00 · The bench
You are about
to be judged.
There is no homepage here. No services grid, no gallery, no scroll. This site does not exist until it has measured you, and then it is assembled out of what it found.
First it works out what your actual audio chain can deliver — with a tone and a tap, and no permission of any kind. Then it checks its own arguments against that result and openly gives up the ones it cannot honestly demonstrate to you. What survives, you sit blind: unlabelled pairs, matched to the same measured loudness so that loudness cannot decide for you.
Everything you hear is synthesised in your browser from measured filters and real cabin geometry. It is a simulation. It is never a recording of a car we built, and this page will keep saying so.
Four minutes. Nothing is stored, nothing is sent anywhere, and no microphone is requested at any point.
This browser has no Web Audio, or no way to render sound offline for measurement. Both are needed: without offline rendering, two sides of a blind test cannot be honestly matched, and an unmatched blind test is worse than none.
01 · The chain
You are
the microphone.
A tone falls in frequency at a fixed digital level. Tap when it disappears. Then one climbs, and you tap again. Between the two numbers is the band of music that can actually reach you right now.
It measures speaker, room, ears, volume setting and attention, all four together, and it cannot separate them. A fixed digital level is not a fixed loudness — your ear's sensitivity falls off a cliff below about 100 Hz — so the tone goes quiet partly because of your chain and partly because of you. Both are reasons an argument cannot reach you, which is all the next section needs to know. It moves if you move the volume. Run it twice and it will disagree with itself. It is not calibrated and we are not going to call it calibrated.
Sweep one of two — descending, 400 Hz down to 30 Hz.
One thing no browser can measure
Nothing on the web reports whether you have headphones on. There is no API for it, and asking for a microphone to find out would be both unreliable and a trust catastrophe on a shop's website. So we ask. It decides whether one section further down opens at all.
02 · The consequence
What we are
allowed to argue.
Five things this business argues about a car, checked against the chain you just measured. The rule, so you can argue with it: an argument survives when its frequency is more than 12% above your low cut and more than 10% below your high cut. Timing has no frequency at all — it is a difference between your two ears — so it is never gated on this test.
Pair 1 · 01
Weight
Listen to both before you can commit. The choice is final, and which cabin is “1” was a coin toss.
04 · The seat
Where is
the band?
Before you listen. This uses one averaged head-and-ear model to put sound around your head. Some people get a solid image out in front of them; some get it inside their skull, or behind them. That is normal, it is unfixable on the web, and you are being told now rather than afterwards — because the failure mode of this technology is the whole argument of the business. It is the same reason a car is tuned to your seat and your ears rather than to a specification sheet.
Untuned, driver’s seat
- Near door → ear—
- Far door → ear—
- Head start—
We are not going to fake it.
Here is the instrument instead. Same engine, same geometry, same speed of sound — you are reading the numbers rather than hearing them, which is honest because a number is not an illusion.
| State | Head start | Modelled image |
|---|
Take the seat with you. Almost everybody has a pair of earphones within reach of the phone in their pocket. Open this page there, put them on, and the seat opens.
05 · Your sheet
The sheet.
EMMA scoring is a person listening blind and filling in a marked sheet by category. That is what an EMMA International head judge does for a living, and it is the instrument this shop's authority rests on. For the last few minutes it has been pointed the other way: you were the car being judged.
This is the same artefact a car we build goes home with. It is yours, and nothing on it was transmitted anywhere.
| Under test | Category | You chose | Result |
|---|
What your ears just told us
In the order the money goes in
Read that list again and notice what is not on it. Not one thing you were tested on was a speaker — every defect was the cabin, the geometry, or the arrival time — so nothing here can recommend a more expensive speaker. That is true by construction, not by assertion.