Is my fingerprint sent or stored anywhere?

No. Every value is read from the browser APIs in the page and rendered on screen. There is no network request, and nothing is written to storage. The point of the page is to show you what a site could collect, not to collect it.

Do these values identify me personally?

Not individually. Nearly everyone shares your colour depth. It is the combination that identifies — each additional attribute narrows the population, and enough of them together produce a value held by you alone or by very few people.

Does blocking cookies stop this?

No, and that is the reason fingerprinting exists. None of these attributes is stored on your device, so there is nothing to clear or refuse. A site reads them fresh on every visit and recognises the combination.

What is the canvas signature?

The tool draws text and an emoji to an off-screen canvas and takes the tail of the resulting image data. The exact pixels depend on your GPU, graphics driver, installed fonts and rasterisation settings, so the value differs between devices that otherwise look identical.

Does private browsing help?

It clears storage, not characteristics. A private window has the same screen size, timezone, CPU count and canvas output as a normal one. It defeats cookie tracking and does almost nothing against fingerprinting.

Browser Fingerprint Test

Overview

Cookie-based tracking requires storing something on your device, which means you can see it, refuse it and delete it. Fingerprinting requires storing nothing: a site reads properties your browser exposes by design, combines them, and uses the combination as an identifier.

This page shows eleven of those properties. Each one is read with an ordinary browser API, none of them prompts for permission, and none of them is covered by a cookie banner.

How to use

Open the page. The values are read on load. Compare them against another browser on the same machine, or the same browser on a different device, to see which parts identify the machine and which identify the browser.

What each attribute contributes

Attribute	What it narrows
`User-Agent`	Browser, major version, OS family. Increasingly reduced by browsers on purpose
`Language`	Preferred language and region. Strongly identifying outside common locales
`Platform`	OS family. Being frozen or removed in current browsers
`Timezone`	Region. Combined with language, often narrows to one country
`Screen`	Resolution and device pixel ratio. Unusual resolutions are highly identifying
`Color depth`	Almost always 24-bit. Contributes nearly nothing
`CPU cores`	Hardware class. A small set of common values
`Device memory`	Hardware class, coarsely bucketed by the browser
`Touch points`	Distinguishes touch devices from desktops
`Cookies enabled`	One bit, and the answer is almost always yes
`Canvas signature`	GPU, driver, installed fonts and rasterisation. The most identifying single value here

The pattern to notice: the attributes that are trivially readable contribute least, and the one that requires actually rendering something contributes most. That is why canvas, WebGL and audio fingerprinting are what commercial fingerprinting services actually rely on.

How identification from attributes works

Each attribute divides the population. If half of all visitors are on Windows, learning your platform halves the candidates — one bit of information. Colour depth divides almost nobody, so it carries close to zero bits. A canvas signature can divide the population into thousands of groups, so it carries many.

Bits add. Around 33 bits is enough to distinguish one person among roughly 8.6 billion, and a browser exposes considerably more than 33 bits in total — which is why fingerprinting works at all, and why removing any single attribute achieves little.

This also explains a counter-intuitive result: making your browser unusual makes you easier to identify. A rare user agent, an uncommon resolution, or a spoofing extension that reports impossible combinations all add bits rather than removing them. Blending in beats standing out.

What actually reduces it

Ranked by effect, honestly:

The Tor Browser. It normalises the values that leak most — window size is quantised, canvas reads are blocked or prompted, timezone is fixed to UTC. Its users are deliberately identical to each other.
Safari and Firefox with protections on. Both randomise or block canvas and audio reads, and Safari reports a reduced user agent. Neither eliminates the surface, but both remove the highest-value bits.
Nothing else, much. Blocking cookies, private browsing and clearing history all target storage, and none of these values is stored.

Spoofing extensions are worth singling out because they are frequently counter-productive. Reporting a Windows user agent from a machine whose canvas output looks like macOS creates an inconsistency that is itself distinctive.

Examples

Convincing someone that a cookie banner is not the whole story. Show them this page with cookies blocked.
Understanding a bug that only affects some users. Screen, Language and Timezone are the variables behind most “only on my machine” layout and date-formatting reports.
Checking what your privacy setup actually changes. Compare the values with and without your extensions and settings, and notice how many of them do not move.
Explaining why a ban followed you. Fingerprinting is how a site recognises a returning visitor after a new IP and a cleared profile.

Notes

Values that a browser withholds show as — rather than being guessed. deviceMemory is unavailable in Firefox and Safari by design, and platform is being frozen or removed across browsers, so a dash here is the browser protecting you rather than a failure.

The canvas signature is the last sixteen characters of the encoded image data, not a cryptographic hash. It is stable on one device and differs across devices; it is shown as an illustration of the technique rather than as a canonical identifier. A browser that randomises canvas reads will produce a different value on each visit, which is exactly the defence working.

Reading these attributes is unavoidable for legitimate purposes too — responsive layout needs the screen size, date formatting needs the timezone. The problem is not that the APIs exist; it is that reading them for identification is indistinguishable from reading them for rendering.

Two attributes here are read differently from the rest. Canvas signature requires actually rendering, so a browser that blocks or randomises canvas reads changes it; everything else is a plain property read that cannot be intercepted without breaking legitimate uses. That asymmetry is why browser defences target canvas, audio and WebGL rather than trying to hide screen size.

For the network half of the same question, see the IP address checker — an address and a fingerprint are complementary, and a site that has both does not need either to be stable.