The rules we wrote down so we could not quietly break them
Every editorial promise below is testable. If we fail one of them in a published piece, the correction appears at the top of that article with the date attached, and the digest that week names the failure explicitly.
Minimum time before we write
A review is only published once the reviewer has reached a defensible stopping point. What counts as defensible varies by form, so we set floors rather than averages.
| Type of game | Minimum before publishing | What must be finished |
|---|---|---|
| Narrative role-playing | 40 hours | Main storyline plus at least two optional regions |
| Linear story game | Full campaign | Credits reached without difficulty reduction mid-run |
| Driving and sport | 25 hours | All handling assists tested off, half the tracks driven |
| Systems and strategy | 30 hours | Two full campaigns at different difficulty settings |
| Live or seasonal game | One full season | Written only after a season closes, never at launch |
Hardware we test on
Mid-range desktop
Our baseline machine, kept deliberately ordinary and replaced only every four years.
Five-year-old laptop
Integrated graphics, thermally limited. If a game is unplayable here we say so plainly.
Handheld
Checked for frame pacing and readable text at arm's length, not just for average frame rate.
Current console
Used for any release where the console build is the version most readers will actually own.
The accessibility pass
Carried out by hand on every game, before the review is drafted, using the same checklist each time. We record what exists, what is missing, and how easy it was to find.
Input remapping
Can every action be reassigned, including menu navigation? Are hold actions convertible to a single press?
Text and subtitles
Minimum readable size, speaker labels, background opacity, and whether the setting survives a restart.
Colour dependence
Any puzzle or warning that relies on hue alone is flagged, with the specific moment named in the review.
Motion and camera
Field of view, head-bob, screen shake and motion blur toggles, tested for whether they actually change anything.
Difficulty granularity
Whether combat, puzzle and timing challenge can be adjusted separately, and whether changes apply mid-run.
Findability
How many menus deep the options are buried. Excellent features nobody can locate are noted as partial.
Independence, stated plainly
Daxureli sells nothing. There is no shop, no membership tier, no affiliate link and no sponsored placement anywhere on this site, and there never will be while the current four of us run it.
- Review copies are declared.
If a publisher provided access, the first line of the review says so. Roughly a third of our archive was bought at retail by the reviewer. - No embargo theatre.
We do not agree to conditions that dictate tone, timing tied to marketing beats, or approval of text before publication. - Travel is refused.
We do not accept paid trips or accommodation to preview events. If we cannot reach it by train at our own cost, we do not cover it. - Conflicts are named.
Where a reviewer knows someone on a development team, another writer takes the piece and the connection is stated in a footnote.
The twelve-month revisit
Games change after release, so a launch verdict has a shelf life. Twelve months after publication the original reviewer replays the game for a minimum of ten hours and adds a dated revisit above the original article.
The original paragraphs are never deleted or silently rewritten. If the revisit contradicts them, both stay visible — that contradiction is the useful part.
Four reviews are due a second look this quarter, including one where we expect to reverse the recommendation entirely.
Corrections, delivered on purpose
The Friday digest always names what we got wrong that week. Subscribe if you would rather hear about our mistakes from us than find them yourself.