How does The Audio Stuff score audio gear?
Every review carries a single score from 0 to 10 plus a one-word verdict, and each verdict owns a score band no other tier shares: Reference 9.2-10.0, Highly Recommended 8.2-9.1, Recommended 7.0-8.1, Mixed 5.5-6.9, Pass 0.0-5.4. The bands do not overlap, so a score names one tier. The single exception is a score exactly on a tier's floor (9.2, 8.2, 7.0 or 5.5): that makes a product eligible for the tier, and the verdict may hold it one tier lower where price, competition or reliability argue against the higher one. Our publishing system will not let a review go live with a verdict its score does not allow. The scale is normalized across categories: every score is built from the same weighted axes, so a 9.0 is the same standing whether the product is a headphone, a DAC or a pair of speakers, while what earns the points on each axis is category-specific - tonal balance on a headphone, measured noise on a DAC, a measurable or audible effect on an accessory - and each category hub lists its criteria. When a new benchmark arrives in a category, an older score can move, and the change is dated on that review.
What does 'Reference' mean in the verdict scale?
Reference is the top verdict, for scores of 9.2 and up: a product that sets the standard in its class, and rare by design. It does not mean the product suits everyone - price, taste and the rest of your system still decide that. It is also a different thing from the test chain, the gear every review is compared on: a piece is in the chain because it stays the same from one review to the next, and three of its six pieces score below 9.2.
What audio gear brands does The Audio Stuff cover?
Gear from 33 brands so far, most of all HIFIMAN, SMSL, Sivga, Taga Harmony, Denafrips, Kiwi Ears, Melodika and Monosaudio; the other 25 are in the review catalogue, searchable by name. Category, merit and reader requests decide what gets reviewed next; brand relationships do not.
Do reviews use measurements or just listening tests?
Both, in that order of weight. Listening is always primary - over weeks, on multiple chains, against the gear each category is compared on. Measurements (frequency response, distortion, output impedance, jitter, sensitivity) are added when they reveal something audible or contradict the marketing claims. We do not chase graphs over what the gear actually sounds like in a real room.