Bandwidth, Throughput, and Latency Basics

"Fast" has several meanings. Three measures describe a network: bandwidth, throughput, and latency — each a different part of the experience.

Bandwidth is the most data a connection can carry per second, like a pipe's width — measured in bits per second (Mbps, Gbps). It is a capacity limit.

Speeds use bits; file sizes use bytes, and 8 bits make 1 byte. So 100 Mbps delivers about 12.5 megabytes per second at best.

Throughput is the speed you actually get, usually below bandwidth. Congestion, distance, and errors all drag it down. Bandwidth is the limit; throughput is reality.

Latency is the delay before data starts moving — how long a packet takes to travel, in milliseconds. Low feels snappy; high feels laggy.

Round-trip time (RTT) is how long a packet takes to reach a destination and reply. The ping tool reports it — that is your ping. Lower is snappier.

Jitter is how much latency varies. Uneven delays make voice and video stutter even when the average is low. Steady timing matters for live media.

Different tasks stress different measures. Big downloads need throughput; video calls and gaming need low latency and jitter to feel natural.

More bandwidth does not fix everything. A call on a 1 Gbps link with 300 ms latency still lags, because delay, not capacity, is the problem.

Latency comes from distance, router processing, and congestion. Distance sets a hard floor — even light takes time, so a far server always feels slower.

Bandwidth is max capacity, throughput is what you get, latency is delay, and jitter is how steady that delay is. Name the failing one to fix it.

Which measure best describes the delay before data arrives?

Bandwidth is max capacity; throughput is your real rate; latency is delay (ping/RTT), and jitter is its variation. Downloads need throughput, calls need low latency.