Ping
Ping (round-trip time)
The round-trip time in milliseconds between your device and a server — the gaming metric that matters most.
Ping measures the round-trip time in milliseconds (ms) for a small data packet to travel from your device to a server and back. It is the primary metric for real-time applications: the lower the ping, the more responsive gaming, video calls, and VoIP feel.
How ping is measured
The classic ping command sends ICMP echo request packets to a target host and records how long each takes to return as an ICMP echo reply. This is the simplest form of latency measurement and is universally supported, but some servers block ICMP for security reasons.
TCP ping measures the time to complete a TCP handshake (SYN → SYN-ACK) to a specific port, useful when ICMP is blocked. HTTP ping measures the time to receive the first byte of an HTTP response — this includes DNS resolution time plus TCP handshake plus server processing, so it is higher than ICMP ping but more representative of real web browsing latency. Speed test tools typically use HTTP or WebSocket-based measurements to the test server.
Ping vs latency vs RTT
These three terms describe overlapping concepts that are often used interchangeably but have precise meanings. Latency is the general term for any delay in a network — it can refer to one-way delay or round-trip delay depending on context. RTT (round-trip time) is the formal networking measurement of the full round-trip delay. Ping is both the name of the ICMP measurement tool and a colloquial term for the RTT result it produces. On a speed test, the number labelled "ping" is the RTT to the selected test server — not a one-way latency figure.
What affects ping
- Physical distance — light and electrical signals travel at finite speed; a server 3,000 km away has a minimum propagation delay of about 15 ms one-way (30 ms RTT), regardless of connection type
- Routing hops — each router hop adds 0.5–5 ms of processing delay; poorly optimised routing paths with unnecessary hops add latency independent of distance
- Server load — a busy server takes longer to process and respond to ping packets; game servers under heavy load show higher and more variable ping than lightly loaded ones
- ISP congestion — cable nodes shared with many neighbours show elevated latency during peak hours as the shared upstream becomes saturated
- Wi-Fi overhead — the 802.11 contention-based access mechanism adds 2–20 ms vs wired Ethernet; Wi-Fi latency also varies with interference level
Good ping benchmarks by use case
| Use case | Excellent | Acceptable | Problematic |
|---|---|---|---|
| Competitive FPS gaming | < 20 ms | 20–50 ms | > 50 ms |
| Casual / RPG gaming | < 40 ms | 40–80 ms | > 100 ms |
| Video calls (Zoom, Teams) | < 40 ms | 40–100 ms | > 150 ms |
| VoIP (voice calls) | < 30 ms | 30–80 ms | > 150 ms |
| General web browsing | < 50 ms | 50–150 ms | > 300 ms |
| Video streaming | Any | Any (< 500 ms) | Rarely matters |
How bufferbloat inflates ping under load
A key distinction is idle ping vs loaded ping. Your idle ping — measured when the connection is quiet — reflects the true propagation and routing delay. Your loaded ping — measured while a download or upload is saturating the link — reflects idle ping plus the queuing delay introduced by your router's oversized buffers. On a router with bufferbloat, idle ping might be 8 ms while loaded ping spikes to 200–400 ms. This is why gaming feels smooth at 3 AM but unplayable at 8 PM when family members are downloading. Use the Waveform Bufferbloat Test to measure both values and see the gap.
Testing ping effectively
- Command-line ping —
ping google.comorping 8.8.8.8gives continuous RTT samples; run for 30+ seconds to see variance - MTR / WinMTR — combines ping and traceroute; shows per-hop latency and packet loss, useful for diagnosing where in the path latency is introduced
- Speed test ping — measures RTT to the speed test server; representative of your ISP connection quality but not your in-game ping, which depends on server location
- In-game ping vs speed test ping — a speed test server is usually nearby (same city or region); game servers may be hundreds or thousands of kilometres away. A 10 ms speed test ping does not mean 10 ms in-game — always check in-game latency directly
Frequently Asked Questions
Does ping affect streaming?
Streaming buffers several seconds of video ahead, so ping under 200 ms has virtually no effect on Netflix or YouTube quality. Ping matters for live, interactive traffic — gaming, video calls, financial trading.
Why is my ping high only in the evening?
Evening congestion on cable ISPs. Cable plants share upstream capacity across neighbours — when everyone is online between 7–10 PM, the shared node gets congested and latency spikes. This is a structural cable limitation, not a fault.
How do I lower my ping for gaming?
Connect via Ethernet, choose game servers in your region, close background applications that use bandwidth, and consider switching to fiber — which consistently delivers 3–10 ms ping vs 10–25 ms on cable.