IP FrogNetwork check
← Back to the blog
Network quality7 min read

What is jitter, and what is a good jitter speed?

Network jitter is variation in latency. Learn what a good jitter result looks like, how it affects gaming and calls, and how to test and reduce it.

Network jitter is the variation in latency from one measurement to the next. It shows whether data is arriving at a steady pace or on an unpredictable schedule. Lower jitter is better, especially for gaming, voice calls, video meetings, and other real-time activity.

People sometimes call this “jitter speed,” but jitter is not a speed. It is measured in milliseconds (ms), while download and upload speed are measured in megabits per second (Mbps).

What is a good jitter result?

As a practical starting point:

  • Under 15 ms: excellent for most uses, including gaming and video calls
  • 15–30 ms: generally good and unlikely to cause obvious problems
  • 30–60 ms: noticeable in some games, calls, and live audio
  • Over 60 ms: likely to cause an inconsistent real-time experience

These are useful guidelines, not universal pass-or-fail rules. An app’s buffering, your baseline latency, packet loss, Wi-Fi conditions, and the size and frequency of spikes all affect what you actually experience.

IP Frog considers jitter up to 15 ms part of a great gaming result and up to 30 ms part of a good one when latency is also low. For video calls, it evaluates jitter alongside upload speed, download speed, and latency rather than judging one number alone.

How jitter works

Suppose a test sends five small requests and measures round-trip times of 20, 21, 19, 20, and 21 ms. The samples are close together, so the connection has low jitter.

Now compare results of 20, 65, 22, 110, and 24 ms. The connection can still respond in about 20 ms, but some packets take far longer. That inconsistency is high jitter.

The exact formula can vary among testing tools. Some calculate the average difference between consecutive latency samples; others use a different statistical summary. Because methods and test servers differ, compare repeated results from the same tool under similar conditions.

What does jitter feel like?

Downloads and buffered video can often absorb uneven packet timing. Real-time apps have less room to hide it.

Gaming

High jitter can make a player or object jump between positions, a behavior often called rubber-banding. An action may feel immediate one moment and delayed the next. A low average ping does not rule this out because the average can hide individual spikes.

Voice and video calls

Calling apps use a small jitter buffer to reorder packets and smooth minor timing differences. Too much variation can produce robotic audio, clipped words, pauses, frozen video, or sudden drops in quality.

Livestreaming

Regular streaming builds a larger buffer before playback, but livestreams try to stay close to the action. High jitter can contribute to stalls or force the player to increase its delay.

Browsing and file transfers

Jitter is usually less obvious here. These activities care more about overall latency, bandwidth, and reliability, although severe instability can still make a connection feel erratic.

Jitter, latency, and packet loss are different

These three measurements are related but not interchangeable:

  • Latency is the time a packet takes to make a round trip.
  • Jitter is how much that latency varies.
  • Packet loss is the percentage of packets that never complete the trip.

A connection can have low latency but high jitter, or stable latency with occasional packet loss. Looking at only one number can miss the actual problem. Our jitter vs. latency guide shows how to read the first two together.

What causes high jitter?

Common causes include:

  • Crowded Wi-Fi: Nearby networks and devices compete for airtime.
  • Weak signal: Retransmissions and changing radio conditions disrupt timing.
  • Network congestion: Too much traffic creates uneven queues and delays.
  • Heavy uploads: Cloud backups and file sharing can fill the upstream link.
  • Bufferbloat: An overloaded router holds packets in long queues instead of managing them promptly.
  • VPN routing: Encryption overhead and a longer route can add variation.
  • Mobile network changes: Signal shifts and tower handoffs can make delay less consistent.
  • An overloaded device or router: Busy hardware may process packets unevenly.

The test path matters as well. A faraway or overloaded test server can report high jitter even when the connection to a nearby service is stable.

How to lower jitter

Start with the changes that are easiest to test:

  1. Use Ethernet for a baseline. If jitter drops, the local Wi-Fi is the likely source.
  2. Pause other traffic. Stop large uploads, downloads, backups, and high-resolution streams.
  3. Improve the Wi-Fi signal. Move closer to the access point and away from interference.
  4. Test without a VPN. This reveals whether the VPN route is contributing.
  5. Update or replace an old router. Modern traffic management can help when the connection is busy.
  6. Enable smart queue management if available. Features labeled SQM or modern QoS can reduce delay under load.
  7. Test at different times. A pattern that appears only during peak hours may point to congestion.

Avoid changing everything at once. Run the same test after each change so you know which one affected the result.

How to test jitter accurately

For a useful reading, test from the device and connection you actually care about. A wired desktop result does not describe the Wi-Fi experience in another room.

Run several tests at roughly the same time and note whether another person or device is using the network. Compare idle results with results taken during a download or upload. If jitter climbs only under load, queueing or bufferbloat is a strong possibility.

See your jitter now

Run IP Frog’s free speed test to see your jitter in milliseconds alongside latency, download speed, and upload speed. You will also get practical quality estimates for gaming, video calls, and streaming, so the number has some context.

For the cleanest comparison, run the test once on your normal connection and once after pausing other network activity.