What Is Latency?

Latency is the amount of time it takes data to travel across a network and receive a response. It is usually measured in milliseconds, or ms. Lower latency means the connection responds more quickly.

Latency is different from Internet speed. Download and upload speeds measure how much data can move through the connection, while latency measures how quickly that data can begin moving back and forth.

This is why a very fast Internet plan can still feel sluggish during gaming, video calls, remote desktop sessions, or other real-time activities if latency is high.

Understanding Internet Latency, Ping, Jitter & Responsiveness

Latency affects how responsive an Internet connection feels. It matters most when devices and servers need to exchange information repeatedly in real time, such as during online gaming, voice calls, video meetings, cloud applications, and interactive websites.

The sections below explain what causes latency, how ping is measured, what counts as good latency, how jitter and packet loss fit into the picture, and why connection type, Wi-Fi, equipment, congestion, and physical distance can all affect responsiveness.

How Is Latency Measured?

Latency is normally measured in milliseconds (ms).

One millisecond is one-thousandth of a second.

A latency result might look like:

• 8 ms
• 20 ms
• 45 ms
• 100 ms
• 500 ms

Lower numbers generally indicate a more responsive connection.

Most consumer speed tests report round-trip latency, meaning the time required for a data packet to travel from your device to a destination and for the response to return.

What Is Ping?

Ping is commonly used as another way of describing network latency.

Technically, ping is a network test that sends a small packet of data to another device or server and measures how long it takes to receive a reply.

If a game reports a ping of 25 ms, for example, it means the round trip between your device and the game server is taking roughly 25 milliseconds.

In everyday Internet discussions, the terms ping and latency are often used interchangeably.

What Is Round-Trip Time?

Round-trip time (RTT) measures how long it takes data to travel from a device to a destination and back again.

This is the measurement commonly shown by ping tests and many Internet speed tests.

For example:

1. Your device sends a data packet to a server.
2. The packet travels through your router, Internet provider, and other networks.
3. The server receives the packet and responds.
4. The response travels back to your device.

The total time required for that complete trip is the round-trip latency.

What Is Good Latency?

There is no single perfect latency number for every application, but lower is generally better.

These are broad guidelines rather than strict thresholds. Streaming video can tolerate relatively high latency because content is buffered, while competitive online gaming may feel noticeably different with even moderate increases in delay.

Is 20 ms Ping Good?

Yes. A 20 ms round-trip latency is very good for almost any normal Internet activity.

It is generally low enough for:

• Competitive gaming
• Video conferencing
• Voice calls
• Remote desktop
• Cloud applications
• Web browsing

Is 50 ms Ping Good?

Yes. Around 50 ms is still a good result for most users.

Many online games, video calls, and interactive applications can work comfortably at this level.

Highly competitive gamers may prefer lower latency, but a stable 50 ms connection can still provide a good overall experience.

Is 100 ms Ping Bad?

100 ms is usable, but the delay may become noticeable.

For browsing and streaming, it may cause little concern. During online gaming, voice calls, or remote desktop sessions, the additional delay can be easier to feel.

The stability of the connection also matters. A consistent 100 ms connection may perform better than one that constantly jumps between 30 ms and 200 ms.

Why Does Latency Matter?

Many Internet activities involve repeated communication between your device and a remote server.

For example, when playing an online game:

1. You press a button.
2. Your device sends that action to the game server.
3. The server processes it.
4. The updated game state returns to your device.

The longer that communication takes, the more delayed the game can feel.

Latency also affects:

• Video conferences
• Voice calls
• Remote desktop applications
• Cloud gaming
• Interactive websites
• Financial and business applications
• Remote control systems

Latency vs Internet Speed

Internet speed and latency measure different things.

Speed describes capacity, while latency describes delay.

A connection can therefore have very high bandwidth and still have high latency.

For example, a satellite connection may provide hundreds of Mbps while still having more delay than Fiber.

See Mbps vs Gbps for more about bandwidth measurements.

Latency vs Upload & Download Speed

Upload and download speeds determine how much data can move through the connection, while latency determines how quickly individual exchanges occur.

Different activities emphasize different measurements.

