broadband at all. The second is that the four overlap on speed. The fastest FTTC line Openreach sells is quicker than the slowest full fibre package most providers sell, so full fibre is not automatically the faster of the two on the plan you actually buy. Which means the technology name on the advert tells you less than the number next to it, and the number next to it tells you less than the guaranteed minimum underneath.

Before you sign anything, ask for the guaranteed minimum download speed in writing. If the seller cannot give you one, you have learned something more useful than any headline speed. That one question does more work than any speed test, because a guaranteed minimum is the only number on the whole sales process that a provider has to stand behind afterwards.

Run the full comparison to see which of these four reaches your address, or start from full fibre from every network or the cheapest deals.

How to tell which one you have now

Direct answer: Look at where the cable enters your home. A white box on the wall with a small fibre socket means full fibre. A cable running into a phone socket means FTTC or ADSL. A coaxial screw connector like an aerial lead means Virgin Media cable. No cable at all means wireless.

Your bill and your contract both name the technology, and the package name usually gives it away: anything sold as Full Fibre, Fibre 900, Gig1 or 1Gb is full fibre or cable, and anything at 76 Mbps or below on an Openreach line is FTTC. The word Fibre on its own settles nothing, because both use it. There is one rule that cuts through it: ask where the fibre stops. At the cabinet in your street is FTTC, at a box on your wall is full fibre, and at a street cabinet with coaxial the rest of the way is Virgin Media cable.

The quickest answer is a postcode check, because it tells you what is available at your address rather than what the country average is. A national figure tells you what is likely and a postcode check tells you what is true, and on this subject the two can be a long way apart: an address can sit inside 82% coverage and still be one of the 39,000 premises Ofcom counts as below decent broadband, because coverage is measured street by street rather than house by house. Openreach publishes its own explanation of why one house on a street can order full fibre and the one next door cannot.

One more figure worth sitting with before you go any further. 24.9 million homes can get full fibre and 12.4 million have taken it, which is 47% of the homes that could. So more than half the households who could move to the best technology on this page have not, and a good number of them are paying the same money or more for FTTC.

Our speed test tells you what you are getting now, what is FTTP explains the best case in plain words, and why can't I get full fibre when next door can covers the address lottery. A sister question this page does not answer is whose network you are on: how to tell which network your provider uses does.

All four compared, side by side

Direct answer: The first five rows are the ones the adverts compete on and the next three are the ones that decide what happens to you when something breaks. Buy a fixed line from a provider that has signed both of Ofcom's schemes and you get three things: a minimum speed in writing before you buy, a way out without penalty if it is missed, and money back automatically when the line goes down. Buy 4G or 5G home broadband and you get none of the three.

Three rows in that table are the whole argument of this page, and they say from a code signatory rather than simply yes for a reason: the protection travels with the company you buy from and not with the wire in the ground. Ofcom runs two schemes, both voluntary, and a provider can be in one and not the other. The next section takes them apart.

And one row deserves a warning rather than a footnote. The latency and peak-time figures are more than three years old. Ofcom closed the programme. Its final report covered March 2023, so the most recent measured comparison of latency and upload speed by technology in this country is three and a half years old, and no regulator has published one since. Nobody is measuring this any more, which is worth knowing before you trust any comparison of these four, including ours.

Go deeper on any one pairing: which is faster, FTTP, FTTC, 5G or Virgin cable, full fibre against part fibre against cable, FTTP against Virgin Media cable and fixed wireless against fibre. Or see FTTC deals, full fibre deals and 5G home broadband deals as they stand today.

Speed, and why the headline number misleads

Direct answer: The fastest FTTC line Openreach sells is quicker than the slowest full fibre package most providers sell, so full fibre is not automatically the faster of the two on the plan you actually buy. Which means the technology name on the advert tells you less than the number next to it, and the number next to it tells you less than the guaranteed minimum underneath.

Start with the number nobody expects. Cable came out fastest on download, at a median of 270.6 Mbit/s against full fibre's 149.2 Mbit/s, because the figures reflect the packages people had bought rather than what each technology can do. Read that once and it looks like proof that coaxial cable beats glass, which would be an extraordinary finding. It is not what it means. Anyone quoting that figure as proof that cable is faster than full fibre has misread it. Ofcom measured the packages its panel had bought, and in 2023 Virgin Media was selling gigabit packages to far more people than the full fibre networks were. The gap was 121.4 Mbit/s, and it measures a market rather than a technology.

