Starlink Ping & Latency

Ping and latency are two names for the same thing: the time it takes a small packet to travel from your device to a server and back, measured in milliseconds (ms). It is different from speed. Download and upload are about how much data moves per second; latency is about how quickly a single message makes the round trip. For streaming and large downloads, speed dominates; for video calls, gaming, and anything interactive, latency and its steadiness matter more. On Starlink, latency has its own character because every packet takes a trip up to a satellite and back down.

Why Starlink Latency Is Different

The number that defines satellite latency is orbital altitude. Traditional satellite internet uses geostationary satellites about 35,786 km above the equator. Even at the speed of light, the round trip up and back down to that altitude alone is close to half a second, which is why older satellite internet feels sluggish — real-world latency is often around 600 ms. Starlink instead uses low Earth orbit satellites only a few hundred kilometres up, so the physics of the trip is a small fraction of that. That is the whole reason Starlink can offer latency in the tens of milliseconds: Starlink has described Standard/Residential latency in roughly the 25–60 ms range, far closer to a wired connection than to legacy satellite. Your own figure still depends on location, obstructions, congestion, and load.

“People compare Starlink's ping to fiber and call it high. The fairer comparison is old satellite internet at ~600 ms. Going from that to tens of milliseconds is the difference between unusable and genuinely interactive.” — Jack Shi, SatSpeedCheck editor

How to Test Your Starlink Ping

There are three good ways to measure latency, and using more than one gives you a fuller picture:

• A speed test. Run our Starlink speed test, which reports latency next to download and upload. Note that a browser test measures HTTP request latency to a nearby server — it tracks real ping closely but is not an ICMP ping; see the methodology for exactly what it does.
• A command-line ping. Open a terminal and run a continuous ping to a reliable host, such as ping 1.1.1.1, for a minute. Watch the average, the highest spikes, and how many replies are missed — that is your real-world latency, jitter, and packet loss in one view.
• The Starlink app. It shows latency and a history of outages, which is useful for connecting a latency spike to a real obstruction event.

Whichever you use, test more than once and at more than one time of day. A single reading can catch a good moment or a bad one; the useful answer is the typical value and how far it swings.

Test Under Load, Not Just Idle

The most common way people misjudge their connection is testing latency only when nothing else is happening. Idle latency looks fine; then a big upload starts and the ping climbs sharply. That effect is called bufferbloat — packets queue up behind the heavy traffic and every round trip gets slower. It is why a video call can stutter the moment a cloud backup kicks in, even though your idle ping was healthy. To see it, start a large upload or download and run a ping at the same time; if latency jumps a lot under load, bufferbloat is the culprit, not the raw Starlink link.

Ping, Jitter, and Packet Loss

Latency is only one of three numbers that decide how a real-time connection feels:

• Ping (latency) — the average round-trip time. Lower is better.
• Jitter — how much the ping varies between samples. Steady beats fast-but-erratic for calls and games.
• Packet loss — the share of packets that never arrive; even a small amount hurts real-time use. See our Starlink packet loss guide for how to test and fix it.

What Your Ping Number Means

Ping has no single “good” value, because what matters depends on what you are doing. These bands are common rules of thumb for interactive use, not measurements or guarantees, and they apply to a typical value with few spikes:

For voice and video there is an actual yardstick behind those bands: the ITU-T G.114 recommendation advises keeping one-way delay under about 150 ms for most voice applications, and treats delays well beyond that as progressively harder on conversation. Ping is a round trip, so on a roughly symmetric path that guidance corresponds to a ping in the neighbourhood of 300 ms. Starlink's typical latency is far under that, which is why the average rarely ruins a call by itself.

Video Calls: Steadiness Beats the Average

Call apps smooth out uneven packet arrival with a small buffer, and a bigger buffer means more delay. So a connection with a low average but a lot of variation can feel worse on a call than a slightly higher but steady one. Three things cause most Starlink call trouble: jitter (see Starlink jitter explained), packet loss (see the packet loss guide), and bufferbloat when something else on the network is uploading. If you work from the dish, the Starlink for remote work guide covers the full call and VPN setup.

A practical check is to run a one-minute ping while doing something that looks like your real use: a test call, or a large upload in the background. If the maximum jumps or replies go missing only under load, fix the load (a router with smart queue management, or schedule the upload), not the dish.

How to Read a One-Minute Ping

Run ping -c 60 1.1.1.1 on macOS or Linux, or ping -n 60 1.1.1.1 on Windows, and read the summary at the end. The wording differs by system but the useful numbers are the same: the minimum, the average and the maximum, plus the percentage of packets lost. macOS prints min/avg/max/stddev, Linux prints min/avg/max/mdev, and Windows prints Minimum, Maximum and Average.

Here is how to interpret the common patterns. The figures are illustrative examples of the shape, not measurements from any dish:

Repeat the test wired and at a different time of day before drawing a conclusion; one minute is a snapshot, and the same dish can look different at noon and at 8 pm.

What Raises Starlink Ping — and How to Lower It

If your latency is higher or spikier than it should be, work through the usual causes in order. A partly blocked view of the sky is the biggest one — every obstruction can cause a brief spike or drop, so check with the obstruction checker and clear the dish's field of view. Evening congestion raises latency during the busy hours in populated cells. Bufferbloat under load is fixable with a router that supports smart queue management (SQM). And local Wi-Fi adds its own delay, so confirm the problem on a wired or close-range test first. For a broader tune-up, see how to increase Starlink speed and, if games are your priority, the Starlink for gaming guide. And if the whole connection feels slow rather than just high-ping, start with why is my Starlink so slow.