Does Faster Internet Lower Latency?

Not necessarily.

Moving from a 100 Mbps plan to a 1 Gbps plan increases bandwidth, but it does not automatically reduce the physical or routing delay between your device and a server.

A faster plan can reduce latency indirectly if the existing connection regularly becomes congested. When a network is overloaded, packets may have to wait in queues before being transmitted.

In that situation, additional bandwidth can reduce congestion-related delay.

However, factors such as connection technology, physical distance, routing, Wi-Fi, and server location still determine much of the latency.

What Causes Internet Latency?

Latency is caused by several types of delay that accumulate as data travels through a network.

Common causes include:

• Physical distance
• Connection technology
• Network routing
• Router processing
• Network congestion
• Wi-Fi interference
• Server response time
• Packet retransmissions
• VPN routing
• Buffering and queueing

Physical Distance & Latency

Distance is one of the most fundamental causes of latency.

Data cannot travel instantaneously. Even signals moving through fiber-optic cable are limited by the speed at which light travels through the material.

A server located nearby will generally have lower network latency than a server located thousands of miles away, assuming other conditions are similar.

This is why many online services use geographically distributed data centers and content delivery networks to place content closer to users.

Network Routing & Latency

Internet traffic rarely travels directly from your home to the destination server.

It may pass through several routers and networks along the way.

The route depends on:

• Your Internet provider
• Peering agreements
• Backbone networks
• Regional Internet exchanges
• Congestion
• The destination network

A geographically nearby server can sometimes have unexpectedly high latency if the network route is inefficient.

Network Congestion & Latency

When a network becomes busy, packets may wait in queues before they can be transmitted.

This can increase latency even if the underlying connection is normally fast.

Congestion can occur:

• Inside the home network
• On the Internet provider’s network
• At network interconnection points
• At the destination server

Latency caused by overloaded queues is sometimes associated with a problem known as bufferbloat.

What Is Bufferbloat?

Bufferbloat occurs when network equipment holds too much data in queues during periods of heavy traffic.

This can cause latency to rise dramatically when the connection is busy.

For example, your Internet may show:

• 15 ms latency while idle
• 150 ms latency during a large upload

The connection still has bandwidth, but packets used by gaming, voice calls, or browsing are forced to wait behind large amounts of other traffic.

Modern queue-management technologies can reduce this problem.

What Is Loaded Latency?

Loaded latency measures latency while the Internet connection is actively transferring substantial amounts of data.

This can be more revealing than idle latency because it shows how responsive the connection remains during actual household use.

A connection may have excellent idle latency but perform poorly when someone starts:

• Uploading a large file
• Downloading a game
• Running a cloud backup
• Using most of the available bandwidth

FCC broadband measurements have shown that latency can rise under heavy downstream or upstream load, particularly on some DSL connections.

What Is Jitter?

Jitter is variation in latency over time.

If several packets have round-trip times of:

• 20 ms
• 21 ms
• 19 ms
• 20 ms

the connection has very little jitter.

If the results look more like:

• 20 ms
• 85 ms
• 35 ms
• 140 ms

the connection has much greater latency variation.

High jitter can cause:

• Choppy voice calls
• Video-call instability
• Gaming lag
• Audio dropouts
• Uneven remote desktop performance

What Is Packet Loss?

Packet loss occurs when some data packets fail to reach their destination.

Networks can often retransmit lost data, but this takes additional time.

Real-time applications may not have enough time to wait for retransmission.

Packet loss can result in:

• Missing audio
• Frozen video
• Gaming stutters
• Disconnected sessions
• Slow transfers

The FCC notes that packet loss around 1% can already significantly affect highly interactive applications such as VoIP.

Latency vs Jitter vs Packet Loss

A good Internet connection should ideally have low latency, low jitter, and very little packet loss.

Latency for Online Gaming

Gaming is one of the activities most sensitive to latency.

In fast-paced games, your actions must reach the server and updated information must return quickly.

Lower latency can improve:

• Input responsiveness
• Movement synchronization
• Hit registration
• Competitive timing

Bandwidth requirements for actual gameplay are usually modest.

This is why upgrading from 300 Mbps to 1 Gbps may not improve a game if latency remains unchanged.