The same trap sits under every headline figure on this page. Providers now report an average maximum download of 285 Mbit/s across the UK, and that sounds like the speed you would get. That is a line capability reported by the providers rather than anything measured, so it is not the number a speed test gives you. It is the line's capability, added up and divided, and the number on your own speed test depends on the package you bought, the wiring in your house, your Wi-Fi and the time of day.

On upload the technologies genuinely do separate, and it is the one place where full fibre wins on every package rather than on the package you happened to buy. Most full fibre is symmetrical or close to it. Virgin Media publishes 20 Mbps on M125 rising to 200 Mbps on Gig2, though you have to dig for it: Virgin Media's main speeds page publishes download figures for all six tiers and no upload figure for any of them. Its own explanation is about demand rather than engineering: “Most internet plans have higher download speeds than upload speeds because people usually download more than they upload.” On FTTC, Openreach publishes up to 76Mbps and upload speeds of up to 15Mbps and Ofcom measured 17.5 to 18.5 Mbit/s on the top tier. Across all technologies the median upload Ofcom measured was 18.4 Mbit/s.

Where the wireless numbers sit is harder to say, because the three companies selling it are publishing different things behind similar figures. Three's 150Mbps is a median at the worst hour of the day, in its own words: “at least 50% of 5G Broadband customers within 5G coverage areas, between the peak hours of 8pm and 10pm”. Vodafone publishes its 150 Mbps as an up to figure, which is a maximum rather than a median. EE publishes no basis for its headline speed. It does not say whether the figure is an average, a median or a maximum, and it does not say at what time of day it applies. A median at peak and an up to maximum are not the same claim, and Three's is by some distance the more useful disclosure even though the number looks identical.

Our guide to upload against download speed covers which one you actually notice, and what speed do I need works back from how your household uses it.

Latency and peak time, where they really differ

Direct answer: Full fibre was the lowest, then FTTC, then cable, with old copper worst by a distance. The spread between the best full fibre and the best cable was 7.6 ms, and small as that sounds it is the difference you feel rather than the one you measure.

A few thousandths of a second sounds like nothing, and for a film or a web page it is nothing. It is the difference you feel on a video call, in a game, and on anything where your computer has to ask a question and wait for the answer before it can ask the next one. Download speed decides how long a big file takes. Latency decides how a video call feels, whether a game is playable, and how quickly a page starts to appear rather than how quickly it finishes. It is the number the adverts never print and the one most worth asking about.

The peak-time column is the other half of the story and it is where cable shows its one structural weakness. At the busiest hours full fibre held 97.6% to 99.3% of its maximum speed and cable held 94.2% to 97.9%. Ofcom's own explanation is worth reading in full: “The variation in speeds at peak-time (8-10pm) is often higher in cable connections due to network congestion occurring nearer to the customer, making it harder for cable operators to add the additional network capacity required to reduce the effects of congestion.” In plain terms, on cable your street shares capacity in a way that full fibre customers do not, so the evening is when you notice it.

Wireless sits in a different category again. Ofcom measured 15 to 21 ms on 5G and 18 to 23 ms on 4G, which is respectable and a long way behind full fibre. But that figure moves with how busy the mast is, how far you are from it and what the weather is doing, and none of the three companies selling 4G or 5G home broadband publishes a latency figure at all, so there is nothing to hold them to.

Our guide to latency, jitter and packet loss explains the three numbers behind this, why speeds drop in the evening covers peak time in detail, and broadband for gaming is where latency matters most.

What you are owed when it goes wrong

Direct answer: Buy a fixed line from a provider that has signed both of Ofcom's schemes and you get three things: a minimum speed in writing before you buy, a way out without penalty if it is missed, and money back automatically when the line goes down. Buy 4G or 5G home broadband and you get none of the three.

Ofcom runs two voluntary schemes and most people have never heard of either. The first is the residential broadband speed code. Its signatories are BT, EE, NOW Broadband, Plusnet, Sky, TalkTalk, Utility Warehouse, Virgin Media and Zen Internet, and signing means three things. You get a minimum guaranteed download speed at the point of sale, specific to your address rather than to the package. On cable and full fibre, Ofcom's rule is that the figure must be at least 50% of the advertised speed for the package that is being enquired about. On technologies where distance matters, which means FTTC and old copper, it is set at the 10th percentile of the ISP's similar customers, adjusted to account for the impact of congestion at peak time. And then the part that gives that figure teeth. If your speed drops below the guaranteed minimum on a daily basis for at least three successive days after you report it, and the provider has not fixed it within 30 calendar days, you can leave without penalty.

