Screen Tearing Test

How to use

1. Read the Simulated tear reference first. It shows an aligned hard vertical edge beside a deliberately offset lower region; it is not a test result.
2. Choose Parallel bars or Moving line, a browser-drawn speed, Left or Right direction, and a Dark or Light background. The speed is shown in CSS pixels per second for this visual surface.
3. Optionally enter browser Fullscreen before Start to enlarge the observation surface. It is not game-exclusive fullscreen and does not bypass the operating-system compositor.
4. Click Start in the stable control strip, keep this tab visible and focused, and watch repeated motion long enough to inspect the moving hard edge or bars. Settings and start geometry stay fixed for the run.
5. Look for a repeated horizontal spatial discontinuity between regions of the pattern. Also note motion that looks uneven without a clear split; that is closer to stutter or judder.
6. Click Stop, or Stop & exit fullscreen, before changing a pattern, speed, direction, background, or canvas geometry. Start a new run after each change instead of mixing conditions.
7. Record an observation only after an explicit clean Stop. Fullscreen exit, resize, hidden, blurred, or reduced-motion interruptions must be repeated.
8. If a tear-like split repeats, compare it in the actual game or app with the same VSync, frame-cap, VRR, and fullscreen settings. This page cannot reproduce that presentation path.
9. Use Frame Skipping Test for uneven presentation intervals, Refresh Rate Test for browser timing, and Monitor Ghosting Test for trailing or smearing.
10. No camera or microphone permission is required. The patterns and your selected observation stay local; the page does not capture or upload your screen.

What the result means

• requestAnimationFrame normally participates in browser repaint and compositor scheduling. This makes the moving pattern useful for direct observation of the current browser path, not a panel timing readout.
• No obvious tear-like split noticed means none was obvious in this run’s browser, compositor, display mode, pattern, speed, direction, window, and fullscreen conditions. It does not prove VSync, VRR, G-SYNC, FreeSync, or a game path is synchronized.
• Possible horizontal tear-like split means you saw a horizontal spatial discontinuity in the browser-generated pattern. Browser compositing, display-path behavior, capture artifacts, and perception can look similar; repeat before comparing elsewhere.
• Uneven / stuttered motion means cadence looked irregular without a clear spatial split. Use Frame Skipping Test or FPS Test when timing or workload is the main question.
• Unsure / repeat means the visual evidence was not clear enough to label. Repeat with the same settings and focused window before changing conditions.
• The simulated tear reference is a static teaching illustration: the lower region is horizontally offset across one visible split. It does not simulate frame timing, verify VSync, or produce pass/fail output.
• The speed presets describe browser-drawn visual motion in px/s. They are not panel scan speed, GPU speed, FPS, refresh rate, or tear-line speed, and the presets are not industry thresholds.
• A positive observation is a reason to investigate the current presentation path and, if relevant, reproduce the issue in the real game or app. It does not identify a root cause or certify a display.
• A negative observation is only evidence about this local browser run. A normal-looking browser path cannot prove that another workload, driver mode, VRR mode, or exclusive game path will be tear-free.
• A normal browser compositor may prevent visible real tearing even when another application or game can tear. This page cannot force VSYNC off or reproduce an exclusive-game presentation path.
• Browser Fullscreen is optional setup before Start. It gives the visual surface more space but remains browser fullscreen, not exclusive fullscreen or a compositor bypass. Exiting it during a run interrupts that run.
• Camera recordings can show rolling-shutter bands, exposure effects, or capture cadence. Direct visual observation is primary here; DeviceMetrik does not request, analyze, or require a camera recording.
• Screen tearing, frame skipping, stutter or judder, and monitor ghosting are different visual intents: a split is spatial, a gap is temporal, uneven motion is cadence, and ghosting is a trail or smear.

Limitations

• This is a visual browser-path observation, not automatic tear detection or a hardware, VSync, VRR, G-SYNC, FreeSync, or game verdict.
• requestAnimationFrame, browser compositor behavior, window mode, display path, capture, and perception can all affect what you see.

Troubleshooting

Nothing happens after Start

Keep the tab visible and focused, check whether Reduced Motion is enabled, then Reset and try again in a supported browser. No device permission is required.

No tear-like split is visible

That can be normal for the current browser compositor and display path. Repeat with one other pattern, speed, or direction, but do not treat a negative result as proof about VSync, VRR, or a game.

The motion looks uneven without a split

Record Uneven / stuttered motion only after a clean run, then use Frame Skipping Test or FPS Test for browser timing and workload context.

A line appears only in a phone video or photo

Rolling shutter, exposure, recording frame rate, and capture timing can create bands or discontinuities. Compare direct observation with the recording; this page does not capture or analyze it.

The result changes after switching tabs or windows

Background tabs can be throttled and blur ends a clean run. Reset and repeat in the foreground with the window focused.

The result changes by speed or direction

Compare repeated runs with the same pattern, speed, direction, background, window, and fullscreen state. The browser-drawn presets are observation conditions, not hardware thresholds.

Fullscreen is unavailable or exits

The windowed canvas still works. Enter Fullscreen only before Start. If it exits during an active run, repeat and use the in-fullscreen Stop & exit fullscreen control for a clean result.

The issue occurs in a game but not here

Use this page only as browser-path context. Reproduce the game issue in that game with the same VSync, frame-cap, VRR, and fullscreen settings; this page cannot inspect or control them.

Related tools

• Frame Skipping Test

  Run a browser timing preflight, then photograph a discrete pattern for possible frame-skipping gaps. Review long frames, P95/P99, and stability—not a hardware verdict.

• Refresh Rate Test

  Estimate the refresh rate reaching this browser window from requestAnimationFrame timing. Review median frame interval, timing consistency, interruptions, and repeatable local results without treating browser Hz as panel telemetry.

• Monitor Ghosting Test — Motion Trails & Inverse Ghosting

  Track high-contrast targets at 480, 960, or 1920 px/s to look for possible trails, inverse halos, dark smearing, color fringes, and motion blur. Visual browser test only—not GtG or MPRT measurement.

FAQ