ifconfigpublic IP, network & connection quality
Your public IP address

Connection quality — latency, jitter, loss & bufferbloat for gaming and real-time

Unloaded ping · ms
Download · Mbps
Upload · Mbps
Loaded ping ↓ · ms
Loaded ping ↑ · ms
Bufferbloat
Jitter · ms
Packet loss · %
Lag spikes · %
Latency p99 · ms

Game-server latency — round-trip from your browser · Steam prioritised · approximate

ServerPingJitterLoss
Run a test to measure latency to game networks…
Download · Mbps Upload · Mbps Ping · ms

Gaming & real-time readiness

Run a test to assess gaming quality…

Latency distribution · ms

StateMinAvgp50p95p99MaxJitterSpikesLossPings
Run a test to measure latency quality…

Also suitable for

Run a test to evaluate connection quality…
What do these metrics mean?

Unloaded ping — round-trip time to the server while the line is idle. Your baseline latency; lower feels snappier.

Loaded ping ↓ / ↑ — latency measured while the download / upload is saturating the link. The gap versus unloaded ping is what you feel as lag during a big transfer.

Bufferbloat — the extra latency added under load (worst of ↓/↑). Oversized buffers queue packets and make a "fast" line feel sluggish during downloads. Graded A+ <5 ms · A <30 · B <60 · C <200 · D <400 · F ≥400 ms added.

Jitter — how much ping varies from one sample to the next (mean consecutive difference). Low jitter keeps calls and games smooth. Good <10 ms · Fair <30 · Poor ≥30 ms.

Packet loss — share of pings that errored or timed out (approx. over HTTP; a healthy line reads 0%). Even 1–2% badly hurts calls and gaming. Good <0.5% · Fair <2% · Poor ≥2%.

Lag spikes — share of pings well above your own median (median + max(2.5×jitter, 15 ms)). These are the momentary jumps that cause rubber-banding, hit-registration errors and deaths in games. Good <1% · Fair <5% · Poor ≥5%.

Latency p95 / p99 — 95% / 99% of all pings were faster than this. p99 is the worst 1% — the tail that a simple average hides and that decides competitive play. p99 Good <70 ms · Fair <180 · Poor ≥180 ms.

Game-server latency — round-trip time from your browser to real game networks (Steam first), measured over HTTPS. Approximate — it reflects the nearest edge of each service, not a specific match server, and can be blocked by ad-blockers or corporate proxies (shown as unreachable). This is closer to in-game latency than the ping to our test server.

Gaming & real-time readiness — verdicts per genre from your game latency (Steam ping when available), jitter, packet loss, lag spikes, bufferbloat and, for cloud gaming, download speed. Competitive titles need the tightest latency and jitter; casual and turn-based are far more forgiving.

Line stability — how steady throughput stayed across the test (100% = perfectly flat); shown in the transfer summary. Low values mean a bursty or congested line.

Latency distribution — min / avg / p50 / p95 / p99 / max, plus jitter, spikes and loss, split across idle, download-loaded and upload-loaded phases.

Runs in your browser and uses real bandwidth. For CLI / pfsense / routers:
curl -s https://HOST/speedtest.sh | sh
by size
by time

Network intel & fingerprint — tap a card to expand

Location

Country
Region
City
Postal
Timezone

Network & WHOIS

ASN
ISP / Org
Reverse DNS
Network
Range / CIDR
Registry
Abuse contact
Edge (colo)

IP intelligence

Type
Hosting/VPN
Tor exit
Proxy hops

Browser fingerprint client-side, nothing stored

Visitor ID (SHA-256)computing…
Canvas
WebGL vendor
WebGL renderer (GPU)
Audio
Fonts detected
Screen
Signals used

Local network WebRTC

Detected IPsdetecting…
ICE candidates; browsers may return a *.local mDNS name.

Open ports expand to scan

Scans 1–1024 plus common service ports on first expand.
Best-effort TCP handshake from the Cloudflare edge; most residential IPs are firewalled/CGNAT, so nearly everything shows filtered.

Request headers

Loading…