Every gaming networking guide, including ours, eventually gives the same advice: plug in a cable. It's correct. A wired link removes an entire category of problems — interference, contention, roaming, radio retries — and it costs about £8.
The trouble is that a lot of people genuinely cannot do it. The router is in the hallway because that's where the master socket is, the console is two floors up, and there's a solid wall, a rented flat with a no-drilling clause, or a partner who is not having a cable stapled along the skirting board.
So the real question isn't "wired or wireless". It's: when a straight Ethernet run is off the table, which alternative actually behaves like a cable, and which only looks like one on the box?
This guide covers the three serious options — MoCA, powerline, and mesh with a proper backhaul — plus the one thing to try first. Throughout, we judge them on the metric that matters for gaming, which is not the one printed on the packaging.
Table of Contents
- Bandwidth Is Not the Problem. Jitter Is.
- Try This First: Run the Cable Anyway
- MoCA: Ethernet Over Your Coax
- Powerline: The One Everyone Buys First
- Mesh With a Proper Backhaul
- How to Test Whichever You Choose
- The Short Version
- FAQ
Bandwidth Is Not the Problem. Jitter Is.
Powerline and MoCA kits are sold on headline throughput — "AV2000", "2.5 Gbps", "Gigabit". For gaming, that number is close to irrelevant.
An online match sends tiny packets, very frequently. A competitive shooter might use under 1 Mbps in each direction. You do not need 600 Mbps to play Apex Legends; you need every one of those small packets to arrive at a consistent interval.
That consistency is jitter. A link averaging 12ms but swinging between 6ms and 45ms feels dramatically worse than one sitting steadily at 20ms, because client-side prediction can compensate for a stable delay and cannot compensate for an unpredictable one. It's the same mechanism behind bufferbloat, where a download drives your ping to 500ms without your bandwidth ever running out.
So the questions below are: how much fixed latency does the link add (Ethernet adds a fraction of a millisecond), how stable is that figure under load, and how predictable is it in your house — the part no review can answer for you.
Try This First: Run the Cable Anyway
Before spending £60 on adapters, it's worth a serious look at whether a cable is genuinely impossible or merely inconvenient. A surprising number of "impossible" runs aren't.
- Flat Ethernet cable is around 1mm thick and goes under carpet edges, under laminate thresholds and beneath door frames without a visible bulge.
- Cat 5e is fine. You do not need Cat 8. Cat 5e carries a full gigabit over 100 metres, and spending more on cable category buys you nothing for gaming — see the USB Ethernet adapter guide.
- Self-adhesive trunking, painted to match the skirting, is close to invisible for a few pounds.
- Existing holes. If there's already a route through the floor for pipework or an aerial downlead, there's usually a route for Ethernet.
If, having actually looked, the answer is still no — the options below are worth having.
MoCA: Ethernet Over Your Coax
MoCA (Multimedia over Coax Alliance) turns existing coaxial television cabling into an Ethernet backbone. You put an adapter at each end of a coax run, and each gives you an RJ45 port.
The performance is genuinely good. MoCA 2.5 delivers actual data rates up to 2.5 Gbit/s and — far more importantly — is specified for low, tightly bounded latency. Published figures put average link delay at around 3.5ms, with manufacturers quoting roughly 3ms best case and 6ms worst. It stays backward compatible with MoCA 2.0 and 1.1, so mixed kit works at the lower standard's speed.
That 3.5ms is the honest caveat: it is not free the way Ethernet effectively is, and adding 3–4ms to a 15ms ping is a 25% increase. But it's far better than a marginal Wi-Fi link, and crucially it is consistent — a fixed toll rather than a random one, which is exactly the trade you want.
The UK problem with MoCA
Here's the part US-written guides won't tell you: MoCA assumes your house has coax running to multiple rooms, the norm in North America because cable television was wired that way. Most UK housing stock is not. The typical British home has a single aerial downlead to the lounge, possibly with a satellite downlead beside it, and no coax at all in the bedroom where the console lives — nothing for MoCA to run over.
Check before you buy. Look for coax wall plates in both the room with the router and the room with the console. If you have them, and they're on the same cable run or splitter tree, MoCA is comfortably the best option on this page. If you don't, it isn't an option at all — skip to powerline.
One caution: if you're a Virgin Media customer, your incoming service also arrives over coax. Fitting MoCA onto the same cabling as a live broadband service needs a proper point-of-entry filter, and is a job worth doing carefully rather than enthusiastically.
Powerline: The One Everyone Buys First
Powerline adapters send data over your mains wiring. They're cheap, they're everywhere, and they are the option most people reach for — partly because unlike MoCA, every home definitely has electrical wiring.
They also have the widest gap between best case and worst case of anything here.
G.hn has replaced HomePlug
There are two competing powerline standards, and the market has moved decisively.
- HomePlug AV2 is the older one, with a theoretical ceiling around 1200 Mbps (kits marketed as "AV2000" count both directions). Real-world throughput typically lands in the 300–500 Mbps range. The HomePlug Alliance has since disbanded, so HomePlug kit is now built on an unmaintained standard.
- G.hn, in Wave 2 form, uses OFDM with MIMO and reaches higher theoretical figures, with real-world speeds nearer 400–600 Mbps on clean wiring. Its more modern LDPC forward error correction helps it hold a stable link in electrically noisy conditions — and stability, not the headline number, is what you're buying.
The two are not interoperable. A G.hn adapter will not talk to a HomePlug AV2 adapter. Buy a matched pair, and in 2026 there's little reason for that pair not to be G.hn.
