Jitter is the variation in ping (round-trip time) from one packet to the next — not the ping itself. Two connections can have the same average ping and feel completely different if one is jittery. This calculator computes jitter from a list of your own ping samples three different ways, because most "jitter calculators" report one number with no stated method — and that's exactly why two tools fed the same numbers disagree.
ping series for the current figure.Jitter Calculator
Liveping -n 20 8.8.8.8. On macOS/Linux: ping -c 20 8.8.8.8.
Show the working
What Jitter Actually Measures
Ping tells you how long one round trip took. Jitter tells you how consistent that time is from packet to packet. A connection averaging 30 ms with almost no variation feels smoother in a fast-paced game than one averaging 25 ms but swinging between 10 ms and 60 ms — the second connection has lower average ping and worse jitter, and worse jitter is usually what players actually feel as stutter or rubber-banding.
Why Three Numbers From the Same Samples
| Method | How it's calculated | Where it's used |
|---|---|---|
| Mean consecutive difference | Average of the absolute difference between each sample and the one before it | The method most consumer network-quality tools use by default |
| Standard deviation | Statistical spread of all samples around the average | Common in monitoring dashboards; treats the whole sample set at once rather than step-by-step |
| RFC 3550 smoothed jitter | Exponentially-weighted running estimate: each new difference is blended in at 1/16 weight | The IETF standard used by RTP/VoIP systems, Wireshark's RTP statistics and many routers' interarrival jitter reports |
The RFC 3550 result starts at zero and converges gradually — early samples are deliberately under-weighted, which is why it reads lower than the other two methods on a short sample set. On a long, continuous stream it converges close to the mean consecutive difference. None of the three is "wrong"; they answer slightly different questions, and quoting one without naming it is how two legitimate tools end up disagreeing on the same data.
Samples: 24, 26, 25, 31, 22, 28, 24, 35, 23, 27 ms. Consecutive differences: 2, 1, 6, 9, 6, 4, 11, 12, 4. Mean consecutive difference = 6.11 ms. Standard deviation = 3.98 ms. RFC 3550 smoothed estimate = 2.83 ms. Same ten numbers, three defensible answers.
What Counts as Good Jitter?
| Jitter (mean consecutive difference) | What it typically feels like |
|---|---|
| Under 5 ms | Effectively imperceptible |
| Under 30 ms | The commonly accepted bar for smooth VoIP calls and competitive gaming |
| 30–50 ms | Occasionally noticeable stutter or audio artifacts |
| Over 50 ms | Regularly disruptive — rubber-banding in games, choppy voice/video |
Common Mistakes
A "12 ms" from one site and "4 ms" from another on a similar connection isn't necessarily a measurement error — check whether each one is reporting mean consecutive difference, standard deviation, or an RFC 3550 smoothed estimate.
A low average ping with high jitter can feel worse in real time than a slightly higher, rock-steady ping. Check both numbers together.
Jitter calculated from 3–4 pings is noisy and unreliable. Run at least 15–20 pings, ideally over a minute or more, before drawing conclusions.
How to Collect Ping Samples
| System | Command |
|---|---|
| Windows | ping -n 20 8.8.8.8 |
| macOS / Linux | ping -c 20 8.8.8.8 |
| Router admin panel | Many routers include a built-in ping/diagnostics tool under Advanced or Tools settings |
Copy the individual "time=" values from the output (not just the summary line) and paste them into the field above, separated by commas or line breaks.
What Actually Reduces Jitter
High jitter with a reasonable average ping is usually a routing or congestion problem, not a distance problem — unlike the raw physics floor a Ping Calculator reports, jitter is created by inconsistency along the path: Wi-Fi interference, a congested link, or an inefficient route that varies packet-to-packet. Wired connections and uncongested networks typically reduce jitter; nothing removes the underlying propagation delay set by distance.
When routing — not your Wi-Fi — is the cause
If your jitter reads high even on a wired, uncongested connection, the instability may come from an unstable route, congestion beyond your ISP, or other factors along the path. When routing between your ISP and the game server is part of the problem, a routing-optimization service like GearUP may sometimes smooth that path out.
Check GearUP →GearUP cannot fix jitter caused by your own Wi-Fi or local network, since that happens before traffic ever reaches the routing service.
Affiliate disclosure: This is an affiliate link. If you purchase through it, Numeros may earn a commission at no additional cost to you. This does not affect our calculations, test results, or editorial conclusions.
Frequently Asked Questions
What is jitter in networking?
Jitter is the variation in latency (ping) between consecutive packets, rather than the latency itself. High jitter means your ping swings around a lot even if the average looks fine, which is often more disruptive for real-time applications than a slightly higher but steady ping.
Why do different jitter calculators give different numbers for the same pings?
Because "jitter" isn't one formula. Mean consecutive difference, standard deviation, and the IETF RFC 3550 smoothed estimate are all legitimate, commonly used methods that produce different numbers from identical data. This calculator shows all three so you can see the gap directly.
What jitter is acceptable for online gaming?
Under 30 ms is the commonly accepted bar for smooth, competitive-feeling gameplay and VoIP calls. Above 50 ms, rubber-banding, stutter and audio artifacts become regularly noticeable.
Is jitter more important than ping?
