If router position changes SINR or RSRP
Keep the location that gives the most repeatable wired ping, jitter and signal values. Place the hub high, in the clear and near the strongest window-facing mast direction before buying an antenna.
This guide helps you test 5G broadband latency with practical benchmark ranges for ping, jitter, packet loss, loaded latency, signal quality and mast congestion.
5G latency test guide
Is your 5G home internet lagging during games or dropping frames on video calls? Download speed alone does not show responsiveness. Use the live browser test below to measure latency and jitter, then compare Wi-Fi, Ethernet, signal quality, loaded latency and peak-time mast performance.
Do not judge 5G by download speed alone. A 5G router can show a fast Mbps result while gaming still feels poor because the radio scheduler, mast load, upload queue or local Wi‑Fi hop is adding delay.
Live browser latency sample
This browser sampler sends repeated lightweight requests to LinkSpeed and measures round-trip response time. It is not an ICMP ping or a mobile-network lab test, but it can reveal unstable browser latency and jitter on the current connection.
Compare one Wi-Fi run with one Ethernet run to identify where the latency begins.
Recommended next step
Use the targeted troubleshooting route below.
Open 5G Latency ActionsQuick benchmark
Use these ranges as a practical guide. A stable result with low jitter and zero packet loss is more useful than one unusually low sample.
| Ping Range | Performance | Best Fit |
|---|---|---|
| Under 30 ms | Excellent Fast, responsive and suitable for latency-sensitive use. | Competitive gaming, cloud gaming and live interaction. |
| 31–50 ms | Good Typical of a healthy 5G home broadband setup. | Most online gaming, 4K streaming and video calls. |
| 51–80 ms | Fair Usable, but fast games may feel less immediate. | Browsing, streaming and casual gaming. |
| Over 80 ms | Poor for gaming Check jitter, packet loss, Wi-Fi, mast load and router placement. | Basic browsing unless the cause is corrected. |
Connection path
One local wired hop removes Wi-Fi retries and airtime contention from the diagnosis.
The extra wireless hop can add retries, interference and jitter before traffic even reaches the 5G radio link.
Result-led diagnosis
The symptom, likely network layer and first useful test are combined in one place so you can move directly from the result to evidence without repeated staging sections.
| What You Notice | Most Likely Layer | Best Next Check |
|---|---|---|
| Games rubber-band or calls break up on Wi-Fi, but Ethernet stays stable | Indoor Wi-Fi. Retries, interference, weak coverage or shared airtime are adding delay before traffic reaches the 5G radio link. | Keep Ethernet as the control and compare router-to-device Wi-Fi separately from the cellular connection. |
| Latency changes sharply when the router moves or turns | Signal penetration or interference. Building materials and router orientation are changing RSRP, RSRQ or SINR. | Try a higher window-facing position and record signal metrics with every latency sample. |
| Spikes line up with a band, cell or 5G-to-4G change | Radio-mode instability. The router is changing its serving band, cell or network mode. | Record the band, cell ID and network mode before and during the spike; only test supported preferences one at a time. |
| The same wired test becomes worse at similar evening times | Mast or backhaul load. Shared mobile-network capacity or routing is busier outside the home. | Compare identical daytime and evening Ethernet results with household traffic paused. |
| Idle ping is acceptable, but uploads or downloads create large spikes | Mobile-router queueing. Interactive packets are waiting behind larger transfers on a variable uplink. | Run a quiet-versus-loaded comparison, then follow the dedicated bufferbloat optimisation guide to resolve loaded-latency constraints. |
| Only one game, server region or app is affected | Destination or route. The local 5G path may be healthy while one external service or route is busy. | Compare another real-time service and a nearer server region before replacing the router or changing provider. |
5G radio evidence
Signal bars are too broad for fault finding. Record these values from the router dashboard or app beside each wired latency test.
