Monitor Input Lag Test: Check Your Display Latency Online
Monitor Input Lag Test. Click Start Trial to run this monitor input lag test and measure your display latency plus reaction time. Run the ips glow test tool to check your display for this issue directly in your browser — no download required.
Interactive Tool
Input Lag Test
Click Start Trial, then click or tap the box the instant it turns green. Run several trials — a single click can be thrown off by a slow reaction.
Use the same method, and the same key if using a keyboard, for every trial. Keyboard methods respond to the key only; the other methods respond to a click or tap on the box.
The Monitor Input Lag Test has you click Start Trial, wait for the box to flip green, then click it as fast as you can — timing the gap with performance.now() across several trials and reporting your median and range. Run it more than once, since a single click can be thrown off by a slow reaction, and keep in mind this measures your browser, operating system, and compositor stacked together with your own reflexes — not your display panel's true input lag, which only specialized hardware can isolate.
If your aim feels a step behind or a fighting-game combo drops for no clear reason, this monitor input lag test lets you put a number on the delay right in your browser. You run five timed trials, and you get back a median response time for your whole setup: your own reaction, the input device, the computer and the monitor together. Keep every other setting fixed, change one thing, and the shift in that median shows how much input lag the change added or removed, so a sluggish or unresponsive feel in gaming becomes something you can measure instead of guess.
How the Monitor Input Lag Test Works
The test is a simple wait-and-respond task. A panel sits idle for a random interval, then turns green, and you respond as fast as you can with the input you picked before starting. The page records the time between the colour change being drawn and your response arriving, and repeats that five times. Because the start delay is random, you can't time your clicks by rhythm, and a response before the cue restarts that trial. Run the monitor clouding test online for a few seconds under normal room lighting for the most reliable result.
- Pick one input method (a pointer click, a touch, or one specific keyboard key) and keep it for all five trials.
- Close heavy background work such as downloads, video calls and game launchers, so the browser isn't competing for time.
- Press Start, wait for the green cue, and respond with the same input every time.
- Read the median and the individual trials, then note the setup you tested.
- Change one setting, such as the monitor's picture mode, and run five more trials under otherwise identical conditions.
The result is the median of the five trials, not the average. The median is the middle value once the trials are sorted, so one slow or lucky attempt can't drag it around. Here is a real-world style run on a 165 Hz monitor with a wired mouse: 231, 218, 244, 209 and 226 ms. Sorted, that is 209, 218, 226, 231, 244, so the median is 226 ms. The plain average is 225.6 ms, almost the same here, but a single 400 ms trial (a missed cue) would have pushed the average to about 263 ms while leaving the median at 231 ms.
Look at the spread as well. A gap of 35 ms between the fastest and slowest trial is normal for human responses. If one run swings by 100 ms or more, treat the odd values as outliers and repeat the set when you are rested; a repeatable result is worth more than a single fast one. Don't swap keys mid-run either, since a keyboard key can fire on press while some buttons fire on release, and that alone moves the number.
Where the Milliseconds Go: Latency From Click to Screen
The number you see is never just the monitor. It is the sum of a chain of small delays, and your own reaction is by far the largest link. The rest comes from the input device, the operating system and browser, the wait for the next frame, the monitor's own processing, and the time pixels take to change. Written out: The monitor geometry test tool is free and runs entirely client-side, so nothing about your display is ever uploaded.
$$t_{\text{measured}} = t_{\text{reaction}} + t_{\text{input}} + t_{\text{system}} + t_{\text{frame wait}} + t_{\text{processing}} + t_{\text{pixel}}$$
Only the last two terms belong to the monitor itself. Everything else is shared with the rest of your hardware, which is why a browser check measures the whole path rather than the panel alone. For the 165 Hz example above, here is how the average device delay builds up:
| Stage | Assumption in this example | Average delay |
|---|---|---|
| Input device | Wired pointer polled at 500 Hz (a report every 2 ms) | 1.0 ms |
| Operating system and browser | Event handling with an idle desktop | 3.0 ms |
| Wait for the next frame | 165 Hz, 6.06 ms frame time, half on average | 3.0 ms |
| Monitor signal processing | Fast gaming monitor, no scaling | 3.0 ms |
| Pixel transition | IPS LCD at its normal overdrive setting | 5.0 ms |
| Device chain total | about 15 ms |
Subtract roughly 15 ms from the 226 ms median and about 211 ms is left for the person. That ratio is the key insight: on a good desktop setup the hardware is a small slice of the total, so you only learn something about the monitor by comparing runs where everything else stays fixed. The GPU matters too. When a game is GPU-bound, frames pile up in a render queue before the monitor ever receives a signal, and a low frame rate stretches every link after it. A browser test with simple graphics avoids most of that, which is exactly why game-level delay can be higher than what you measure here.
Human Reaction Time Sets the Floor
Laboratory studies put simple visual human reaction time at roughly 180 to 250 ms, and young adults practised on the task sit near the lower end. Online results usually come out higher than lab figures because the device chain sits on top. Your own number also moves with alertness: the first trial of a session is often the slowest, and fatigue late at night can add 20 ms or more. That is the reason the test uses several trials and a median, and why you should compare runs taken minutes apart, not days apart. A faster reaction time between runs on different days says more about you than about the monitor.
