Jitter and latency are two different network performance metrics that measure different things about how data travels across a connection. Latency is the time it takes for data to get from one point to another. Jitter is how much that travel time varies from one packet to the next. This page explains what each metric means, how they affect things like video calls and online gaming, and how to measure them. By the end, you’ll know which issue is affecting your connection and what to do about it.
Understanding Latency: The Total Delay in Network Communication
Latency is the total time it takes for data packets to travel from your device to a destination and back. Think of it like the total delivery time for a package: the time from when you send it until you get confirmation it arrived. This round-trip measurement captures every delay along the network path, which makes it a basic indicator of how responsive your connection feels.
What Latency Measures in Network Performance
Network latency is measured in milliseconds (ms) and comes from three separate components that add up to the total delay you experience. The ping command measures latency by sending test packets and recording how long the round trip takes.
Propagation latency: The physical travel time across distance as signals move through cables or wireless transmission
Transmission latency: The time it takes to push all the packet bits onto the network
Processing latency: The delay while routers and switches read packet headers and figure out where to send them
Common Causes of High Latency
Latency comes from physical limits and network infrastructure problems. Usually, several factors combine to create the total delay you measure.
Physical distance: The farther apart two endpoints are, the longer signals take to travel between them
Network congestion: Too much traffic on the network creates queuing delays as packets wait their turn
Routing inefficiency: Packets taking a longer-than-necessary path through multiple hops add processing time at each router
Transmission medium quality: Poor cabling, low-frequency wireless, or degraded fiber slows things down
Insufficient bandwidth: Limited capacity forces packets to sit in buffers before they can be sent
Server processing delays: Overloaded servers take longer to respond to incoming requests
Acceptable Latency Benchmarks by Application Type
What counts as acceptable latency depends a lot on what you’re doing. These benchmarks are target ranges, not hard cutoffs. Your experience gets worse gradually as latency climbs past the optimal levels.
Application Type
Acceptable Latency
Optimal Latency
User Impact of High Latency
VoIP/Voice Calls
<150ms
<30ms
Talk-over, echo, unnatural pauses
Video Conferencing
<150ms
<50ms
Audio-video desync, delayed reactions
Online Gaming (FPS)
<50ms
<20ms
Lag, missed actions, competitive disadvantage
Online Gaming (Strategy)
<100ms
<50ms
Input delays, reduced responsiveness
Video Streaming
<200ms
<100ms
Slow initial buffering (minimal impact once streaming)
Web Browsing
<200ms
<100ms
Slow page loads, delayed interactions
These benchmarks assume consistent latency. When latency varies a lot, you get a different set of problems even if the average looks fine. That’s where jitter comes in.
Understanding Jitter: The Variability in Network Timing
Latency measures total delay. Jitter measures how inconsistent that delay is over time. That inconsistency in packet arrival timing creates its own set of problems, and for real-time applications, it’s often more disruptive than latency alone.
What Jitter Measures: Inconsistency in Packet Arrival
Jitter measures how much packet arrival times vary from one packet to the next. If latency is your average delivery time, jitter is the unpredictability: packages arriving sometimes in 2 days, sometimes in 5 days, with no consistent pattern. Packets sent at regular intervals (every 10ms, for example) may arrive with irregular spacing (12ms, 8ms, 15ms, 9ms) because network conditions keep changing.
Here’s a concrete example: Packet 1 arrives 20ms after sending. Packet 2 arrives 50ms after sending. That 30ms gap between expected and actual arrival time is jitter. Jitter is measured in milliseconds (ms) and represents the standard deviation or variance in latency measurements over time.
Jitter matters more than latency for real-time applications because your brain can adapt to a consistent 100ms delay. But when latency swings between 20ms and 150ms with no pattern, the experience becomes choppy and unusable.
Common Causes of Network Jitter
Jitter usually comes from network instability and resource contention, not physical distance. The root causes are different from what drives latency. Understanding these causes is especially important when diagnosing intermittent network connectivity problems, where jitter and packet timing irregularities are often the first signs of a deeper fault.
Network congestion and microbursts: Sudden traffic spikes overwhelm device buffers, causing packets to be forwarded at irregular intervals
Wireless interference: Radio frequency noise causes packet drops and retransmissions with variable timing
Faulty network equipment: Degraded routers, switches, or network interface cards process packets inconsistently
The distinction matters: jitter comes from variability and instability, while latency comes from capacity and distance. That difference shapes how you troubleshoot each one.
