Distance table
At the same approximation, 500 km is about 4.9 ms RTT, 2,000 km about 19.6 ms RTT, and 5,000 km about 49.0 ms RTT.
Using approximately 204,000 km/s in fiber, 1,000 km of optical path adds about 4.9 ms one way and 9.8 ms round trip before routing or equipment overhead.
Editorial model reviewed · Internet Analysis Editorial Team
At the same approximation, 500 km is about 4.9 ms RTT, 2,000 km about 19.6 ms RTT, and 5,000 km about 49.0 ms RTT.
Straight-line city distance underestimates the physical cable path. Subsea landing points and terrestrial detours can materially increase it.
These are calculated propagation floors using 204,000 km/s, not observed internet measurements. One-way delay in milliseconds = route length in km ÷ 204. Round-trip delay is twice that value, assuming the same modeled length in both directions.
| Optical path length | One-way delay | Round-trip delay |
|---|---|---|
| 1 km | 0.005 ms | 0.010 ms |
| 100 km | 0.490 ms | 0.980 ms |
| 500 km | 2.451 ms | 4.902 ms |
| 1,000 km | 4.902 ms | 9.804 ms |
| 2,000 km | 9.804 ms | 19.608 ms |
| 5,000 km | 24.510 ms | 49.020 ms |
At this approximation, 1 km adds about 0.0049 ms one way, which is 4.9 microseconds. Confusing microseconds with milliseconds introduces a factor-of-1,000 error. These values describe optical path length, not a complete internet connection.
For an illustrative 1,000 km geographic separation, assuming a 1,200 km optical path gives 5.882 ms one way and 11.765 ms RTT. The 20% detour is a chosen scenario, not a measured route. A 20 ms entered RTT would leave 8.235 ms above that model; it does not identify which network component caused the difference.
Increasing Mbps does not change propagation speed in the same fiber. It can reduce serialization or queueing when a link is busy. If timing changes with comparable endpoints, investigate congestion, Wi-Fi and endpoint work rather than assuming the distance changed.
Calculate your own route and compare measured RTTSee New York–London scenarios. For unstable timing, follow jitter troubleshooting or check connection stability.
That is approximately the speed of light in vacuum. Light travels more slowly in glass because of its refractive index.
No. Higher bandwidth changes how fast bits can be serialized, not the propagation speed through the medium.
Using 204,000 km/s, 1 km adds approximately 0.0049 ms one way or 0.0098 ms round trip. Those are microsecond-scale propagation values, not complete network latency.
Values use transparent approximations described in the article.
Reference constants: NIST speed of light in vacuum. Fiber propagation uses a disclosed engineering approximation of 204,000 km/s; actual group velocity varies with the fiber.