How This Test Measures — Sweep Method and Accuracy
Logarithmic sweeps, the method of limits in both directions, handling anticipated presses, median records and how this differs from a clinical hearing test.
Published 2026-09-26
Logarithmic sweep
The sweep changes frequency at a constant 'ratio'. Moving from a start frequency f0 to an end frequency f1 over T seconds, the frequency at time t is:
f(t) = f0 × (f1 / f0)^(t / T)
This covers the same number of octaves in the same amount of time, so the rate of pitch change you perceive is constant from start to finish. The speed setting (20, 40 or 60 seconds) changes T. The slower the sweep, the smaller the error from reaction time.
Method of limits in two directions
Like the method of limits, a classic technique in psychophysics, the test works in both directions: from audible to inaudible and from inaudible to audible.
- When it fades out: press when a tone you could hear disappears. The high limit rises from 8 kHz; the low limit falls from 250 Hz.
- When it becomes audible: after a random wait of 1–3 seconds the tone begins; press when you start to hear it. Pressing before the sound starts is treated as an 'anticipated press' and not recorded.
It takes time to notice a change and move your hand, so the two directions are skewed in opposite ways. Measuring both lets you assume the true limit lies somewhere in between.
Scoring rules
- In the fade-out method, pressing within 1.2 seconds of the start is marked 'right after start'. You may not have heard even the starting frequency.
- If you never press, the fade-out method records 'audible to the end frequency' (≥ for the high limit, ≤ for the low limit), and the becomes-audible method records 'not heard'.
- The frequency is calculated at the moment you pressed minus the output latency reported by the browser.
- Each value in the results table is the median of the last 5 valid attempts for that ear and test, so a single outlier moves it less.
Volume display (dBFS) and real loudness
dBFS is a relative unit where the digital maximum is 0. The same −30 dBFS can produce very different sound pressure at your ear (dB SPL) depending on device volume and earphone sensitivity. That's why results from this site are hard to compare across devices or between people.
How it differs from a clinical hearing test
A clinical pure-tone test uses a calibrated audiometer and headphones in a sound-treated room to find the quietest sound you can hear (hearing threshold, dB HL) at each frequency from 250 Hz to 8 kHz, following standard procedures such as ISO 8253-1. This site doesn't ask 'how quiet a sound' — it estimates 'up to which frequency', without calibration.
If your results worry you, or your hearing has dropped suddenly, get a proper test from an ENT doctor or audiologist.
Tips for more accurate results
- Use wired headphones in a quiet room.
- Start at a low volume, set it so the start of the sweep is comfortably audible, and don't change it during testing.
- Test each ear separately, 2–3 times with each method.
- Only compare results measured on the same device with the same volume settings.