Acceptable Jitter Benchmarks by Application Type
Jitter tolerance is generally much lower than latency tolerance. An application that handles 100ms latency fine might fall apart with 30ms jitter. Jitter also compounds with latency, so high jitter is worse when latency is already elevated. Low jitter can partially make up for moderate latency.
Applications use jitter buffers to smooth out small amounts of jitter by holding packets briefly before playing them. That helps, but it adds latency to your connection as a trade-off.
Jitter vs Latency: Key Differences and How They Interact
Jitter and latency both involve delays, but they measure different things and point to different problems. Knowing how they relate helps you read test results correctly and figure out what to fix first.
Side-by-Side Comparison: Jitter vs Latency
These metrics are independent. You can have any combination of high or low latency and high or low jitter. Each combination creates different problems and points to different causes.
What It Measures: Latency measures total round-trip delay for packets; jitter measures variation in packet arrival timing
Calculation: Latency is the average time from source to destination and back; jitter is the standard deviation of latency measurements
Primary Causes: Latency comes from physical distance, bandwidth limits, and congestion; jitter comes from network instability, equipment issues, and QoS problems
Impact on Real-Time Apps: Latency creates noticeable lag and delayed responses; jitter creates choppy, inconsistent experiences
Impact on Non-Real-Time Apps: Latency slows downloads and page loads; jitter has minimal impact
Typical Acceptable Range: Latency <150ms for most applications; jitter <30ms for real-time applications
Optimal Range: Latency <50ms for real-time applications; jitter <10ms for real-time applications
Measurement Tool: Latency uses ping and traceroute; jitter uses ping with multiple samples and specialized jitter tests
Primary Solution Focus: Fixing latency means increasing bandwidth, improving routing, or reducing distance; fixing jitter means stabilizing the network, replacing faulty equipment, and configuring QoS
Which Is Worse: High Jitter or High Latency?
For real-time applications, high jitter is typically worse than high latency. People can adapt to consistent delays, but unpredictable timing is much harder to work around. A video call with a steady 120ms latency feels like a natural pause in conversation. A call where latency swings between 20ms and 150ms becomes unusable because nobody can tell when to speak.
The answer depends on what you’re doing:
Real-time interactive applications (VoIP, gaming, video conferencing): Jitter is worse — consistency matters more than speed
Streaming applications (video, audio): Latency is worse for initial buffering, but jitter causes mid-stream interruptions
Non-real-time applications (web browsing, file downloads): Latency is worse — jitter has minimal impact
If you have both high jitter and high latency, fix jitter first for real-time applications. Stabilizing the network often brings latency down too.
How Jitter, Latency, Ping, and Packet Loss Relate
These four metrics get confused a lot, but they form a connected picture of network performance. Ping is a tool that measures latency by sending test packets and recording round-trip time. Latency is what ping measures: the total delay. Jitter is calculated from multiple ping measurements by looking at how much those latency results vary. Packet loss is a separate metric that tracks the percentage of packets that never arrive, and it’s often caused by the same things that create jitter.
These metrics affect each other in important ways:
High packet loss forces retransmissions, which increases both latency and jitter
High jitter often goes hand in hand with packet loss because both come from network instability
You can have high latency with low jitter and no packet loss (a long-distance connection with stable routing)
You can have low latency with high jitter and packet loss (a local network with faulty equipment)
When you run network tests, look at all four metrics together. The pattern tells you what’s wrong. High latency with low jitter points to distance. High jitter with packet loss points to instability. High latency that only shows up during peak hours points to a capacity problem.
How to Measure Jitter and Latency on Your Network
Measuring both metrics is straightforward with tools that are already built into your operating system. These steps work on Windows, Mac, and Linux.
Measuring Latency Using Ping (Windows, Mac, Linux)
Ping is the simplest latency measurement tool available, and it’s built into every operating system.
Open your command-line interface: Windows users press Win+R, type “cmd”, and press Enter. Mac and Linux users open Terminal from Applications/Utilities.
Run a basic ping test: Type ping google.com (or any reliable server) and press Enter to send continuous test packets.
Observe the latency results: Look for the “time=” value in each line. That’s your round-trip latency in milliseconds for each packet.
Calculate average latency: Let the test run for 30 to 60 seconds, then press Ctrl+C to stop. The summary shows minimum, maximum, and average latency.
Interpret the results: Average latency under 50ms is excellent. 50 to 100ms is good. 100 to 150ms is acceptable for most uses. Above 150ms may cause problems with real-time applications.
