Understanding Your Device's Network Connection

Your connection is one of the biggest factors in how the web feels, so browsers offer a small window into it through the Network Information API. It is genuinely useful for adapting a site — loading lighter images on a slow link, for instance — but it is important to know what these numbers are, and what your browser fundamentally cannot see about your network.
What the Network Information API reports
Where it is available, the API exposes four things. The effective connection type is a class — "4g", "3g", "2g" or "slow-2g" — that summarises overall quality rather than the literal radio technology, so good home Wi-Fi typically reports as "4g". Downlink is an estimated bandwidth in megabits per second. Round-trip time (RTT) is an estimate of latency in milliseconds. And the data-saver flag tells a site whether you have asked the browser to economise on data.
Together these let a site make sensible choices without measuring anything itself. DeviceScope simply reads and displays them on the Network panel, updating if your connection changes while the page is open.
Why they are estimates, and not everywhere
These values are the browser's rolling estimates based on recent traffic, not a live speed test. Downlink is typically capped and rounded for privacy, so you will not see a precise figure, and it reflects recent conditions rather than the maximum your line could achieve. If you want a true measurement, run a dedicated speed test that actually transfers data.
Support is also uneven. The Network Information API is available in Chromium-based browsers but is largely absent in Safari and Firefox, which decline to expose it for privacy reasons. When your browser withholds it, DeviceScope shows the fields as not available rather than filling in a plausible-looking guess.
Online, offline and what navigator.onLine really means
Alongside the connection estimates sits a far simpler signal: navigator.onLine, together with the online and offline events that fire when it flips. It is useful, but it answers a narrower question than most people assume. A true value means only that the browser has a network interface it believes is usable — an association with a Wi-Fi network, a live Ethernet link, a cellular data connection. It does not mean packets are actually reaching the internet.
So you can sit on an unpaid captive-portal hotspot, or behind a router whose upstream link has died, and still read true. The false value is the more trustworthy half of the pair: when the browser says you are offline, you almost certainly are. If a page needs to know whether one particular service is reachable, the only dependable test is to request something from it and see what comes back — which is precisely how offline-capable apps and service workers are written.
What your browser cannot tell you: the public IP
Here is the honest limit. Your public IP address, your ISP's name and whether you are behind a VPN are all properties of the network path to a server — they only become known when you actually connect to one. A page running purely in your browser has no server of its own, so it cannot discover any of them. That is why DeviceScope, which has no backend, never displays your IP: showing it would mean routing your traffic through someone else's server, which is exactly the privacy cost the tool avoids.
The same logic explains why a local page cannot name your ISP, place you by IP, or tell whether a VPN is active: each of those is something a remote server infers from the address it can see, not something your browser is handed. The one geographic hint you do get is your time zone, and it comes from your operating-system settings rather than from the network — which is why it keeps pointing at your real region even while a VPN is switched on.
Your local network is off-limits too
The public IP is not the only network fact a page cannot help itself to. Devices on your local network — your router's admin page, a printer, a network drive, anything living on 192.168.x.x or on localhost — were once reachable from any website's JavaScript, which made the browser a convenient way to probe and attack home equipment from outside. Private Network Access closes that door: before a page from the public internet may send a request to a private or local address, the browser issues a preflight and the target device has to opt in with a specific response header. Consumer hardware almost never does.
The practical effect is that a web page cannot enumerate what sits on your network, read your router's model, or find your local addresses by scanning for them. WebRTC once leaked local IP addresses as a side effect of setting up a connection, and browsers now hide them behind randomised mDNS hostnames instead. Taken together with the public-IP limit, this means a purely client-side tool can describe the quality of your connection but never its topology or its identity.
The takeaway
The browser offers an estimated connection class, downlink and latency through the Network Information API, but it is a rounded hint, missing in some browsers, and never includes your public IP. See what yours exposes on the Network panel, and use a real speed test for exact figures.
Effective connection type, downlink, latency and data-saver mode.
More guides

What Your Browser Reveals About You (and What It Doesn't)
A web page can read a surprising amount about your setup with no permission at all — and there are hard limits it cannot cross. Here is exactly where the line sits.

Screen Resolution, DPR and Retina Displays Explained
Why the resolution your browser reports may not match the pixels in the panel, what device pixel ratio really means, and why "Retina" is just marketing for a number you can read yourself.

What Is a User Agent String?
Every request your browser sends carries a strange line of text claiming to be several browsers at once. Here is what it means, why it is full of historical fibs, and why it is slowly being retired.