What Is a Good Ping for Gaming?

As a general guideline:

• Under 20 ms: Excellent
• 20–50 ms: Very good
• 50–100 ms: Usually playable
• 100–150 ms: Increasingly noticeable delay
• 150 ms+: Often problematic for competitive games

The actual experience depends on the game and server.

A stable 40 ms connection is often preferable to one that fluctuates between 10 ms and 150 ms.

Latency for Video Calls

Video conferencing requires real-time two-way communication.

High latency can cause:

• People talking over each other
• Delayed responses
• Awkward pauses
• Audio and video synchronization problems

Jitter and packet loss can be equally important because unstable delivery can create broken audio or frozen video.

Latency for VoIP Calls

Voice over IP is particularly sensitive to delay because conversations depend on rapid back-and-forth communication.

High latency makes conversation feel unnatural and can cause participants to interrupt each other unintentionally.

Packet loss and jitter may cause audio to become distorted or incomplete.

Latency for Streaming Video

Streaming services are much less sensitive to latency than gaming or voice calls.

Video platforms buffer content ahead of playback, which hides much of the network delay.

For streaming, download bandwidth and connection stability are generally more important than very low latency.

Latency may still affect how quickly playback begins or how responsive menus and controls feel.

Latency for Remote Work

Remote work can involve several applications with different sensitivity to latency.

Latency matters particularly for:

• Video meetings
• VoIP calls
• Remote desktop
• Virtual desktops
• Interactive cloud software

Large file transfers depend more heavily on bandwidth than latency.

See How Much Internet Speed Do I Need? for a broader look at remote-work requirements.

Latency on Fiber Internet

Fiber Internet generally provides very low latency.

In the FCC’s recent fixed-broadband measurements, sampled Fiber providers had median idle latency of approximately 7–14 ms, the lowest range among the Fiber, cable, and DSL services measured.

Actual Fiber latency still depends on:

• Provider routing
• Server location
• Home networking
• Congestion
• Wi-Fi

Latency on Cable Internet

Cable Internet generally provides low latency suitable for gaming, video calls, and other interactive applications.

The FCC’s sampled cable providers showed median idle latency in roughly the 12–24 ms range.

Performance can change during congestion or heavy household traffic.

Latency on DSL Internet

DSL Internet typically has higher latency than Fiber and somewhat higher latency than many cable connections.

The FCC’s sampled DSL services showed median idle latency of approximately 23–34 ms.

DSL connections can also experience substantial latency increases when the limited upstream or downstream capacity becomes saturated.

Latency on Fixed Wireless & 5G Home Internet

Fixed Wireless and 5G Home Internet can provide good latency, but performance varies with:

• Signal conditions
• Distance from the tower
• Spectrum
• Network congestion
• Wireless scheduling
• Network architecture

Latency may therefore change more from moment to moment than on a stable wired connection.

Latency on Satellite Internet

Satellite Internet varies dramatically depending on the satellite orbit.

Low Earth orbit satellite systems can provide latency in the tens of milliseconds.

Traditional geostationary satellites orbit much farther from Earth, creating much higher unavoidable delay because signals must travel enormous distances.

This physical distance is why geostationary satellite service has historically been poorly suited to highly interactive applications.

Does Wi-Fi Increase Latency?

It can.

Wi-Fi introduces another wireless link between your device and router.

A strong modern Wi-Fi connection may add very little delay, but latency can increase because of:

• Weak signal
• Interference
• Network congestion
• Many active devices
• Distance from the router
• Retransmissions
• Older Wi-Fi equipment

For applications where consistency matters, Ethernet can often provide lower and more predictable local-network latency.

Ethernet vs Wi-Fi Latency

Ethernet provides a wired connection between the device and router.

Compared with Wi-Fi, Ethernet avoids:

• Radio interference
• Signal attenuation through walls
• Wireless contention
• Many retransmission problems

This is why competitive gamers and other latency-sensitive users often prefer Ethernet where practical.

Can a Router Cause High Latency?

Yes.

