Ping, Latency and Packet Loss: What the Numbers Mean | Check Host

What is a good ping? Learn how to read ping results — round-trip time, jitter and packet loss — and what values are normal across the world.

Ping is the oldest network test there is: send a small packet, wait for the echo, measure how long it took. It answers two questions — is the host reachable, and how fast and stable is the path to it.

The numbers in a ping result

RTT (round-trip time) — milliseconds from sending the packet to receiving the reply. Ping results show min, average and max.
Jitter — how much the RTT varies between packets. The gap between min and max is a quick proxy.
Packet loss — the percentage of packets that never got a reply.

What is a good ping?

Distance sets a floor: light in fibre covers roughly 200 km per millisecond, and the signal has to go there and back. Rough guide values:

Path	Typical RTT
Same city	1–5 ms
Same continent	10–50 ms
Europe ↔ US East Coast	70–100 ms
Europe ↔ Asia / Australia	150–300 ms

So "is 120 ms good?" depends entirely on where you are measuring from. That is why the online ping test shows results per location on a map: a host can be fast in one region and slow in another for perfectly normal reasons.

Packet loss

Any sustained loss above about 1% is noticeable in calls and games; above 5% connections feel broken. When you see loss:

Loss from every location — congestion or a problem close to the target.
Loss from one region only — a congested link or bad peering on that route.
100% loss everywhere, but the website works — the host simply blocks ICMP. Use a TCP port check instead; it measures the connect time to a real service port.

Ping blocked? Use TCP ping

Many cloud providers and firewalls drop ICMP echo requests. A TCP ping measures the time to complete a TCP handshake on a port such as 443. It travels the same path as real traffic, so it is often the more honest measurement.

Find where latency is added

Ping tells you that a path is slow or lossy, not where. For that, run a traceroute — see how to read a traceroute for what to look for.