If your broadband came in a bundle you get the choice: leave the whole bundle, keep the whole bundle, or leave only the parts that depend on the broadband connection. That matters more than it sounds, because a slow line is often sold with TV and a mobile SIM attached, and without that rule a speed problem on the broadband would leave you locked into the rest.

The second scheme is automatic compensation, and the word automatic is the whole point. You do not have to claim. The payment appears as a credit on your bill within 30 days, and it is paid whether or not you asked for it. The published rates are £10.34 for each day a total loss of service is not repaired, £32.31 for a missed engineer appointment and £6.46 for each day a new service starts late. Its members are BT, EE, Hyperoptic, Plusnet, Sky including NOW Broadband, TalkTalk, Utility Warehouse, Virgin Media, Vodafone and Zen Internet.

Now the finding this page exists for. Ofcom's speed code sets out its own scope, and it is worth quoting exactly because almost nobody has read it: “This covers broadband using Asymmetric Digital Subscriber Line (ADSL), fibre-to-the-cabinet (FTTC) broadband services, G.fast, fibre-to-the-premises (FTTP) broadband services, cable services, and fixed broadband delivered via fixed wireless technology, including satellite services.” Ofcom's speed code covers fixed wireless by name, so the gap is not technical. It is that the companies selling 4G and 5G home broadband have not applied it to those products, and the automatic compensation scheme's own scope is limited to fixed products. Three has signed neither Ofcom's residential broadband speed code nor its automatic compensation scheme. It does publish a 30-day money-back guarantee of its own, which is a commercial promise rather than a right, and it is not the same thing at all.

The automatic compensation scheme is a scope question rather than a signature one. Ofcom's own wording is “This scheme covers all residential fixed broadband and landline products.” EE and Vodafone are both members and both sell 4G or 5G home broadband, and the scheme's own words limit it to fixed products, so the wireless product falls outside it even at a member. None of Three, EE or Vodafone publishes a minimum guaranteed download speed for its 4G or 5G home broadband, so there is no figure for a speed complaint to be measured against.

Our guide to the minimum guaranteed download speed explains how yours is worked out, up to speed against guaranteed minimum takes the two numbers apart, can poor speeds let you leave early covers the exit right step by step, and the automatic compensation rates and compensation and service failure cover what you are owed.

4G and 5G home broadband, fairly

Direct answer: 4G and 5G home broadband is a genuinely good answer in two situations: an address a fixed line has not reached, and a home you are only in for a few months. It installs in minutes, it moves with you, and at £18 it is cheaper than most fixed lines.

Fixed wireless access means a router in your window that picks up the same 4G or 5G signal a mobile phone uses, so no cable is laid to your home at all. It reaches 96% of UK premises, which is more than full fibre and more than cable, and it is the only one of the four that needs no engineer, no appointment and no hole in the wall. For a rented flat, a house move in six months, a caravan, a building site or an address the fixed networks have simply not reached, it is the right answer and this page will not pretend otherwise.

Read the two speed rows together, because that is where the adverts stop being comparable. All three headline figures land near 150 Mbps and only Three says what its figure means: “at least 50% of 5G Broadband customers within 5G coverage areas, between the peak hours of 8pm and 10pm”. Vodafone publishes its 150 Mbps as an up to figure, which is a maximum rather than a median. EE publishes no basis for its headline speed. It does not say whether the figure is an average, a median or a maximum, and it does not say at what time of day it applies. A median at the worst hour of the day is a far stronger promise than a maximum, so on published evidence Three's number is the one to trust and it is the one that looks least impressive in a list.

EE is the only one of the three that tells you in advance what happens when the network is busy: past 1000GB in a month your traffic can be slowed at peak times. That is a limit rather than a benefit, and publishing it is still better practice than the silence from the other two. On what happens when the mast is busy and you are not over any threshold, none of the three publishes anything at all.

Two practical points before you buy one. Do not assume the switching service moves you off mobile broadband the way it moves you between fixed lines. Three directs leavers to One Touch Switch and also asks for 30 days' notice, so give notice as well as starting the switch. Then the silence that matters most. None of the three publishes a latency figure, a guaranteed minimum speed, or anything at all about telecare and personal alarms. If anyone in the house relies on a personal alarm, a lift line or a telecare device, that silence is the reason to choose a fixed line if one is available to you.

See what is on sale today in 5G home broadband deals and 4G home broadband deals, compare the two in 4G against 5G at home, and if you are waiting for an installation, mobile broadband as a temporary bridge covers the short-stay case.