Polling Rate, Debounce and the Input Device Chain
A mouse or keyboard only reports to the computer at its polling rate. At 125 Hz a press can wait up to 8 ms to be reported; at 1,000 Hz the wait is at most 1 ms. Switches also have a short debounce window to filter contact chatter, which adds a millisecond or a few. A wired connection or a proprietary 2.4 GHz USB receiver stays close to that minimum, while a Bluetooth link can add noticeably more. If the goal is to isolate the monitor, use wired peripherals or a low-delay receiver and keep the same one for every run, because a wireless sleep state between trials can add delay to the first press.
Refresh Rate and the Wait for the Next Frame
Your response can only appear on the next frame the monitor draws. The time between frames is the refresh interval, and it follows directly from the refresh rate:
$$t_{\text{frame}} = \frac{1000}{f_{\text{Hz}}}\ \text{ms} \qquad \text{average wait} \approx \frac{t_{\text{frame}}}{2}$$
- At 60 Hz, one frame lasts 16.7 ms, so a change waits 8.3 ms on average.
- At 144 Hz, a frame lasts 6.94 ms and the average wait is 3.47 ms.
- At 165 Hz, a frame lasts 6.06 ms and the average wait is 3.03 ms.
- At 240 Hz, a frame lasts 4.17 ms and the average wait is 2.08 ms.
Going from 60 to 240 Hz trims about 6 ms of average wait. That is real, and it is felt in fast gaming, but it is small next to a 200 ms human response, so a higher refresh number alone won't transform your median.
Display Lag vs Pixel Response Time
Reviewers use the term in a narrower sense than the delay you feel. To them, display lag is only the time from the video signal arriving at the monitor to a visible change starting on the panel. Pixel response time is a separate figure: how long the pixels take to finish moving from one shade to another once that change starts. Both affect how responsive a monitor feels, but mixing them up leads to bad buying decisions, because a monitor with a 1 ms response claim can still buffer a whole frame of image processing before drawing anything.
| Panel type | Typical pixel transition | What to expect |
|---|---|---|
| QD-OLED monitor | Well under 1 ms | Near-instant transitions; processing delay decides the result |
| WOLED monitor | Under 1 ms | Same as above, with slightly different processing per model |
| Fast IPS LCD | About 3 to 6 ms | Good overdrive keeps it inside one frame at high refresh |
| VA LCD | About 5 to 12 ms, slower in dark shades | Dark transitions can smear and trail |
| TN LCD | About 2 to 5 ms | Quick, but narrower viewing angles |
An OLED panel takes pixel transition almost out of the equation, so on an OLED monitor nearly all of the display's share comes from processing. On an LCD, the slower dark transitions of a VA panel can add a visible tail even when the processing delay is short. Processing grows with scaling (feeding a non-native resolution) and with extra picture features.
Two other terms show up in reviews. Click-to-photon measurements, such as NVIDIA's LDAT, time the whole route from a button press to light leaving the panel. AMD's measurement tooling goes further and separates the monitor's share from total system latency. A browser test sits closest to click-to-photon, with your reaction added on top. That is also why one OLED monitor can feel sharper than another OLED monitor with the same refresh: the panel type matches, the processing doesn't.
Ways to Measure End-to-End Delay at Home
A browser check gives you the full end-to-end delay including your reaction, which is ideal for before-and-after comparisons. If you want to measure monitor input lag on its own, you need a method that removes the person from the chain. Here is how the home options compare:
| Method | Cost | What it isolates | Best use |
|---|---|---|---|
| Browser reaction trials | Free | Whole chain plus you | Comparing settings on one setup |
| Timer photo test | Free with a second screen | Difference between two screens | Spotting a frame or more of extra lag |
| SMTT software photo | Low | Difference with fewer timing errors | Comparing modes on the same monitor |
| Hardware sensor or camera | Moderate to high | Signal to light, or click to light | Validating a single monitor |
A spec sheet rarely helps here: manufacturers quote pixel transition, not processing, so the figure you need usually isn't printed anywhere.
Timer Photo Test With a Mirrored Second Display
The classic timer photo test runs a millisecond counter on two screens fed a mirrored image, then photographs both at once. If your monitor trails a known fast reference by about 17 ms across several photos, it is roughly one frame behind at 60 Hz. The method is cheap if you already own a second display, but it is a comparison, not an absolute number. Camera shutter speed, V-sync on either output, and imperfect synchronization between the two outputs all skew single shots, so take at least ten photos and use the middle values.
SMTT-Style Software Comparison
Purpose-built lag-timer software improves on a generic stopwatch: it exists only to show how many ms your monitor trails a known-fast reference in the two-screen photo setup above. It runs without vertical sync limits and renders digits that stay readable in a photo, so the readings repeat better from shot to shot. The result is still only as good as the reference monitor, but it is the strongest way to get the monitor's own share of the delay without dedicated hardware. Reach for it instead of the five-run reaction check when you need that share on its own rather than your whole chain.
