Earbud test

Check each bud, then a low tone. If bass appears only when you press the bud in, the tip is not sealing. Swap tips before you call the earbud thin.

A headphone test is about the two cups and the seal, not about a score. Left must be left, right must be right, and a centered tone should sit in the middle of the head. Speech plus a hard-panned tone is harder to fool than a tone alone, because a mono Bluetooth profile will say “left” in both ears. If the word is centered, you are not on a stereo media path. Do that check before you trust a sweep.

Balance lives in two places. The operating system has a balance slider, and this page has a pan. Set the system slider to the center before you decide the headphones are lopsided. A page cannot override a Windows or macOS balance that is already parked on one side. After the system is centered, a tone that still favors one cup is the pair, the cable, or the ear tip.

Bass on earbuds is mostly a seal. A 70 Hz tone that appears only when you press the bud in means the tip is leaking, not that the driver has no low end. Change the tip size before you call the earbud thin. Open-back headphones leak bass in a different way: the room hears them, and a loose pad does the same job as a loose tip.

Polarity is wiring. In phase, a short low click sits in the center. With one side inverted, the image smears and bass thins. Summing that inverted pair to mono makes the click nearly vanish, which is the clearest proof. A taste for “more bass” will not fix an inverted driver. Swap the wiring, or stop using a cable that flips one side.

Sweeps are for rattles and dropouts. A log sweep spends time in the octaves you actually hear. A buzz at one pitch is often a hair on a driver or a loose plate in the cup. It is not a frequency-response plot. This page does not have a measurement microphone in your ear, and it will not draw a curve and pretend the curve is yours.

A missing high tone is not a diagnosis. It can be the driver, the volume cap on this page, your age, or a real hearing change. Only a clinician with calibrated equipment can test hearing. Do not turn a system volume to maximum to “find the limit.”

Latency is the delay you notice against a video. The browser can report its own audio buffer, often a handful of milliseconds. Bluetooth codecs commonly add something closer to a tenth of a second or more, and the page cannot read that number from the headphone. Film the flash-and-click if you need a rough count. At 30 frames per second, each frame is about 33 milliseconds. That method includes your camera, so treat it as an estimate.

Keep the level down while you hunt for a high tone. The slider on these pages is capped, and that cap is deliberate. People damage hearing by turning the computer up to chase a pitch the driver was never loud at. If you cannot hear 12 kHz at a comfortable level, leave it there. A clinician’s booth is the next step, not a louder browser.

Work a pair in a fixed order. Confirm the output device, play the spoken left and right, then a centered tone, then one low tone for the seal, then polarity if the center sounds hollow. Write down which step failed. A swap is a cable. A seal failure is a tip. A hollow center is polarity. A delay against video is the radio, and it belongs on the latency page rather than on a new pair of headphones. Stopping at the first failure saves you from “fixing” three things that were never broken.