What jitter is, in plain terms
Imagine a courier who delivers a parcel to the same house every day. On a good route, the journey takes about half an hour, give or take a minute. On a bad route, the same journey takes twenty minutes on Monday, fifty on Tuesday, and thirty-five on Wednesday, the average still looks respectable, but nobody relying on those deliveries can plan anything around them.
Your game connection has the same two properties. Ping is the journey time for data between you and the server. Jitter is how much that journey time varies from packet to packet. A connection sending packets that arrive after 30, 31, 29 and 30 milliseconds has almost no jitter. A connection sending packets that arrive after 24, 71, 26 and 95 milliseconds has a great deal of it, even though if you averaged the two sets, they would look surprisingly similar.
That is the essential point: jitter is not a second kind of slowness. It is inconsistency, and it is measured separately from ping because it harms games in a separate way. A player can adapt to a steady delay, your hands learn it within minutes. Nobody can adapt to a delay that keeps changing, because there is nothing stable to learn.
How jitter is measured
Jitter is expressed in milliseconds, like ping, and at heart it is simple arithmetic: look at how much the latency changes between consecutive packets, and average those changes over time. If packets arrive at 30, 45, 33 and 60 ms, the swings between neighbours, 15, 12, 27 ms, are what jitter describes.
You rarely need special software to see it. Run a continuous ping test and read the individual reply times rather than the average at the end: the spread between the lowest and highest replies is jitter showing itself. Some connection tests report a jitter figure directly. The per-tool methods, and how to record results properly, are set out in our guide to testing your ping before playing online games; what matters here is learning to read the pattern:
| Jitter level | What the readings look like | What it feels like in game |
|---|---|---|
| Under ~3 ms | Replies within a few ms of each other | Rock-steady; effectively invisible |
| ~3–10 ms | Small, regular wobble around the average | Fine for most games and players |
| ~10–30 ms | Noticeable swings between readings | Fast games feel inconsistent; timing feels unreliable |
| Over ~30 ms | Readings leaping between low and high values | Stutter, teleporting opponents, shots that feel randomly late |
Treat those bands as a rough guide rather than a law of physics, sensitivity varies by game and by player, but the direction is universal: the wider the spread, the worse the experience, whatever the average says.
A worked example makes it concrete. Suppose two players each take five readings during a match. Player A records 41, 43, 40, 42, 41 ms: average roughly 41 ms, jitter just 1–3 ms between neighbours. Player B records 18, 64, 22, 70, 31 ms: the average works out near 41 ms too, but neighbouring readings differ by 40 ms or more. Given the choice, experienced players would take Player A's connection every time, and if you swapped their connections without telling them, Player B would report "lag" within a minute while Player A noticed nothing at all. The average called them equal; the jitter knew better.
Why a good average ping can still feel awful
Averages hide the very thing jitter is. Consider two connections, both averaging 50 ms across a match. The first delivers almost every packet at 48–52 ms. The second alternates between 15 ms and 85 ms. On paper, identical. In play, the first is smooth and predictable; the second is a connection where your view of the game keeps lurching between "ahead" and "behind", opponents seem to skip, and your inputs land at inconsistent moments.
This is why players so often say "my ping says 30 but the game feels laggy", and why they are usually not imagining it. The single number on the screen is typically an average or a spot reading; the suffering is in the spread it conceals. If that sentence describes your evenings, measure the spread before you measure anything else. And if the number on screen swings wildly between matches rather than within them, that is a different pattern again, the troubleshooting method in our unstable-connections guide applies.
Jitter and ping spikes: cousins, not twins
These two are related, and this site treats them in separate guides deliberately, so the division is worth stating plainly. Jitter, this article, is the constant, small-scale unevenness in your latency: the background wobble present in every second of play, caused by the everyday behaviour of your network path. Ping spikes are sudden, large jumps, a connection sitting at 30 ms that leaps to 300 ms for five seconds, usually because something specific happened: a housemate started an upload, interference hit the Wi-Fi, a route congested hard. Spikes are a symptom with triggers to hunt down, and they have their own guide: why your ping keeps spiking in games.
The practical difference: jitter is treated by making the whole path steadier, while spikes are treated by finding and removing the trigger. Many troubled connections have both, a jittery baseline with spikes on top, and it is worth dealing with the baseline first, because spikes are much easier to spot once the background wobble is gone.
What causes jitter
Wi-Fi, first and most often
Wireless links are jitter machines by nature. Radio interference from neighbouring networks makes the link retry transmissions; walls and distance weaken the signal; power-saving features wake and sleep. Each retry adds a slightly different delay, which is precisely what jitter is. This is why the same connection often measures beautifully on a cable and wobbles on Wi-Fi in the next room.
