Gaming

Input Lag

The time from a local input, such as a mouse click, to a visible change on the display. Game processing, graphics rendering and display behaviour all contribute; network delay is an additional factor in online play.

How it works

Input lag is the local delay from an action to the resulting visible change. The signal passes through the input device, game processing, rendering and display. A game can have a high average frame rate but still queue work that makes controls feel slower than that average suggests.

Important differences and limits

Network ping is not a measurement of this complete local path. Monitor response time is also only one part: a quoted pixel-transition time does not include every processing stage. Frame generation can raise displayed FPS without shortening the response of the underlying game simulation.

A useful practical check

Use a repeatable offline action to separate local response from online connection problems. Check display refresh rate, game frame caps, synchronisation settings and any supported low-latency option. Change one setting at a time. Prefer stable frame times and comfortable controls over a higher counter that introduces visible tearing or inconsistent movement.

An example in practice

If a menu reacts slowly even offline, changing an internet plan is unlikely to address that local delay. If local actions respond promptly but remote events arrive late, investigate the online path separately instead of applying the same graphics adjustment to both symptoms.

Additional examples and checks

First reproduce the action locally, then compare its online equivalent. Record frame rate, graphics settings, display mode and network conditions so tests can be repeated. If using a measurement tool, read its metric definition. Choose settings that reduce the relevant delay without creating unstable frame delivery or compromising the controls you need.

For example, a click-to-visible-response test and a network ping test can produce very different numbers on the same PC. Neither number is wrong solely because they disagree: the test boundaries differ. Label the measurement so later comparisons address the same delay. Record whether the action is local or requires server confirmation, so a later test preserves the same measurement boundary.

Technical sources

← All glossary terms