Methodology
How Testpartout measures
These are the principles behind every test on the platform. Each test page then documents its own procedure and its own limits in detail.
Platform-wide principles
Measurements use browser APIs
Every Testpartout test is built on standard web platform APIs — pointer and keyboard events, requestAnimationFrame, and related interfaces. What the browser can observe is what the test can measure; nothing is simulated to look more precise than it is.
Timing-sensitive tests use performance.now()
Where a test depends on elapsed time — reaction time, click speed, time-to-target and similar — the measurement is anchored on performance.now(), the browser's monotonic high-resolution clock, rather than wall-clock time.
Methodology is documented per test
This page describes platform-wide principles only. The specific procedure of each test — trial counts, exclusion rules, scoring formulas — is documented on that test's own page, next to the tool itself, and kept in step with the implementation.
Your environment affects the result
Browser, device, display refresh rate and input-device latency all influence a measurement. A score taken on one machine is not directly comparable to a score taken on another, and small differences between runs are normal.
Browser measurements, not laboratory instruments
Testpartout tests are practical browser measurements. They are useful for comparing your own runs and spotting changes, but they are not calibrated laboratory equipment and should not be read as clinical or scientific measurements.
No invented population averages
Testpartout does not publish population averages, global percentiles or rankings that are not backed by real aggregated measurements. Until such data genuinely exists, no comparative statistics are shown.
Personal bests are currently local
Where a test keeps a personal best, it is stored in your browser's local storage on your device. It is not uploaded, and clearing your browser storage removes it.
Future benchmarks will use real data only
If aggregated benchmarks are ever introduced, they will be computed from real measured results and labelled as such. Synthetic or placeholder statistics will not be published in their place.
How we validate Testpartout tools
Calculation verification
Pure functions can be tested with known inputs and expected outputs, including event-rate math, grid coverage, held-key comparisons and checklist state. These checks verify calculations, not physical hardware.
Browser-flow verification
Routes, controls, API state and cleanup can be exercised in an actual browser. Automated input verifies the browser flow; it does not establish that a real device generated the input or that sound reached a speaker.
Hardware-dependent verification
Real microphones, cameras, speakers, controllers, displays, mice and touchscreens are needed to evaluate their physical behavior. We do not claim every tool has been checked with professional equipment, calibrated laboratory instruments or every hardware model.
User observation
Pixel defects, larger display patterns and audio routing require you to look or listen. Browser-generated patterns and tones are not automatic fault detection. Local checklists organize your observations without creating a hardware-health score. Optional result cards are created locally and shared only after your explicit choice.
Per-test methodology
The detailed method for each live test — including what the implementation does and the limits of a browser measurement — is on the test's own page:
- Touch Screen TestMap touch coverage, compare simultaneous contacts and watch for unexpected touch events reported by your browser.
- Keyboard Ghosting & Key Rollover TestHold combinations together and compare browser-observed keys, presets and maximum simultaneous input.
- Mouse Polling Rate TestEstimate browser-delivered mouse event rate over ten seconds with median intervals and sample context.
- Monitor TestInspect manually selected colors, gradients, uniformity and geometry patterns without automatic defect claims.
- Microphone TestCheck that your microphone is detected and picking up sound, with a live input level meter.
- Speaker TestPlay left, right and reference tones to check each speaker is working, on the correct side and free of obvious distortion.
- Left / Right Headphone TestVerify stereo channels, wiring and balance by playing audio to the left and right sides separately.
- Webcam TestSee your camera feed, confirm the right device is selected and check framing before a call.
- Keyboard TestPress keys to confirm every one registers, and find stuck, dead or repeating keys.
- Mouse TestCheck buttons, double-clicks, scrolling and pointer movement for a failing or erratic mouse.
- Gamepad / Controller TestCheck buttons, triggers and analog sticks, including stick drift and dead-zone behaviour.
- Dead Pixel TestRun full-screen solid colours to reveal dead, stuck or discoloured pixels on your display.
- Screen Resolution TestSee the resolution, viewport size, pixel ratio and colour depth your browser reports.
- Refresh Rate TestEstimate the refresh rate your display is actually running at, with the limits of the method stated.
- Reaction Time TestMeasure how fast you respond to a visual cue, in milliseconds, across repeated attempts.
- Typing Speed TestWords per minute, accuracy and mistake patterns over a timed typing passage.
- Click Speed TestClicks per second over a fixed interval, with consistency across attempts.
- Visual Memory TestRecall increasingly complex patterns to measure your visual memory span.
- Sequence Memory TestRepeat growing sequences to find the longest order you can hold and recall.
- Aim / Accuracy TestHit targets under time pressure to measure accuracy, hits, misses and time-to-target.