A busy connection: queueing and bufferbloat
When your connection is carrying more traffic than it can move instantly, a download, a stream, someone else's upload, packets wait in a queue at your router before leaving. Waiting time varies with the length of the queue, so a busy connection produces uneven delays automatically. Routers that queue too generously make this worse: packets sit in an oversized buffer, arriving late and bunched together. Gamers know this family of problems as bufferbloat, and its fingerprint is jitter that appears exactly when the household is busy, and vanishes when the house is quiet.
Congestion further out
The same queueing happens on your provider's network and along the route to the server, especially at peak hours. You cannot see these queues, but you see their effect: evening jitter on a connection that is steady in the morning, on a wired link, with nobody else home, that pattern points away from your house.
Route changes
Your traffic's path is not fixed. When routing shifts mid-session to a longer or busier path, the base latency steps up or down, and during the instability around the change, packets arrive unevenly. Frequent route flapping, the path repeatedly re-deciding itself, shows up as jitter that no home change touches.
Variation at the server end
Finally, some unevenness is born where your data arrives. A game server under heavy load does not process every update at perfectly even intervals, and a match hosted far down a struggling region's priority list can hand you inconsistent timing that your connection faithfully delivers. You can identify this kind of jitter because everyone in the match tends to feel it together, and it is the one kind no setting, cable or tool at your end can correct.
How games react to jitter behind the scenes
Game developers know the internet is uneven, and modern netcode is built to hide it. Understanding the hiding explains the symptoms. To keep movement looking smooth, your game does not show you other players exactly where the last packet said they were; it predicts and interpolates, drawing motion between known points and guessing slightly ahead. To do that, it holds incoming information in a small buffer, deliberately adding a touch of delay so that late packets can still be woven in.
Jitter attacks exactly this mechanism. If packets arrive evenly, a small buffer suffices and play feels immediate. If they arrive unevenly, the game faces a choice: keep the buffer small, and movement stutters or snaps whenever a packet arrives too late for its slot; or grow the buffer, and everything becomes smoother but more delayed. When the buffer runs dry entirely, you see the correction as a teleport, the family of symptoms covered by our guide to fixing lag and rubberbanding in online games. Jitter is also why hit registration can feel inconsistent between two fights at the same average ping: in one fight your packets arrived in rhythm, in the other they did not.
What actually reduces jitter
- Use a wired connection. The single most effective jitter fix, because it removes the most jitter-prone link in the whole chain. If your jitter survives a wired test, you have learned it is not born in your wireless, valuable knowledge in itself.
- Quiet the connection while you play. Pause downloads, cloud backups and other uploads. Less traffic in the queue means shorter, more even waits for your game's packets.
- Let the router prioritise. If your router offers quality-of-service (QoS) or gaming prioritisation settings, they work by moving game packets to the front of the queue, attacking queueing delay, which is jitter's home-grown source. Worth testing, honestly measured.
- Choose a nearer server region. A shorter journey passes through fewer queues and fewer networks, leaving fewer opportunities for unevenness to creep in. It is not a guarantee, a congested short route can jitter more than an idle long one, but it stacks the odds.
- Consider the route itself. Where jitter is born on the path between your provider and the server, a routing tool can help by choosing a steadier path. This is squarely what GearUP Booster describes its Adaptive Intelligent Routing as doing, monitoring game server and network conditions in real time and routing accordingly, and GearUP's own product description lists jitter reduction among its aims alongside ping and packet loss. Judge it as you would any fix: measure your spread with it off and on, across several sessions.
Frequently asked questions
What jitter is acceptable for gaming?+
As a rough guide, under about 5 ms is effectively invisible, up to roughly 10 ms is comfortable for most games, and once readings routinely swing by 30 ms or more, most players feel the inconsistency clearly in fast titles.
Is jitter worse than high ping?+
Often, yes. A steady 70 ms can be learned and played around; a connection averaging 35 ms but swinging between 15 and 90 ms cannot, because the delay never stays still long enough to adapt to. Consistency beats raw speed in most real matches.
Can I see jitter in a normal speed test?+
Some speed tests report a jitter figure, but the simplest view is a continuous ping test: ignore the average for a moment and look at how far apart the individual replies are. That spread is jitter, visible to the naked eye.
Does an Ethernet cable fix jitter?+
It fixes the share of jitter born in your Wi-Fi, which for most households is the largest share. Jitter caused by household congestion, your provider or the route will survive the cable, which is exactly what the wired test tells you.
Why is my jitter worse in the evening?+
Because queues fill up. In the evening your household is busier and so is your provider's network, and packets waiting in fuller queues wait for uneven amounts of time. Evening jitter with a quiet-house morning is the fingerprint of congestion rather than equipment.
Jitter is the quiet half of connection quality, the half the headline ping number never mentions. Once you start reading the spread between your readings instead of the average alone, a whole class of "my ping is fine but the game feels wrong" problems suddenly makes sense, and the fixes become obvious in kind: steady the path, quiet the queues, and measure honestly to prove each step worked.
