Ping (latency)
The time, in milliseconds, a packet takes to reach the server and come back. The lower it is, the more "responsive" the connection feels.
- < 30 ms Great for gaming and video calls
- 30–80 ms Good for everyday use
- > 100 ms Noticeable lag
Free test · No sign‑up · Privacy first
Measure ping, jitter, download and upload right in your browser: nothing to install, no sign-up, under a minute. Then learn how to read the numbers and what you can improve.
A speed test measures the quality of your Internet connection in real time by opening several connections to neutral servers (here we use Cloudflare's global infrastructure) and timing how fast data travels. RealMbps does it entirely in your browser, with nothing to install and without storing your results.
The test returns four numbers. Ping (latency) is the delay, in milliseconds, for a packet to reach the server and come back: it matters for gaming and video calls. Jitter is how much ping varies over time: the lower it is, the more stable the connection. Download is how fast you receive data (streaming, browsing, updates). Upload is how fast you send data (video calls, cloud backups, posting photos and videos).
To read the results well, compare them with your plan's speed, remembering that 1 byte = 8 bits: a "1 Gig" fibre line (1000 Mbps) downloads at most around 125 MB/s. Differences of 10-30% from the contracted value are normal due to Wi-Fi, router and device. In the sections below you'll find how to interpret each value and how to improve your connection.
Full-fibre (FTTP) is the gold standard for speed and stability. In the UK, BT, Virgin Media, Sky and TalkTalk sell 100 Mbps to 1 Gbps plans, while much of the US runs on cable (Xfinity, Spectrum) or fibre (AT&T Fiber, Verizon Fios) with fast download but slower upload. Where wired broadband isn't available, 5G home internet is the usual fallback — cheaper to deploy but higher latency. RealMbps shows what you actually get, beyond the advertised number.
Compare the result with your plan, keeping in mind that Wi-Fi, the ISP router and time of day all matter: a 10-30% gap from the theoretical peak is normal. If you're far below what you pay for, you can escalate — to Ofcom or ADR in the UK, or check the FCC Broadband Nutrition Label in the US. Re-test at different times and over Ethernet to tell whether your Wi-Fi or the line is the bottleneck.
The fastest connections in the community. Measure, register, climb the ranks.
No results for this country yet — be the first!
Average download speed of published tests, by country. A community-built ranking, not an official statistic.
Take on the leaderboard: who has the fastest line? →
Results submitted by users, not verified. Just for fun.
You will appear on the wall with your nickname and country.
Four numbers tell the story of your network. Here is what they really mean.
The time, in milliseconds, a packet takes to reach the server and come back. The lower it is, the more "responsive" the connection feels.
How much ping varies from one packet to the next. High jitter means choppy audio/video, even if average ping looks fine.
How fast you receive data: streaming, browsing, updates. The most "spectacular" number, but on its own it isn’t enough.
How fast you send data: video calls, cloud backup, online gaming, live streaming. Often overlooked, but crucial for remote work.
Beyond the speed test: find your IP, measure ping and manage DNS.
Small changes that often make a big difference — before calling your provider.
An Ethernet cable removes interference, packet loss and Wi‑Fi swings. For desktop PCs, consoles and smart TVs it is almost always the best choice.
Centre of the house, up high, away from metal, microwaves and aquariums. Thick walls and mirrors are sworn enemies of Wi‑Fi.
2.4 GHz reaches further, but 5 GHz is much faster and less congested. For devices close to the router, always pick it.
If your router is more than 4–5 years old, it is probably the bottleneck. Look for Wi‑Fi 6 / 6E models and keep firmware up to date.
Mesh systems spread the signal intelligently without halving bandwidth like old range extenders.
Close apps using the network in background, pause cloud sync during calls and check no device is downloading updates.
Try public DNS like 1.1.1.1 (Cloudflare) or 8.8.8.8 (Google). They do not raise download speed, but make browsing snappier.
Power the modem off for 30 seconds before turning it back on. It sounds trivial, but it frees memory, sessions and routing tables built up over days.
If real numbers are far from your contract, run more tests at different times (cable + Wi‑Fi). With data in hand, file a complaint or switch plans.
Your provider measures speed to the nearest node under ideal conditions. RealMbps measures what actually reaches your device, through Wi‑Fi, router, OS and browser. Differences of 10–30% are normal.
Connections are measured in Megabits per second. 1 byte = 8 bits, so 100 Mbps ≈ 12.5 MB/s. That is why downloads show "smaller" numbers.
Netflix 4K needs ~25 Mbps, 4K HDR ~50 Mbps. A remote‑working family with several HD TVs lives comfortably on 100 Mbps symmetrical. Beyond that, stability matters more than the headline number.
Almost certainly yes if your line is above 300–400 Mbps. Repeat the test on Ethernet near the router: if numbers go up, the issue is Wi‑Fi, not the connection.
No. The test runs entirely in your browser, against public Cloudflare servers. We do not store results, use tracking cookies or profile anything.
Download drives streaming and browsing, but upload carries video calls, cloud backups, online gaming and photo/video sharing. On asymmetric lines (e.g. FTTC 200/20) a low upload is often the real bottleneck when working from home.
Yes. The algorithm detects mobile networks and adapts chunk size to measure correctly on 4G/5G, where many speed tests fail. Keep in mind radio signal varies: repeat the test in different spots for a reliable average.
For a realistic picture, run 3-5 tests at different times: morning, afternoon and during the evening peak (9-11pm), when your ISP network is busiest. A single test says little; consistency over time reveals the true health of your line.
Practical articles to measure, understand and improve your connection.
Honest comparison of top models for fiber up to 1 Gbps, with selection criteria the marketing won't tell you.
Read → TutorialReliable procedure, common mistakes that ruin a test, what to do if speeds are below expectations.
Read → ReferenceReal Mbps for streaming, gaming, video calls, remote work. Summary table and family bandwidth math.
Read → UK ISPsReal speeds per UK broadband provider and how to escalate to Ofcom or ADR.
Read → US ISPsReal speeds per US ISP, the FCC Broadband Nutrition Label and how to file a complaint.
Read → BasicsBits, bytes and the factor-8 rule. Why your download is 8x slower than the Mbps in your contract.
Read → Optimization12 concrete fixes: Wi-Fi, router, DNS, cabling. What actually works and what is just a myth.
Read → CoverageWhole-home coverage with no dead zones: the best mesh kits for large homes and multiple floors.
Read → SignalWhen an extender is enough and which one to pick to extend your signal reliably.
Read → AlternativeInternet in every room with no rewiring, using your electrical wiring: recommended models.
Read → WiredCat 6/7/8, switches and adapters: stability and low latency for gaming and work, without overspending.
Read → ComfortChair, adjustable desk and accessories: a comfortable setup for working online for hours.
Read →