How a number gets onto a page.
This site exists because of a gap: a spec sheet tells you a dash cam shoots 4K, and nothing on it tells you whether you can read the number plate of the car in front at night. Independent laboratories answer that second question. Their results are scattered across a dozen sites, in a dozen formats, under conditions that are rarely stated in the same place as the number. We collect them, name who took them, and state what they were taken under.
What we do not do
We do not test products. We own no laboratory, no sound chamber, no gloss meter and no dynamometer. Nothing on this site should ever be read as our own measurement, and if you find a sentence that reads that way, it is a mistake and we want to hear about it.
This is also why the site is called a bench and not a lab. A bench is where you lay things out next to each other and compare them. That is the work we actually do.
We also do not accept payment, product or review copy from a manufacturer in exchange for coverage, a placement or a rating. How we are paid is set out at the bottom of this page and on the disclosure page.
The four marks
Every figure we publish about a named product carries one of four marks. They are not decoration and they are not a rating — they tell you how much weight the number can hold.
A figure with a named source use it
Someone independent measured it, we name them, and we state what they measured under. “12.6 dB(A) at 1 m, idle” is a measurement. Go and check it — the link is there for that, not for decoration.
A figure from the manufacturer marked
Sometimes the only number that exists is the one on the box. We use it, and we say so every single time. A claim is not a measurement, and a spec sheet has never once reported a disappointing result.
A figure set by regulation the rule
Some numbers are not measurements at all — they are the legal minimum a product must clear before it can be sold. We print them as what they are. A brake fluid that boils at exactly its DOT minimum passed a law, not a test.
Two labs, two numbers not merged
When two independent sources measure the same thing and disagree, both appear, side by side, each with its conditions. Averaging them would invent a precision that neither one has.
What makes a source independent
We take a figure from a source when it satisfies all four of these:
- It is not the seller. A manufacturer, a retailer or a shop blog can be quoted, but never as an independent measurement — it gets the marked flag.
- The conditions are published. A brightness figure without a window size, a noise figure without a distance and a runtime without a load are not measurements. If we cannot find the conditions, the number does not go on the page.
- It is attributable. A named publication or laboratory, not an aggregated “experts say”. We link to the exact page, not to the site’s front door.
- It is current enough to matter. Every figure carries the date it was measured. Firmware changes what a dash cam does at night; a 2022 result on a 2026 unit is a different product.
Where the recommendations come from
We do not run a scoring formula that turns a pile of specifications into a single number out of ten. Those scores look objective and are mostly a record of how the author weighted the inputs.
Instead each page states the question a buyer is actually asking — can this camera read a plate at night, will this charger fit the panel I already have, does this coating survive a winter — and then shows the measurements that bear on it, with their sources. Where the evidence does not settle the question, the page says so. “Nobody has measured this” is a finding, and we would rather print it than fill the gap with an opinion in the shape of a fact.
What we cover, and why
Three areas: home EV charging, dash cams, and detailing equipment. They have one thing in common, which is why they are here: in all three, the number printed on the box is a poor predictor of what the thing does in a driveway. Amperage is not the amperage you can use, resolution is not legibility, and a coating’s advertised durability is measured in months by people who did not leave it outside for one.
Corrections
When we get something wrong we change it on the page and say what changed. A correction that is quietly edited in is indistinguishable from never having been wrong, which would make every other page here worth less.
9 August 2026
Premium fuel — a percentage range removed. The page said carbon deposits fell by “45 to 72 per cent … over a 5,000-mile interval”, crediting Consumer Reports as reporting the same work AAA did. Consumer Reports is not among that page’s sources and no entry for that report exists in our registry, so the sentence read as sourced and was not. That is the whole of it: we are not saying the figure is wrong, only that nothing here let a reader check it.
AAA’s qualitative finding stays, because that one is cited: in most cases deposits already present can be reduced by switching to a TOP TIER fuel. We would rather publish the finding without a number than a number nobody can check.
A first draft of this notice also said the 5,000-mile interval was wrong, on the grounds that AAA’s report describes 4,000 miles. It does — for the comparison of TOP TIER fuels against others. The question about removing deposits an engine already has is described as a separate inquiry and no interval is given for it, so that was an unsupported claim of our own, written into a correction. It is removed.
How we are paid
Links to retailers on this site are affiliate links, mostly to Amazon. If you buy through one, we earn a commission at no extra cost to you. As an Amazon Associate we earn from qualifying purchases.
The commission is the same whichever product on a page you buy, and it does not change what we recommend or what order things appear in. The full statement is on the disclosure page.