Free tool
Latency checker: find your closest location.
Pick the city where most of your users are. We estimate the round-trip time to each of our 8 data centers and rank them for you.
From New York, your best gate is New York
- NYC New YorkUnited States · Best match ~2 ms 10 configs in stock · from $59/mo
- MIA MiamiUnited States ~25 ms 4 configs in stock · from $59/mo
- LAX Los AngelesUnited States ~53 ms 10 configs in stock · from $59/mo
- LON LondonUnited Kingdom ~74 ms 5 configs in stock · from $59/mo
- AMS AmsterdamNetherlands ~78 ms 10 configs in stock · from $59/mo
- FRA FrankfurtGermany ~83 ms 10 configs in stock · from $59/mo
- TYO TokyoJapan ~143 ms 3 configs in stock · from $115/mo
- SIN SingaporeAsia-Pacific ~201 ms 7 configs in stock · from $115/mo
New York is far from all 8 locations: the closest is about 2 ms away. That is fine for websites, APIs and backups; players of real-time games will notice it.
Estimated round-trip times, not measurements. See how they are calculated below.
Method
How the estimate is calculated
Light travels through optical fiber at about 200 km per millisecond, two thirds of its speed in a vacuum. Cables rarely follow the shortest path, so we multiply the great-circle distance between the two cities by 1.3, count the round trip, and add 2 ms for routers and switches along the way.
The result is close to what well-connected networks see. Your own numbers depend on your provider’s routes and peering, Wi-Fi and the last mile, so treat them as a guide, not a promise.
Estimated round trip
RTT ≈ 2 × distance × 1.3 / (200 km/ms) + 2 ms
- Frankfurt → Amsterdam
- ~7 ms
- New York → London
- ~74 ms
- Los Angeles → Tokyo
- ~117 ms
FAQ
About latency
All questionsHow accurate are these numbers?
They are estimates from geography, close to what well-connected networks see. Real latency depends on your provider’s routing, peering and last-mile connection, so a specific route can be faster or slower.
What latency is good enough?
As a rule of thumb, under 30 ms feels instant for games and interactive apps, under 80 ms is comfortable for most websites and APIs, and above 150 ms starts to show in real-time use. Pages that need many round trips suffer first.
Can I measure the real latency myself?
Yes, once a server is running: use mtr or ping from your users’ networks, and iperf3 for throughput. Our guide explains how to read the results and what to send us if a route looks wrong.
Found your gate?
Every location has servers ready to deploy, with the same price and DDoS protection everywhere.