This scrolling smoothness test measures the kind of motion you meet more than any other: a page or list moving under your finger or wheel. Scrolling is where jank shows up first, because it happens constantly and your eyes are practised at reading it. A page that hitches as it moves, a feed that skips as you flick it, a document that judders line by line — all of it comes down to whether the browser can deliver a fresh frame on every refresh while the content is in motion. This test scrolls a tall band of numbered rows at a fixed step per frame and counts the frames that arrive late.
Because the rows carry clear numbers and hard edges, your eyes can judge the smoothness directly, while the meter puts a figure on it. That lets you compare browsers, extensions, and machines with data instead of a vague impression.
The band scrolls on its own as soon as the page loads. To read it well:
Use Reset meter to start a fresh count. Leaving the test running for thirty seconds gives a more honest picture than a quick glance, because background hitches come and go.
Smooth scrolling asks the browser to produce a new frame every refresh — every 16.7 milliseconds at 60 Hz, every 8.3 at 120 Hz — while also running the page. When anything delays that work, a frame is missed and the scroll jumps. The usual culprits are specific and fixable.
Large images, complex layouts, and busy scripts make each scrolled frame expensive to draw. Pages that run code on every scroll event are especially prone to jank. This is a property of the page as much as your machine, which is why some sites scroll smoothly and others do not on the same computer.
Content-blocking, shopping, and accessibility extensions inspect the page as it moves. A single poorly optimised extension can add just enough work per frame to cause visible stutter. Testing in a private window, where most extensions are disabled, is a quick way to check.
Browsers try to hand scrolling to the GPU so it stays smooth even when the page is busy. When hardware acceleration is off or a page forces work back onto the main thread, scrolling falls back to the slower path and janks. Turning hardware acceleration on in your browser settings usually helps.
As with any motion, a laptop on battery saver, a hot device throttling itself, or a background task stealing CPU time will all make frames arrive late. The scrolling test reflects the same system health that governs every other kind of motion.
Healthy scrolling is a high, steady frame rate with a jank count near zero and a worst frame gap close to the expected frame time. The table below shows the target frame time for common refresh rates — the value the test should hover around while scrolling.
| Refresh rate | Target frame time | Smooth scroll budget per frame |
|---|---|---|
| 60 Hz | 16.7 ms | Everything must finish within 16.7 ms |
| 120 Hz | 8.3 ms | Half the time, twice the smoothness |
| 144 Hz | 6.9 ms | Tighter budget, needs a capable page |
A higher refresh rate makes scrolling look noticeably smoother, but only if the browser can keep filling that tighter budget. Confirm the refresh rate you are actually running with our refresh rate guide, and measure your raw frame rate on the main FPS test.
If scrolling is janky here but a game runs smoothly, the cause is usually the browser or the page rather than your hardware. If motion is rough everywhere, measure it in isolation with the stutter test, which uses the same frame-time meter without the scrolling content.
Because smoothness depends on the page as well as your machine. Heavy scripts, large images, and scroll-event handlers make some sites expensive to scroll. A clean result here means your browser and hardware are capable; the busy site is the bottleneck.
Yes, provided the browser can deliver a frame every refresh. A 120 Hz screen glides more smoothly than a 60 Hz one, but only while the page keeps filling the shorter frame budget. If it cannot, you simply see the jank sooner.
Browsers differ in how they composite and hand scrolling to the GPU, and in how extensions hook into the page. It is common for one browser to scroll a page smoothly while another janks on identical hardware.
They share the frame-time meter, but this test drives real scrolling content through the browser's compositor, which is the everyday case most people care about. The stutter test isolates raw frame timing without the scrolling layer.
Mostly your browser and the machine it runs on. The rows move through the browser's rendering and compositing path, so a poor result usually points to the browser, an extension, or system load rather than the panel. Your display still sets the ceiling — a 60 Hz screen can never scroll as smoothly as a 120 Hz one — but assuming the refresh rate is adequate, this test is judging how well the software keeps up.
Most browsers and operating systems offer a smooth-scrolling option that animates each scroll step instead of jumping. It makes motion look gentler, but it does not create frames the browser cannot draw — if the page is heavy, smooth scrolling simply animates the jank. The real gain comes from a high refresh rate paired with a page light enough to keep up. On a 120 Hz screen, scrolling a simple page is visibly smoother than on 60 Hz, because twice as many frames describe the same motion. This test lets you confirm your browser is genuinely delivering those frames rather than dropping half of them.
How you scroll changes what you feel. A mouse wheel moves in discrete steps, so smoothness depends on how well the browser animates between them. A trackpad or touchscreen scrolls continuously and is handed to the compositor, which is usually the smoothest path — but also the most obvious when a frame is dropped, because your finger sets an expectation the screen must match. Phones and tablets are especially sensitive here, since scrolling is the primary way you move through everything. If a page janks under your finger but not under a wheel, the compositor is being forced to wait on the main thread.
Smooth scrolling is a chain, and the weakest link sets the result: the page must be light enough to draw in time, the browser must hand scrolling to the GPU, the system must not be busy or throttling, and the display must have the refresh headroom to show the frames. This test exercises the whole chain at once. When the number is good, every link is holding; when it is poor, the sections above narrow down which link to look at first. Because scrolling is so common, small improvements pay off everywhere — a browser update, a disabled extension, or a switch to hardware acceleration can lift a janky result to a clean one, and you will feel the difference on every site you visit afterwards.
Scrolling is one everyday slice of motion performance. Continue with these companion tools: