Aller au contenu

Our test protocol

This page describes exactly what we do, and what we do not do, before writing that a product convinced us.

Three statuses, never mixed up

Every product we list carries a status visible at the top of its page. It is the first thing we show, before the specifications and before the prices.

StatusWhat it meansWhat we rule out
Not tested by usA spec sheet built from the manufacturer specifications, sourced and dated.No score, no impressions, no "we recommend".
Hands-onShort, dated use in conditions we did not control (a show, a session at someone else’s place).No overall score, no final verdict, no judgement on reliability.
Full reviewThe protocol below applied in full, over a stated duration.No restriction.

The test bench

The exact setup used for each review is listed at the top of the article: frame, reference base, reference pedals, display and software settings. Without that, a result is not reproducible, so it is not verifiable.

The games we use

We always use the same titles, so that two reviews stay comparable:

  • iRacing: the reference for online racing, force feedback with little filtering
  • Assetto Corsa Competizione: modern GT, heavy aerodynamic load
  • Le Mans Ultimate: endurance, tyre management over time
  • EA Sports WRC: loose surfaces, fast lateral loads

Minimum duration

A full review means at least twenty hours of use spread over two weeks. Heat behaviour, software stability and perceived wear do not show up in one session.

Scoring grid

Six criteria, each scored out of 10, weighted as follows. The overall score is computed automatically from these six values: it is never typed in by hand.

CriterionWeightWhat it measures
Build15 %Materials, assembly, finish, mechanical noise
Setup and software15 %Assembly, drivers, ergonomics and software stability
Force feedback / performance30 %Detail, latency, linearity, heat behaviour
Ecosystem10 %Accessories, room to grow, longevity
Compatibility10 %Platforms, games, standards, adapters
Value for money20 %At the real price recorded, against the alternatives

Our scale is deliberately demanding: **6/10 means a good product**. A scale where everything sits above 8 carries no information.

For a category with no force feedback (cockpit, seat), the weight of the "force feedback" criterion is redistributed to "build" and "value for money". The grid used is then restated on the review page.

Where the unit came from

Every review states, at the top of the article and not in a footnote, how we obtained the product:

  • Product bought by us
  • Product lent by the manufacturer, with the return date
  • Product supplied and kept
  • Hands-on at an event organised by the manufacturer

Corrections

A factual error is corrected and flagged publicly at the end of the article, with its date. We do not rewrite an error in silence: correcting in public costs five minutes, being caught costs the credibility of the site.

Updated on September 20, 2026