A router can contribute to high or unstable latency if it is:

• Overloaded
• Outdated
• Running old firmware
• Handling more traffic than it can process efficiently
• Using poor queue management

The equipment supplied by the provider may also combine several functions into one gateway.

See Modem vs Router for more about how home networking equipment works.

Can a Modem Cause High Latency?

Yes, depending on the Internet technology.

A modem or gateway can contribute to poor performance if it is outdated, incompatible with the provider’s current network, malfunctioning, or overloaded.

On Cable Internet, using appropriate DOCSIS equipment is important for getting the performance supported by the provider.

Fiber connections generally use an ONT instead of a traditional cable or DSL modem.

Can VPNs Increase Latency?

Yes.

A VPN sends traffic through an additional VPN server before it reaches the final destination.

This can increase latency because:

• The route may become longer
• The VPN server must process encrypted traffic
• The chosen VPN server may be far away
• The server may be congested

A nearby, lightly loaded VPN server may add relatively little delay, while a distant server can add much more.

Can DNS Affect Latency?

DNS does not normally change the ongoing latency of an established connection, but slow DNS resolution can delay the beginning of a website request.

When you enter a domain name, DNS helps find the server’s IP address.

Once that lookup is complete, the actual network latency depends on the path between your device and the destination.

Can Server Location Affect Ping?

Yes.

Server location can have a major effect on latency.

A gamer in California will generally have lower latency to a server on the U.S. West Coast than to a server in Europe.

This is why many games and cloud applications allow users to choose a region or automatically connect to a nearby server.

Why Does Ping Change During the Day?

Latency can vary throughout the day because network conditions change.

Possible causes include:

• Peak-hour congestion
• Wireless network load
• Household traffic
• Provider routing changes
• Server congestion
• Wi-Fi interference

If latency rises only when the connection is heavily used, loaded latency or bufferbloat may be contributing to the problem.

Why Is My Ping High When Uploading?

Uploading can increase latency if the upstream connection becomes saturated.

This is especially common on services with limited upload capacity.

For example, a household may have:

• 500 Mbps download
• 20 Mbps upload

A cloud backup consuming nearly all 20 Mbps of upload capacity can create queues that delay gaming, web browsing, and voice traffic.

This is one reason strong upload performance can improve the overall experience even for activities that do not use much upload bandwidth themselves.

How to Test Internet Latency

Many Internet speed tests report latency along with download and upload performance.

A useful test may include:

• Idle latency
• Download speed
• Upload speed
• Loaded latency
• Jitter
• Packet loss

For troubleshooting, test several times and compare:

• Ethernet vs Wi-Fi
• Different times of day
• Idle vs heavily loaded conditions
• Different servers

How to Test Ping Manually

Most operating systems include a command-line ping utility.

A ping test can show:

• Round-trip latency
• Minimum latency
• Maximum latency
• Average latency
• Packet loss

Results can differ depending on the destination being tested, so pinging one server does not describe the latency to every service on the Internet.

How to Reduce Internet Latency

Some latency is unavoidable, but several steps may improve a poor connection.

• Use Ethernet instead of Wi-Fi for latency-sensitive devices
• Move closer to the Wi-Fi router
• Reduce wireless interference
• Stop unnecessary uploads and downloads
• Pause cloud backups during gaming or calls
• Update router firmware
• Replace outdated networking equipment
• Choose nearby game or application servers
• Use a closer VPN server or disable the VPN when appropriate
• Consider a lower-latency Internet technology if available

Will Buying More Internet Speed Reduce Latency?

Only in some situations.

If the connection becomes congested because there is not enough bandwidth, upgrading can reduce queueing delays.

If latency is caused by:

• Physical distance
• Network routing
• Satellite orbit
• Wi-Fi interference
• Server location

buying a faster speed tier may make little difference.

See How Much Internet Speed Do I Need? before assuming that a more expensive plan will solve a latency problem.

Latency Frequently Asked Questions

What is Internet latency?

Internet latency is the amount of time required for data to travel across a network and receive a response. It is usually measured in milliseconds.

Is latency the same as ping?

Not exactly, but the terms are commonly used interchangeably. Ping is a test that measures round-trip network delay, while latency is the underlying delay itself.

What is a good ping?