Neither is more important in every case; they measure different things. Ping sets the baseline delay; jitter sets how consistent that delay is. A steady 60 ms connection can feel better in practice than an average-25 ms connection that spikes unpredictably to 90 ms.
How many ping samples do I need for an accurate jitter reading?
At least 15–20 samples is a reasonable minimum; a handful of pings produces a noisy, unreliable estimate. Running the test for a minute or more captures more realistic variation, including any periodic congestion.
What is RFC 3550 jitter?
RFC 3550 is the IETF standard defining RTP (Real-time Transport Protocol), used for VoIP and video. It specifies a smoothed interarrival jitter formula — an exponentially weighted running average with a 1/16 smoothing factor — used by routers, Wireshark's RTP statistics, and RTCP quality reports.
Can Wi-Fi cause high jitter even with good average ping?
Yes. Wi-Fi interference, signal strength fluctuations and channel contention with other devices are common causes of jitter that a wired Ethernet connection typically avoids, even when the average ping looks similar on both.
Does jitter affect file downloads the way it affects gaming or calls?
Much less. File transfers use TCP, which buffers and reassembles data regardless of arrival timing, so jitter mainly costs a little throughput rather than causing visible glitches. Real-time applications like gaming and VoIP feel jitter directly because they can't wait for buffering.
Why does the RFC 3550 result look lower than the other two numbers?
The RFC 3550 formula starts at zero and blends in each new difference at only 1/16 weight, so it converges slowly. On a short sample list it reads lower than the mean consecutive difference; on a long, continuous stream the two typically move closer together.
Can a game booster or VPN reduce jitter?
It can, specifically when the jitter is caused by an inefficient or congested route between your ISP and the game server — routing traffic over a more stable path can smooth out variation. It cannot remove jitter caused by your own local Wi-Fi or home network congestion, since that happens before traffic ever reaches the routing service.
How Games and VoIP Actually Cope With Jitter
Real-time systems don't just suffer from jitter passively — most build in a small delay on purpose to absorb it, called a jitter buffer. Incoming packets are held for a few milliseconds and released in order at a steady rate, smoothing out arrival-time variation before it reaches the game engine or your speakers.
| Buffer size | What it trades off | Typical use |
|---|---|---|
| Small (10–20 ms) | Lower added delay, but can't absorb large jitter spikes — they still show up as stutter | Competitive gaming, where every millisecond of added delay matters |
| Medium (40–60 ms) | Balances smoothness against responsiveness | Most VoIP calling apps by default |
| Large (100 ms+) | Very smooth playback, but adds noticeable delay | One-way streaming (video/audio) where delay matters less than smoothness |
This is also why a jitter reading well within "Good" still occasionally causes a stutter under real load: a jitter buffer sized for your typical variation can be overwhelmed by an unusually large single spike, even if the mean consecutive difference across the whole sample looks fine. It's part of why the RFC 3550 method — which reacts gradually rather than to any single spike — is the one built into real-time protocols rather than a simple average.
Which Games Actually Need Low Jitter
Jitter tolerance isn't the same across every genre — the timing precision a game demands determines how much variation actually breaks the experience.
| Genre | Jitter sensitivity | Why |
|---|---|---|
| Competitive FPS, fighting games | Very high | Frame-precise inputs and hit registration make even 20–30 ms of inconsistency feel like input lag |
| MOBAs, real-time strategy | Moderate | Actions queue and resolve on a slightly longer cycle, giving some room to absorb variation |
| MMORPGs, survival games | Low to moderate | Server-authoritative movement and longer action windows tolerate more jitter before it's noticeable |
| Turn-based, strategy, most single-player | Minimal | No real-time synchronization requirement, so jitter rarely affects the experience at all |
This is worth knowing before spending money or effort chasing a lower jitter number: the same 15 ms reading that's a real problem in a competitive shooter may be completely invisible in a turn-based strategy game. Match the fix to what you actually play.
Sources
- The RFC 3550 interarrival jitter formula. J(i) = J(i−1) + (|D(i−1,i)| − J(i−1)) / 16, defined in IETF RFC 3550 (RTP: A Transport Protocol for Real-Time Applications), Appendix A.8 — the method Wireshark's RTP statistics module implements.
- Jitter comfort thresholds. Under 30 ms as the accepted bar for VoIP and gaming, over 50 ms as regularly disruptive, cross-checked against PacketProbe's public network-testing methodology and IT community consensus on acceptable real-time conferencing parameters.
- Jitter buffer sizing. General real-time-protocol engineering practice for balancing added delay against smoothing capacity — reflected in how VoIP and streaming applications configure default buffer sizes, not a single named standard.
- Standard deviation as a jitter method. General statistical definition, cross-referenced against the RFC 3550 documentation above for how it differs from the smoothed interarrival method.
Related Calculators
- Ping Calculator — the physics-floor and routing-adjusted RTT this tool's jitter builds on
- Packet Loss Calculator — the other network-quality metric jitter doesn't cover
- Gaming Network Quality Calculator — combines this jitter figure with ping and packet loss into one rating
- Subnet Calculator — work out IP ranges and network sizes
- Standard Deviation Calculator — the general-purpose statistics behind one of the three jitter methods above