Download Speed Estimator

Divide by eight first

Speeds are quoted in bits per second and file sizes in bytes. A 100 Mbps connection moves at most about 12.5 MB per second, so a 5 GB file takes roughly 400 seconds at the theoretical maximum. Almost every complaint that a download is slower than the advertised speed is this conversion rather than a fault.

Then subtract the overhead

Protocol headers, acknowledgements and retransmissions consume capacity, so effective throughput is typically 90 to 95 percent of the line rate on a wired connection and lower on Wi-Fi. Realistic planning uses about 11 MB/s on a 100 Mbps link.

The bottleneck is rarely your connection

Transfer speed is limited by the slowest link in the path: the server's upload capacity, an intermediate network, a congested peering point, or your own Wi-Fi. A gigabit line pulling from a server that caps each connection at 5 MB/s downloads at 5 MB/s. This is why a single slow download alongside a fast speed test is normal rather than contradictory.

Distance limits a single stream

A single TCP connection is bounded by window size divided by round-trip time, so a transfer from a distant server can use a fraction of available bandwidth no matter how much there is. Download managers open parallel connections for exactly this reason, and it is why a CDN edge nearby beats raw bandwidth.

Estimates early in a transfer are unreliable

Remaining-time figures extrapolate from recent throughput, which fluctuates most at the start while TCP is still ramping up. The estimate after the first ten seconds is usually meaningless, and it stabilises once the connection reaches steady state.