Under 20 ms is excellent, 20–50 ms is very good, and under 100 ms is generally usable for most applications. The ideal number depends on what you are doing online.

Is 10 ms ping good?

Yes. 10 ms is an excellent latency result for consumer Internet.

Is 30 ms ping good?

Yes. 30 ms is very good for gaming, video calls, browsing, and most other activities.

Is 50 ms ping good?

Yes. 50 ms is generally considered good and should work well for most applications.

Is 100 ms ping bad?

It is usable, but delay can become noticeable in gaming, voice calls, remote desktop sessions, and other real-time applications.

Is 200 ms ping bad?

200 ms is high for interactive use and can cause obvious delay in gaming and live communication.

Does higher Mbps mean lower ping?

No. Mbps measures bandwidth, while ping measures latency. More bandwidth may help if congestion is causing delay, but it does not automatically reduce ping.

Why is my ping high even with Gigabit Internet?

Possible causes include Wi-Fi, routing, server distance, congestion, VPN use, bufferbloat, or the destination server itself.

Does Fiber have lower latency?

Fiber generally provides very low latency. Recent FCC measurements found lower median idle latency for sampled Fiber providers than for the sampled cable and DSL providers.

Does Wi-Fi increase ping?

It can. A strong Wi-Fi connection may add very little delay, but interference, distance, congestion, and retransmissions can increase latency and jitter.

Does Ethernet lower ping?

Ethernet can reduce local-network delay and make latency more consistent by avoiding wireless interference and contention.

Does a VPN increase ping?

Usually. A VPN adds an extra network route and processing step, although the amount of added latency depends on the VPN server and network path.

Does satellite Internet have high latency?

Traditional geostationary satellite service has high latency because signals must travel tens of thousands of miles. Low Earth orbit systems have much lower latency.

What is jitter?

Jitter is variation in latency from one packet to another. High jitter can cause unstable gaming, voice, and video performance.

What is packet loss?

Packet loss occurs when data packets fail to reach their destination. This can cause dropouts, retransmissions, frozen video, and gaming problems.

What is loaded latency?

Loaded latency measures network delay while the connection is actively transferring substantial amounts of data.

What is bufferbloat?

Bufferbloat is excessive delay caused by large network queues, often becoming noticeable when uploads or downloads saturate a connection.

Why does my ping rise when someone downloads a game?

The download may create congestion or queueing in the router or Internet connection, increasing loaded latency for other users.

Learn More About Internet Speed & Performance

Latency is only one part of Internet performance. Bandwidth, Wi-Fi quality, equipment, packet loss, and connection technology all affect the experience.

• How Much Internet Speed Do I Need? – Choose a practical speed tier based on household use.
• Mbps vs Gbps – Learn how Internet bandwidth is measured.
• Upload vs Download Speed – Understand downstream and upstream bandwidth.
• Modem vs Router – Learn how home networking equipment can affect performance.
• What Is Wi-Fi? – Learn how wireless conditions can affect latency, speed, and stability.
• What Is Fiber Internet? – Learn why Fiber generally provides low latency and high capacity.
• What Is Cable Internet? – Learn how modern cable broadband performs.
• What Is DSL Internet? – Learn how DSL performance differs from newer broadband technologies.
• Fixed Wireless & 5G Home Internet – Learn how wireless conditions affect home broadband.
• What Is Satellite Internet? – Compare low Earth orbit and geostationary satellite latency.

Why Low Latency Matters

Latency determines how quickly an Internet connection responds, while Mbps and Gbps determine how much data it can carry. Both measurements matter, but they affect online activities in different ways.

Streaming and large downloads benefit mainly from bandwidth. Gaming, voice calls, video conferencing, remote desktop, and other interactive applications depend much more heavily on low and consistent latency.

A good connection therefore is not simply the one with the largest advertised speed. Low latency, low jitter, minimal packet loss, sufficient upload and download capacity, and a strong local network all contribute to overall performance.

Latency varies by Internet technology, provider network, routing, server location, Wi-Fi conditions, congestion, customer equipment, VPN use, and other factors. The latency ranges in this guide are general reference points rather than guaranteed thresholds for every application.