High-Speed Camera, Photodiode or Latency Analyzer
For a number you can trust, use hardware. A photodiode taped to the panel, triggered by a device such as the Leo Bodnar lag tester, times signal-to-light directly; this is how lab reviewers produce their figures. A high-speed camera filming a button press and the panel together gives click-to-light timing at 240 to 1,000 frames per second. A dedicated latency analyzer or AMD's Frame Latency Meter logs per-frame timing in DirectX 11 and 12 games to a file, which makes repeated tuning practical. These capture more of the lived experience, so they are ideal for tuning but include delays beyond the monitor.
Reading Your Display Latency Result Against Lag Classes
Once you have a display-only figure from hardware, or a difference between two browser runs, compare it with long-standing review lag classes:
- Under 16 ms
- Less than one frame at 60 Hz (16.67 ms). Suitable for competitive play, including esports-level shooters.
- 16 to 32 ms
- One to two frames. Acceptable for most players and fine for casual and controller games.
- Over 32 ms
- Two frames or more. Delay becomes noticeable, especially in fighting games and rhythm games where timing windows can be a single frame wide.
For browser results, the useful number is the change between runs, converted to frames: divide the difference by the frame time. A 46 ms gap measured on a 60 Hz set is $$\frac{46}{16.67} \approx 2.8\ \text{frames}$$ which lands it in the worst class on its own. Context still matters: a monitor used for spreadsheets and streaming doesn't need the same standard as a 240 Hz esports panel, and moderate lag can be a fair trade for better brightness or colour. For ranked play, responsiveness should win.
Cutting Monitor Processing Delay With Game Mode and Native Resolution
Most of the delay you can remove lives in the display's processing, and televisions are the worst offenders. Here is a comparison of the same 55-inch television at 60 Hz, run five trials in each setting with nothing else changed:
- Standard picture mode: 287, 301, 279, 312 and 294 ms, median 294 ms.
- Game Mode: 248, 239, 261, 244 and 252 ms, median 248 ms.
The 46 ms difference is monitor processing delay that Game Mode removed, close to three frames. To get the same gain on your own monitor or TV:
- Turn on Game Mode, a low-latency mode, or a PC mode input label, which bypasses most image processing and motion processing.
- Run at native resolution and the refresh you actually play at, so the monitor doesn't scale the image.
- Use NVIDIA Reflex or AMD Anti-Lag in games that support them; they cut queued frames waiting behind the GPU.
- Use frame-rate limiters a few frames below the refresh ceiling to keep the render queue short. These last two trim the system's share of the delay, not the monitor's, so check them in a before-and-after run of their own.
Test each change separately. If you flip three settings and the median drops, you won't know which one did the work, and that knowledge is what guides future upgrades.
Browser Check or Hardware Lag Tester: How Accurate Each Is
A browser check reliably shows your combined response through the whole path, and it is excellent at detecting changes between two settings on one setup. It cannot isolate the display's share, because your reaction varies by more than most monitors' processing delay. Browser scheduling and system load add a few milliseconds of noise as well. For a panel-only figure, you need the photodiode or camera route above.
Using Input Lag Database Figures as a Baseline
Review sites publish input lag data for hundreds of models, and an input lag database is a sensible baseline before you buy. Remember how those numbers were taken: many are measured at 60 Hz in the fastest mode, so they aren't automatically what you get at 144 or 240 Hz. Use a published figure to set expectations, then confirm your own monitor's settings with before-and-after runs here.
Frequently Asked Questions
- What is a good result on a monitor input lag test?
- There is no fixed target. The median combines your reaction time with the delay of your mouse or keyboard, computer, browser and monitor, so the useful comparison is between your own runs: same person, same input method, one setting changed, such as game mode on and off.
- Why is my first trial usually the slowest?
- Most people react more slowly on the first attempt while they settle into the task. That is why the test uses five trials and reports the median, which a single slow or lucky trial can't pull far.
- Will a 240 Hz monitor make my result faster?
- A little. The wait for the next frame drops from up to 16.7 ms at 60 Hz to up to 4.17 ms at 240 Hz. That is a real gain in fast games, but it is small next to a human reaction of around 200 ms.
- Can a browser test measure the monitor's own input lag?
- Not on its own. A browser test measures the whole path from the on-screen cue to your response. Isolating display-only lag needs a photodiode, a high-speed camera or a dedicated lag tester synchronized to the video signal.
- What does input lag mean in monitor reviews?
- Reviewers use it for the display's part only: the time from the signal arriving at the monitor to a visible change starting on the panel. Pixel response time, how fast pixels finish changing shade, is a separate figure.
- How can I reduce the delay I measure?
- Turn on game mode or PC mode, run at native resolution and your highest refresh rate, use a wired mouse or a low-latency receiver, and in games try NVIDIA Reflex, AMD Anti-Lag or a frame-rate limiter. Change one thing at a time and rerun the test.
- Why do my results differ from an input lag database?
- Databases often measure at 60 Hz in the fastest picture mode with hardware tools, and they exclude your reaction time. Use their figures as a baseline, then confirm your own settings with before-and-after runs.