Ping measures latency to a specific server, so test multiple destinations (your router, your ISP, major websites) to figure out whether problems are local or external.
Measuring Jitter Using Extended Ping Tests
Jitter requires multiple latency measurements to calculate the variation. Extended ping tests give you the data you need.
Run an extended ping test: Use ping -n 100 google.com (Windows) or ping -c 100 google.com (Mac/Linux) to send 100 packets and collect enough data for a jitter calculation.
Record the latency range: When the test finishes, note the minimum and maximum latency values in the summary.
Calculate a basic jitter estimate: Subtract the minimum from the maximum latency (45ms max – 20ms min = 25ms jitter range). This isn’t statistically precise, but it gives you a practical indicator.
Use online speed tests for automated jitter measurement: Visit speedtest.net, fast.com, or dslreports.com/speedtest. These tests automatically calculate and display jitter alongside latency and bandwidth.
Interpret jitter results: Jitter under 10ms is excellent. 10 to 30ms is acceptable for most real-time applications. 30 to 50ms may cause noticeable quality issues. Above 50ms typically makes VoIP and gaming unusable.
For more precise jitter measurement, network professionals use tools like iperf. But for most troubleshooting, online speed tests are accurate enough. For a deeper look at how these measurements fit into a broader testing workflow, the guide on data communications testing methods and tools covers the full range of techniques used to verify network performance.
Advanced Testing with Traceroute and Specialized Tools
Traceroute shows you latency at each hop along the network path, which helps you pinpoint where problems are coming from.
Run traceroute to map your network path: Use tracert google.com (Windows) or traceroute google.com (Mac/Linux) to see latency at each router between you and the destination.
Identify problem locations: Look for sudden latency jumps at specific hops. That tells you whether the issue is in your local network, your ISP, or external routing.
Test during different times: Run tests during peak hours (evenings) and off-peak hours (early morning) to see whether congestion is driving your jitter and latency problems.
Use continuous monitoring tools: Install PingPlotter (Windows/Mac) or smokeping (Linux) for long-term latency and jitter tracking that shows patterns over hours or days.
If latency is consistently high at your first hop (your router), the problem is on your local network. If problems show up at ISP hops, that’s a service provider issue and you’ll need to contact support.
Optimizing Jitter and Latency for Better Network Performance
The best fix depends on whether you’re dealing with latency (a capacity or distance problem) or jitter (a stability or consistency problem). Your measurement results tell you which to tackle first. Start with local network fixes like upgrading equipment, switching to a wired connection, and configuring Quality of Service (QoS) settings before going after ISP-level issues. Local changes are faster, cheaper, and entirely in your control. If you rely on a wireless setup, the physical layout and equipment placement in your home can significantly affect both metrics — reviewing best practices for designing reliable wireless networks can help you reduce interference-related jitter at the source.
Is 20ms jitter bad for gaming?
Yes, 20ms jitter is a problem for competitive gaming, especially fast-paced FPS games where optimal jitter should be under 10ms. You’ll likely see rubber-banding, hit detection issues, and inconsistent player movement that puts you at a competitive disadvantage.
Can high jitter cause packet loss?
High jitter doesn’t directly cause packet loss, but both often come from the same root causes: network congestion, faulty equipment, or wireless interference. When jitter gets severe enough that packets arrive outside the application’s buffer window, they’re effectively lost even though they technically made it through.
What’s the difference between ping and latency?
Ping is a network diagnostic tool that sends test packets to measure latency. Latency is the actual round-trip delay time that ping measures. Think of ping as the thermometer and latency as the temperature reading.
Does a VPN increase jitter or just latency?
A VPN typically adds 10 to 50ms of latency because of encryption overhead and routing through VPN servers. Good VPN services keep jitter low by maintaining stable, consistent connections. Poor VPN services or overloaded servers can push both latency and jitter higher.
How does packet loss affect latency and jitter measurements?
Packet loss pushes latency measurements higher because lost packets trigger retransmissions that take extra time. It also inflates jitter calculations by creating gaps in the timing data. For accurate latency and jitter readings, packet loss should be under 1%.
Can you have good latency but bad jitter?
Yes. Low average latency doesn’t mean your connection is smooth. If your packets arrive at wildly inconsistent intervals, local network instability or misconfigured QoS settings are likely the cause, not your distance from a server. If inconsistent performance is affecting you, looking into dedicated connection options could make a real difference.
