unlag.me
Guide

Redstone lag: how to find the machine that is eating your server

One twelve-block clock can cost more server time than every mob on your map. Component count will never find it. Here is what actually causes redstone lag, how to locate it, and how to deal with it without breaking someone's base.

Published 2026-09-06 · 3 min read · unlag.me
In this guide
  1. Why component count is wrong
  2. What actually causes it
  3. How to find it
  4. Fixing it without losing the player
  5. Preventing the next one
  6. Where this fits

Why component count is the wrong measure

The instinct when a server lags is to look for the biggest redstone build and blame it. That build is usually innocent.

Redstone costs nothing when it is idle. A 900-component display that changes once an hour is free. What costs time is redstone events — block updates propagating through the world — and those come from things that change state constantly.

A twelve-block observer clock firing every tick can generate more block updates per second than an entire town of static redstone. Size is not the signal. Frequency is.

What actually causes it

Observer clocks

Two observers facing each other retrigger forever. This is the single most common accidental lag machine — often built by someone who wanted a clock and did not realise it would run at maximum speed with nothing to stop it.

Hopper clocks and item loops

An item passing back and forth between two hoppers generates comparator updates continuously. Cheap individually, constant by nature.

0-tick farms

Deliberately abuse piston update timing to force rapid growth. Extremely effective, and extremely expensive per block.

Update suppression

A more advanced technique that can put the server into states it handles badly. Rare, but worth knowing about if you run anarchy.

Large piston arrays

Every piston move recalculates the blocks around it. A flying machine or door array firing repeatedly multiplies that cost.

How to find it

The reliable approach is to measure redstone events per chunk per tick and rank chunks by that. The problem chunk usually stands out by an order of magnitude — not 20% busier than the next one, but twenty times busier.

If you have a profiler, redstone shows up under block-entity ticking and block physics. If you can attribute those samples to a chunk, you have your coordinates.

Without tooling, the crude method still works: unload regions and watch MSPT. Stop the world from ticking a suspect area and see whether the number moves. It is slow and disruptive, but it is definitive.

What a runaway looks like: one chunk producing thousands of redstone events per second while your next busiest produces a couple of hundred. That gap is the tell.

Fixing it without losing the player

The reflex is to break the machine. Resist it — the person who built it is usually one of your more invested players, and they normally have no idea it is causing a problem.

In order of preference:

  1. Throttle the chunk. Cap redstone events per second in that area. The build keeps working, just slower. Nothing breaks and nobody loses anything.
  2. Tell the owner, with coordinates. Most players will happily redesign a clock if you explain what it costs. Many will find it interesting.
  3. Offer a replacement design. A clock gated by a lever costs nothing when off.
  4. Remove it only when the first three have failed, or it is clearly deliberate.

Preventing the next one

Rate limits work better than bans. A per-chunk cap on redstone events means an accidental clock throttles itself instead of taking the server down at 3am, and you find out from an alert rather than from players.

Set the limit generously — high enough that ordinary farms never touch it, low enough that a runaway is caught. Somewhere around a few hundred events per second per chunk suits most survival servers.

The important part is that legitimate technical players never notice the limit exists. Blunt global restrictions are why many servers ban redstone entirely and lose exactly the players who would have built the most interesting things.

Where this fits

Redstone is worth checking, but check it in order — most lag is not redstone. Work through the diagnostic checklist first, and see TPS and MSPT explained for reading the numbers you will be looking at.

unlag.me tracks redstone events per chunk in real time, ranks hotspots with coordinates and the owner's name, and can throttle a single chunk without touching the rest of the world.

Stop guessing which plugin it is

unlag.me samples your server thread thousands of times a second and names the exact plugin, machine or setting costing you ticks — then tells you what to change. Run one command in game and read it in your browser.

Open the dashboard