What Is Good Latency for Gaming?
Under 30 ms of network latency is very good, 30-60 ms is comfortable, above 120 ms you are compensating rather than reacting. But network latency is only part of the delay between your hand and the screen — and often not the part that is hurting you.
The short answer
- Under 30 ms — a competitive baseline. Nothing about the network is holding you back.
- 30-60 ms — comfortable in every genre. Most players sit here.
- 60-100 ms — perfectly playable, but you will lose some close duels in tactical shooters against lower-latency opponents.
- Above 120 ms — you start leading targets and pre-firing out of habit rather than reacting.
Why the number on your screen is not the whole delay
When people ask what good gaming latency is, they are usually thinking about the ping counter in the corner of the screen. That counter measures one leg of the journey: your machine to the server and back. The delay you actually feel — the gap between pressing a button and seeing the result — is longer, and it is assembled from at least five separate contributions.
The industry term for the whole chain is click-to-photon latency: the time from a physical input to the corresponding photons leaving your monitor. Understanding how it breaks down is what separates people who fix their lag from people who buy a faster internet plan and feel no difference.
| Stage | Typical cost | What controls it |
|---|---|---|
| Input device | 1-10 ms | Polling rate. A 1000 Hz wired mouse costs about 1 ms; a low-power wireless controller can cost ten times that. |
| Game engine sampling | 3-16 ms | Frame rate. Your input waits until the next frame begins, so higher fps genuinely lowers latency. |
| Network round trip | 10-150 ms | Distance to the server and route quality. This is the part the ping counter shows. |
| Server tick | 4-16 ms | Tick rate. 128-tick averages 4 ms of wait, 30-tick averages 16 ms. Set by the game, not by you. |
| Render queue | 5-50 ms | Pre-rendered frames and V-Sync. The single biggest hidden cost on a default configuration. |
| Display | 3-25 ms | Refresh rate plus panel processing. A 60 Hz TV in a non-game picture mode is the worst common case. |
What that means in practice
Add the optimistic column and a well-tuned setup lands around 26 ms of non-network latency. Add the pessimistic column and a console on a 60 Hz television with a wireless controller is carrying roughly 100 ms before the network contributes anything at all.
That is the important conclusion. A player at 70 ms ping on a 144 Hz monitor with a wired mouse and V-Sync off is very likely reacting faster than a player at 25 ms ping on a TV with frame interpolation enabled. The ping counter shows one of them winning and the stopwatch shows the other.
So before you chase the last 10 ms of network latency, check the cheap wins: turn off V-Sync and any motion smoothing, enable your display's game mode, cap frame rate slightly below your refresh rate rather than at it, and plug the controller in.
Genre targets, and why they differ so much
A tolerance number only makes sense against how the game resolves disagreements between clients. Shooters resolve them with lag compensation, which rewinds the server world to the moment you fired — generous to the shooter, punishing to the person being shot at. That asymmetry is why latency feels worse when you are holding an angle than when you are pushing one.
Fighting games use rollback netcode, which predicts your opponent's inputs and corrects when the real ones arrive. Rollback hides delay remarkably well up to a point, then produces visible teleporting once the prediction window is exceeded, which is why the community talks in frames rather than milliseconds: at 60 fps, every 16 ms is one frame of rollback.
MMOs and strategy games queue actions on the server and confirm them afterwards, so a 150 ms delay is an inconvenience rather than a defeat. And cloud gaming inverts the whole model: every input travels to a data centre and returns as a video frame, so the network budget has to absorb encoding and decoding on top of the round trip.
How to tell which part is actually hurting you
Run a latency test and note your closest region. If that number is under 50 ms, your network is not the problem and no amount of router tuning will change how the game feels — look at the render and display stages instead.
If the closest region is high, check whether it is high everywhere or only in one game. A connection that is slow to every region has a local or ISP problem. A connection that is fine everywhere except one game has a routing problem to that publisher's data centres, which usually means picking a different server region in the game's own settings.
And if the average looks fine but matches still feel unreliable, look at jitter rather than latency. Consistent delay is something netcode was designed to absorb; inconsistent delay is not.
Common questions
What is good latency for gaming in one number?
For the network portion alone: under 30 ms is very good, 30-60 ms is comfortable for nearly every game, 60-100 ms is playable but noticeable in shooters, and above 120 ms you are compensating for the delay rather than reacting. Total click-to-photon latency is a different figure — a well-configured setup at 30 ms network latency still ends up around 55-75 ms from mouse click to visible reaction.
Is 40 ms latency good for gaming?
Yes for almost everything. 40 ms of network latency is comfortable for battle royales, MOBAs, MMOs, racing and casual shooters. It is only a genuine handicap in tactical shooters played against opponents sitting at 15-20 ms, and in fighting games where the rollback window is measured in frames rather than milliseconds.
Why does my game feel laggy at 30 ms latency?
Because network latency is one component of several. A 60 Hz display adds up to 16 ms, a triple-buffered render queue can add 30-50 ms, V-Sync adds another frame or more, and a wireless controller adds around 5-10 ms. If the game feels sluggish while your ping is genuinely low, the milliseconds are being added after the packet arrives, not before.
Does higher FPS reduce latency?
Yes, and by more than people expect. Frame rate sets how often the game samples your input and how quickly a new frame reaches the display. Going from 60 to 144 fps removes roughly 10 ms of input-to-display delay even if the network is unchanged. This is why competitive players cap frame rate above their refresh rate rather than enabling V-Sync.
Is latency or jitter more important?
For a competitive player, stable and slightly higher usually beats fast and erratic. Netcode compensates for a consistent delay through prediction and interpolation, but it cannot predict a route whose delay jumps between 20 ms and 90 ms. That variance is what produces rubber-banding and hits that register a moment after you moved out of the way.
Does server tick rate count as latency?
It does, and it is often overlooked. A 64-tick server processes a new world state every 15.6 ms and a 128-tick server every 7.8 ms, so on average you wait half a tick before your action is even considered. That is 8 ms versus 4 ms of latency added by the server itself, entirely outside your control.