Loading Atlas data
Trust before hype
Handheld performance changes with updates, thermals, silicon, background tasks and the occasional Windows gremlin. Our job is not to pretend those variables vanished. It is to document the target, show the evidence and explain the compromise clearly enough that another player can reproduce it.
Principle 1
A result only means something when the game, handheld, resolution, power target and important settings are known. We avoid treating a loose FPS number as holy scripture.
Principle 2
Atlas Score, community game ratings, preset upvotes, Worked for me confirmations and Atlas Verified are different signals. They are shown separately because mixing them into one mystery number would be bullshit.
Principle 3
A useful preset says what it targets, why settings changed and what the player gives up in image quality, latency, battery life or stability.
Principle 4
Missing measurements and early community samples are labelled instead of being polished into fake certainty. Confidence should grow with evidence, not marketing volume.
Settings Impact Database
Graphics menus love renaming the same idea. The Settings Guide keeps one public page for each canonical setting and connects aliases such as “Shadows”, “Shadow Quality” and “Shadow Detail” behind the scenes. That prevents duplicate pages while still matching old and new presets automatically.
General FPS, visual and VRAM scores are HandheldAtlas editorial estimates. The vocabulary and technical descriptions are informed by official engine and GPU-vendor documentation, while a game-specific measured result takes priority whenever the game, handheld, TDP and resolution match.
Signal map
A popular preset can still be weakly documented. A verified preset can still perform differently after a game patch. The labels below answer different questions on purpose.
Editorial
A curated compatibility and handheld-experience score for a game. It is not the same as the community's enjoyment rating.
Evidence model
Scores how complete and reproducible a preset is using test targets, FPS data, settings detail, explanations, verification and community proof.
Players
A one-rating-per-account opinion score for games or presets. It measures sentiment, not laboratory certainty.
Matching users
A confirmation that a preset worked on the listed target. Users should only confirm after matching the handheld, TDP and resolution.
Editorial review
A staff-reviewed signal that the preset contains a coherent target, usable settings and credible supporting data.
Benchmark minimum
Limitations
A preset is a documented starting point, not a blood oath from the silicon gods. Game updates, firmware, drivers, ambient temperature, background software and individual device variance can shift the result.
When data is estimated, incomplete or based on a small sample, the interface should say so. Reports that conflict with the listed target can be moderated, and confidence can change as better data arrives.
The honest answer is sometimes “we need more data.”
That answer is less sexy than fake precision, but considerably more useful.