| 5G Signal Indicator | Healthier Target | Degraded Boundary | Diagnostic Meaning |
|---|---|---|---|
| SINR | Around 20 dB or higher | Low single digits or negative values | Signal-to-interference-plus-noise ratio is the clearest placement clue. Low values suggest interference or a poor indoor radio path. |
| RSRP | Better than about -90 dBm | Near -110 dBm or worse | Reference signal received power shows how strongly the cellular reference signal reaches the router. |
| RSRQ | Closer to -10 dB | Near -15 dB or worse | Reference signal received quality is affected by interference and cell load, so it adds context that RSRP alone cannot provide. |
| Band / cell / mode | Stable during repeat tests | Frequent band, cell or 5G-to-4G changes | A change that occurs at the same time as a ping spike is stronger evidence of cellular instability than one low speed result. |
About n78: the 3.4–3.8 GHz capacity layer commonly associated with n78 can be fast, but the best band is the one that remains stable at your address. Do not force a band blindly; lower-frequency layers may penetrate the building more reliably.
Controlled comparison
Keep the same test device and change one variable at a time. The goal is to identify the first layer where the result changes.
Establish a quiet baseline. Run the live sample with downloads, uploads, streaming and cloud sync paused. Record average ping, jitter, packet loss and the current signal values.
Compare Wi-Fi with Ethernet. Repeat the same test over a direct cable. If only Wi-Fi is unstable, the cellular link is not the first place to make changes.
Move the 5G router and retest. Try a higher window-facing position and compare ping, jitter, RSRP, RSRQ and SINR rather than download speed alone.
Record band, cell and network-mode changes. Check whether a spike aligns with a band change, cell change or fallback from 5G to 4G.
Compare quiet and peak-time results. Repeat the same Ethernet test during the day and in the evening with household traffic paused.
Test under upload load. Repeat the sample while another device uploads a large file. A sharp rise from the quiet baseline points to queueing rather than radio strength alone.
Use the result
These actions stay focused on the cellular boundary. Deeper Wi-Fi, upload and queue-management instructions are linked to their dedicated guides rather than repeated here.
Keep the location that gives the most repeatable wired ping, jitter and signal values. Place the hub high, in the clear and near the strongest window-facing mast direction before buying an antenna.
Record the exact change and test any router-supported preference one at a time. Do not assume the fastest band is the most stable, and avoid unsupported locking that breaks carrier aggregation.
The 5G radio path is not the first suspect. Keep latency-sensitive devices wired where practical and use the slow Wi-Fi diagnostic guide for local spectrum, placement and shared-airtime checks.
Collect comparable daytime and peak-time results on at least two days, including ping, jitter, packet loss, signal metrics, band and cell ID. This is stronger provider evidence than one speed-test screenshot.
That pattern is mobile-router bufferbloat: the air link may be healthy while interactive packets wait in a filled queue. Follow the dedicated bufferbloat optimisation guide to resolve loaded-latency constraints.
Compare another mobile network or check whether a lower-variance fixed line is available. The 5G broadband versus full fibre guide explains the access-path trade-offs.
External antenna caution: only consider an antenna or outdoor hub when the exact router supports the connector, bands and MIMO arrangement. Prove that indoor placement is the limiting factor before buying hardware.
A stable result under about 40 ms is strong for gaming, 40 to 70 ms is generally usable, and repeated results above 80 ms or large jitter spikes can cause noticeable lag. Consistency and packet loss matter as much as the average.
5G traffic shares scheduled radio resources. Signal quality, retransmissions, carrier aggregation, band or cell changes, mast load and the local Wi-Fi hop can add more latency variation than a stable optical access path.
Start with Ethernet, then compare router position, SINR, RSRP, RSRQ, band or cell changes and quiet versus peak-time results. If delay appears only under load, use the dedicated bufferbloat guide rather than changing several router settings at once.
Yes. Large background transfers can fill a mobile router queue and push loaded latency far above idle ping. Compare quiet and busy tests, then follow the dedicated bufferbloat optimisation guide for configuration steps.