A better connection
for the work you do.

01 / At a glance

Two ways to reach the network.

These describe common designs, not a guarantee for every device or service. Both should be secured, monitored, and tested in the environment where people use them.

02 / The wired connection

What does a cable change?

In a typical modern switched Ethernet network, each device has its own physical link to a switch. A full-duplex link can send and receive at the same time. The connection does not compete with nearby Wi-Fi networks for radio airtime, which generally makes its local performance easier to predict.

A port’s speed is not a promise that every application will run at that rate. An older 100 Mbps adapter, an incompatible cable or optic, a congested uplink, firewall limits, storage, or the destination can constrain performance. For example, a device connected at 100 Mbps cannot obtain 1 Gbps of useful throughput through that link, even if the office Internet service is faster.

Wiring requires suitable cabling, ports, installation, and sometimes adapters or docks. Cable length and category, port capability, and the installation environment matter. Confirm the specifications before planning an upgrade.

03 / The wireless connection

Why can Wi-Fi change during the day?

Wi-Fi uses radio channels. Access points and client devices share limited radio resources, including airtime. As more devices actively send or receive data, they can compete for those resources. Modern Wi-Fi can coordinate simultaneous users, but its radio capacity is still finite.

Several factors affect the usable connection:

• Concurrent work. Video meetings, large transfers, backups, and software updates can use airtime at the same time. The number of connected devices alone does not describe their demand.
• Signal and placement. Distance, walls, floors, furniture, and access-point placement can weaken a connection or make it less consistent.
• Other radio traffic. Nearby networks using the same or overlapping channels and non-Wi-Fi devices can compete with or interfere with wireless communication.
• Device capabilities. Access-point design, client adapters, supported bands, channel width, and software affect the connection each user can achieve.
• Network beyond the radio. An access point’s wired uplink, switches, firewall, or Internet service may be the bottleneck instead.

A strong signal indicator does not by itself show how busy a channel is or whether the path beyond the access point is healthy. The access-point dashboard and a wireless site survey can help distinguish coverage, interference, and capacity issues.

04 / In a busy office

What happens when everyone gets online?

Imagine an office where several employees join video meetings while others upload design files and synchronize cloud storage. If they share a busy wireless channel, each active device needs a share of the available radio resources. Delays or retries can grow, even when the Internet circuit itself has capacity available.

New York City offices may be surrounded by other businesses and wireless networks. Depending on the building, channels, and traffic, this density can make wireless planning harder. A site survey and measurements can establish the conditions in a particular office; an address alone cannot.

Connecting fixed workstations, conference-room systems, and other high-usage devices by Ethernet can move their network traffic off Wi-Fi and leave more radio capacity for mobile devices. How much that helps depends on the traffic, access-point design, and other users of the channel.

05 / Beyond the office

The Internet connection is a separate part of the path.

Both wired and wireless users typically share switches, a router or firewall, and the business Internet service. If the Internet circuit or another shared device is congested, connecting a laptop by cable may not fix the issue. Similarly, an underperforming Wi-Fi connection can limit a device even when the Internet circuit is working well.

Measure busy-period upload and download separately. Include simultaneous meetings, cloud applications, remote access, and large transfers, then check that network equipment can carry the intended load. Our business bandwidth guide explains how to estimate capacity.

06 / Packets & delay

What happens to a packet?

Whether an employee sends a message, joins a video meeting, or opens a cloud application, the device sends information in small pieces called packets. Those packets move through the local network and, when needed, across the Internet to the service. Responses travel back along a path that may differ.

Latency is the time taken across a measured path; jitter is variation in that delay. When a wireless channel is busy, waiting and retries can add delay and variation. This can be noticeable in calls, interactive applications, and other time-sensitive work. Conditions differ by device, access point, workload, and location, so measure the actual connection rather than assuming a fixed penalty.

A wired connection can also experience queueing, packet loss, or delay when a switch, uplink, firewall, or Internet connection is congested. Wi-Fi and Ethernet both depend on the networks and services beyond the office. The difference is not that radio signals travel slowly; it is how each connection shares resources and handles traffic in the environment.

For more on measuring delay and interpreting results, see latency, jitter, and packet loss.

07 / Measure the difference

Compare the paths fairly.

1. 01

   Choose a representative device

   Use a capable laptop or workstation and, for the wired test, an Ethernet adapter and port that support the intended speed. Confirm which network is active for each test.

2. 02

   Test under similar conditions

   Compare wired and Wi-Fi from the same location, around the same time, to the same appropriate destination. Record the access point or Wi-Fi band, link speeds, time, and other traffic when known. Repeat during a representative busy period.

3. 03

   Look beyond one number

   Measure upload and download, latency, and the actual applications employees use. A speed test reflects its test server, device, and path; it does not guarantee the behavior of every service.

4. 04

   Investigate the difference

   If Wi-Fi alone is affected, examine signal quality, channel utilization, interference, client capabilities, and access-point uplinks. If both paths are affected, also investigate shared equipment, the Internet circuit, and the destination.

Run high-throughput tests only when appropriate and coordinated, because they can consume substantial capacity. See our speed testing guide and latency and packet loss guide for tools and interpretation.

08 / Put it into practice

Plan for people and the work they do.

• Which fixed devices depend most on consistent performance, and can they be wired?
• How many users are active at once, and what applications and transfers overlap?
• Do adapters, switch ports, cabling, and uplinks support the intended speeds?
• Where do employees need Wi-Fi coverage, and what do channel and site measurements show?
• Can the access-point uplinks, firewall, and Internet service handle the combined traffic?
• Are guest access, security policies, and network monitoring planned for both wired and wireless devices?
• When will the team measure performance again as the office changes?

For help with an existing Stealth service, contact support@stealth.net or call +1 212 232 2020.