What actually kills powerline performance
Powerline is the only technology here whose performance depends on something you can't see. The things that matter:
- Both adapters must be on the same consumer unit, and ideally the same ring main. Cross a distribution board and throughput can collapse.
- Plug directly into a wall socket. Never into an extension lead or surge-protected strip — surge protection filters out exactly the high-frequency signal powerline uses.
- Noisy loads on the same circuit — cheap chargers, LED dimmers, washing machines, older fridge compressors — inject interference. A kit that tests perfectly at 2pm can degrade badly when the tumble dryer runs.
- Old wiring. Pre-1980s installations, or anything with a lot of junction boxes, tend to perform poorly and unpredictably.
Realistically, powerline adds in the region of 2–8ms in a good installation: a large improvement over weak Wi-Fi, and a clear step down from Ethernet or MoCA. In a bad installation it will be worse than the Wi-Fi you replaced. Which is why you must test it — and why you should buy from somewhere with a straightforward returns policy.
Mesh With a Proper Backhaul
The third route is to keep it wireless but fix the right part of the problem.
A mesh system's weak point is the backhaul — the link between the satellite node and the main router. If the satellite relays your traffic over the same radio it uses to talk to your console, you pay the airtime cost twice, and both latency and jitter suffer.
Three variants, in descending order of usefulness:
- Wired backhaul — the satellite connects to the router by Ethernet. Excellent, and obviously requires the cable you don't have. Worth mentioning because a short run to a hallway node is often possible when a long run to a bedroom isn't.
- Dedicated backhaul band — a tri-band mesh reserves one radio purely for node-to-node traffic. This is the meaningful upgrade for gaming, and the reason a tri-band kit outperforms a dual-band one by more than the price difference suggests.
- Wi-Fi 7 with MLO — multi-link operation lets a device use several bands at once, improving reliability rather than raw speed. Genuinely useful for jitter, though as we covered in Wi-Fi 7: hype vs reality, the gap between the marketing figures and the practical benefit is wide.
For a console that can't be cabled, a tri-band node in clear line of sight, with the console on 5GHz, will often beat a poor powerline installation. It won't beat a good one.
How to Test Whichever You Choose
This is the step almost everyone skips, and the one that tells you whether you've solved the problem or just moved it.
Do not use a speed test. It measures throughput at a moment of your choosing; you need consistency under load, over time.
The test that matters
From a PC on the same link as the console, run a long continuous ping to your own router — not a game server, because you're testing the link in your house, not the internet:
ping -t 192.168.1.1
(Substitute your router's address. On macOS or Linux, drop the -t.)
Leave it running for at least ten minutes and watch the spread, not the average. You want a tight cluster: every reply within a couple of milliseconds of the last. Occasional single spikes are normal; what condemns a link is a persistent spread — 3ms, then 40ms, then 7ms, then 60ms.
Then repeat it under realistic conditions:
- While someone streams 4K video on the same connection.
- While a game is downloading, which is when bufferbloat shows up.
- In the evening, with the household's usual appliances running — this matters enormously for powerline.
A kit that passes all three is a keeper. One that only passes the first is a kit to return.
Crucially, run the same test over your existing Wi-Fi before you install anything, and write the numbers down. Without a baseline you cannot tell whether the adapters improved anything, and a placebo upgrade is a common outcome. If neither link looks stable, the fault may be upstream of all this — our internet troubleshooting checklist is the better place to start.
The Short Version
- Coax in both rooms → MoCA 2.5. The best performance short of Ethernet, at a predictable ~3.5ms cost. Rare in UK homes, excellent when available.
- No coax, reasonably modern wiring on one consumer unit → a G.hn powerline pair, bought somewhere you can return it.
- Old or noisy wiring, or a rented flat you can't audit → a tri-band mesh node with a dedicated backhaul band, in line of sight of the console.
- A short cable run turns out to be possible → do that. It beats everything above, costs less, and never degrades.
None of these three matches a cable. They're compromises, and the right one depends on the physical facts of your home rather than on which has the best reviews. What makes the difference is running the ten-minute ping test afterwards, so you know which compromise you actually made.
If the router itself is the weak link rather than the path to it, our gaming router guide covers what's worth paying for — particularly SQM support, which fixes a problem no adapter here can touch.
FAQ
Does a powerline adapter add lag? Yes — typically a few milliseconds on a good installation, more on a poor one. The added delay is less important than its consistency. A powerline link that adds a steady 5ms is fine for gaming; one that swings between 3ms and 50ms is not, even if its average looks acceptable.
Is MoCA better than powerline for gaming? Where you have usable coax, yes, comfortably. MoCA runs over shielded cable designed for high-frequency signals, so its latency is tightly bounded and predictable — around 3.5ms — whereas powerline runs over unshielded mains wiring never designed to carry data, and its performance depends on your electrical environment. The catch is that many UK homes lack coax in the rooms you'd need it.
Will an "AV2000" powerline kit be faster than an "AV1000" one for gaming? Almost certainly not in any way you'll notice. The number refers to theoretical combined throughput, and gaming uses a tiny fraction of even the lower figure. Latency behaviour on the same wiring is broadly similar across tiers — you're paying for bandwidth you won't use.
Do powerline adapters work between floors? Often, yes — vertical distance matters much less than which circuit each socket is on. An upstairs socket on the same ring main as the router will usually work better than a downstairs socket on a different one.