The copper switch-off and what it means for FTTC

Direct answer: The old copper telephone network is being switched off, so landline calls move to a digital service that runs over your broadband connection. 31 January 2027 is the date BT has given for retiring the old network.

Start with the part that stops people worrying, because it gets lost in most of the coverage. Stop sell does not cut anyone off. It means that at those exchanges nobody can buy a new copper line or move to a different copper package, so if you change provider or change package you will be moved to full fibre. It is already running at 1,281 exchanges covering about 12.5 million premises, which Openreach describes as 51 per cent of Openreach's total full fibre footprint.

The second thing to say is the part Openreach has not made clear. Openreach's own pages are not consistent about when FTTC broadband itself stops. One says the copper network closes in January 2027 and another runs superfast broadband on to the exchange closures at the end of 2030. Your landline has a published date. The broadband that shares the same copper pair does not, and the two Openreach pages that address it do not agree with each other. Until that is settled, one thing is worth saying plainly. Buying a new FTTC line in 2026 is buying into a technology with a published end date and no published replacement date for the broadband half of it.

For scale: Openreach can now reach 23.4 million premises with full fibre, 9.4 million are actually connected, and it is building towards 25 million by December 2026. So the network to move people onto largely exists, and the gap between what is built and what is connected is the whole switch-off programme in one pair of numbers.

One correction to a common belief: this is not only an Openreach event. Virgin Media O2 also plan to retire their old landline networks, but dates for switching those have not yet been set. So a Virgin Media customer is not exempt, only further back in the queue, and no date has been published to plan around.

Check your own address with our landline switch-off checker, read what the copper switch-off actually changes and what happens to your landline, and if anybody in the house uses an alarm, will my care alarm still work is the one to read first. What changes at home covers the move from FTTC to full fibre.

Why the names are such a mess

Direct answer: There is one rule that cuts through it: ask where the fibre stops. At the cabinet in your street is FTTC, at a box on your wall is full fibre, and at a street cabinet with coaxial the rest of the way is Virgin Media cable.

Part fibre, fibre broadband, superfast and even fibre to the cabinet are all used for a line that ends in copper, and Virgin Media uses FTTC for its coaxial network, which is a different technology with the same acronym. Advertising rules have allowed fibre in the name of a part-copper product for years, so the word on the advert is not a guide to the technology behind it. So a package called Fibre Broadband at 67 Mbps and a package called Full Fibre at 900 Mbps are different technologies wearing overlapping names, and the only reliable way to tell them apart is the question above.

Virgin Media's naming deserves its own paragraph, because it is the one that catches people who already know what FTTC means. Virgin Media calls its own coaxial network fibre to the cabinet, or FTTC, on its own help page. That is the same acronym Openreach uses for an entirely different technology running over copper telephone pairs. Two different things, one label. Its own description of what it runs is clear enough: “So, most of your broadband runs over a fibre link, while our coax cables finish the job.” That is a hybrid fibre coaxial network, it is genuinely capable of gigabit speeds, and calling it FTTC in its own help pages helps nobody.

There is a second wrinkle at Virgin Media that matters if you are comparing upload speeds. Virgin Media sells over two different networks and tells you which only on its fastest tier, where it names nexfibre. On every other tier it does not say whether you are getting cable or fibre. So two Virgin Media customers on different tiers can be on entirely different technologies, and only one of them is told.

On FTTC, the honest description is Openreach's own. “Generally, the closer you are to your street cabinet the better. Speeds can start to drop noticeably beyond around 300 to 400 metres from the cabinet.” Because some of the journey still uses copper wire, your speed can vary slightly depending on the distance between your home and the cabinet. That is why an FTTC quote is personal to your address in a way a full fibre quote is not, and why Ofcom's code sets the minimum for it at the 10th percentile of the ISP's similar customers, adjusted to account for the impact of congestion at peak time rather than at a flat share of the advertised speed. On the folklore that rain and damp slow an FTTC line, there is less published than you would think. Openreach says copper is more susceptible to weather-related disruption than fibre, but neither it nor Ofcom publishes any figure for how much weather, damp or water in a duct actually affects an FTTC line.

Our full fibre against standard broadband guide takes the naming apart in more detail, and is full fibre worth it answers the question the names are usually obscuring.

Which one suits which household

Direct answer: The right technology depends less on how fast you want to go and more on what you do with it, and on what happens when it stops. Find your household below.

If your address has none of the fixed options, our rural switching guide covers what to do next, and the full comparison is still the fastest way to find out which of the four can actually reach you.

Live deals across all four technologies at your postcode